JP5544571B2 - Hierarchical structure management apparatus, method thereof, and program - Google Patents

Description

  The present invention relates to a hierarchical structure management apparatus that manages files in a multidimensional hierarchical structure.

  In modern offices, a large number of electronic files are stored unorganized on personal computers and file servers. Companies need to increase the productivity of intellectual workers to create new value, but information workers spend an average of 24% of their working hours searching and analyzing information There are also results ("The Hidden Costs of Information Work", IDC (http://www.idc.com/), refer to April 2006). Technology that easily grasps the whole picture is important to increase the productivity of intellectual workers. The open government initiative launched by the government is highly important for improving the efficiency of government and local government operations while fulfilling accountability to the public. Files need to be organized systematically. Since the importance of IT governance is also increasing in private companies, there is a great need for systematically organizing a large number of electronic files accumulated in the course of business.

  As a method for organizing files stored on a computer, a method using a folder hierarchy is generally used. However, since various classification viewpoints are mixed in a single folder hierarchy, (1) I do not know where the necessary files are, (2) Only the creator of the folder hierarchy can understand the classification hierarchy, (3) There is a problem that the location where the created file should be stored cannot be specified. For this reason, ingenuity such as introduction of full-text search engines and tagging of files may be performed, but files that do not match the search keyword specified by the user or files with low search order are buried, which is essential. It does not solve the problem.

  The inventors have proposed a multidimensional classification method for managing the hierarchical structure of files in a multidimensional manner with respect to the above problems (see Non-Patent Documents 1 and 2). The multidimensional classification method is a method of classifying entities to be classified in association with a plurality of tree structures. An entity is arbitrary as long as it is information held electronically. Each tree structure represents entity classification criteria based on a specific point of view so that IS-A relations (child nodes correspond to special cases of parent nodes) are established between parent and child nodes. Build (IS-A constraint). Also, sibling nodes of the same tree structure are mutually exclusive (exclusive constraint). The IS-A constraint can keep the classification criteria consistent, and the exclusion constraint makes the classification criteria clear. As a result, a MECE tree structure is constructed and the tree structure can be easily understood, so that the efficiency of file classification work and search work increases.

  The characteristics of the multidimensional classification method include the following. (1) Since the individual tree structure used for classification is consistent and relatively small, it is easy for the user to understand and maintain the tree structure. (2) A user can freely specify an individual tree structure or a node of the tree structure to search for an entity. Since the tree structure and the node specification order can be specified freely, the degree of freedom in searching is great. (3) The user can easily grasp the overall image of the stored information. This makes it possible to quickly determine whether desired information is registered in the system.

As a technique for facilitating access to a file, a detecting means (CPU or the like) detects a plurality of HDDs connected to a PC, and a reading means (CPU or the like) detects a directory tree from each of the detected plurality of HDDs. The creation means (CPU or the like) searches the directory tree for the same name folder in the same hierarchy in order from the highest level in the directory tree, sets the searched same name folder in the same hierarchy as one folder, and stores the same folder in the folder. Patent Document 1 discloses a technique in which a virtual directory tree is created by repeating an operation of writing a folder in a lower hierarchy that is associated, and a management unit (such as a CPU) manages access to a file based on the virtual directory tree. Yes.

Kakeshita Tetsuro, Harayuki Yuki, "Multidimensional Classification: A Combined Approach of Tree Structure Classification and Keyword Classification", Transactions of Information Processing Society of Japan: Database, Vol. 42, no. SIG1 (TOD8), pp. 131-139, 2001 INOguchi, Tetsuro Kakeshita, “Automating tree structure in multidimensional classification”, Transactions of Information Processing Society of Japan: Database, Vol. 44, no. SIG12 (TOD19), pp. 33-42, 2003

JP 2008-226137 A

  However, the technologies shown in Non-Patent Documents 1 and 2 make it much easier to manage files, but migrate a large amount of files stored in existing file servers or personal computers to a multidimensional tree. To achieve this, there is a problem that it takes a lot of labor and time.

  The technique disclosed in Patent Document 1 searches for the same name folder in the same hierarchy in order from the highest level, and makes the searched same name folder one folder to facilitate access to the file. It is not a technology that can manage the hierarchical structure based on the system, but is not sufficient technology to facilitate access to the file for the hierarchical structure that is managed by mixing various classification viewpoints. Have

  Therefore, the present invention automatically generates a hierarchical structure based on an arbitrary viewpoint from a hierarchical structure managed by mixing various classification viewpoints, and provides a user-friendly hierarchical structure management apparatus and the like.

  The hierarchical structure management device disclosed in the present application includes a file storage unit that stores a plurality of management files to be managed, correspondence information that associates the management file with each node in an existing old hierarchical structure, and the management Correspondence information storage means for storing correspondence information in which a file is associated with each node in the newly generated new hierarchical structure, and when the word set composed of any one or a plurality of words is input, the correspondence information Deletes a word that is not included in the input word set from the names of the nodes in the node, deletes the empty node for which the deleted name is empty, and deletes the empty node, and sibling-related nodes having a common parent And reconfiguring means for reconfiguring the correspondence information by integrating the nodes having the same name, and when the reconfiguring means deletes the empty node, It is intended to correspond to the nearest ancestor node for the child nodes and the leaf nodes of the spatial nodes.

As described above, in the hierarchical structure management device disclosed in the present application, the correspondence information that associates the managed file with each node of the existing old hierarchical structure, and the newly created new file and the managed file. Correspondence information that associates each node in the hierarchical structure is stored, and when the word set consisting of one or more arbitrary words is input, this correspondence information is included in the word set input from the node name Delete the unrecognized word, delete the empty node for the empty node whose name is empty, and merge the nodes with the same name among the sibling nodes that have a common parent. When reconfiguring the correspondence information, integrating the nodes with the same name among sibling nodes with a common parent, and deleting the empty node, the child node and leaf node of the empty node In order to correspond to the nearest ancestor node, it is possible to generate a hierarchical structure based on any one or more input word sets, and the user can create a hierarchical structure based on an arbitrary viewpoint from the existing hierarchical structure. It is possible to customize and generate and to improve usability.

  The hierarchical structure management device disclosed in the present application includes a word information generation unit that generates word information by dividing the name of the management file and the name of each node in the old hierarchical structure by morphological analysis, and all of the old hierarchical structure As for the word information related to the node, for each hierarchical level, a duplicate word that appears redundantly in the name of the node is extracted, and when the duplicate word is extracted, the word information other than the duplicate word is extracted. The processing of the reconstruction means is executed for the word set, the duplicated word extraction is performed for all hierarchical levels, and the correspondence information between the nodes and files finally remaining after the processing of the reconstruction means is executed. A first hierarchical structure determining means for determining the first new hierarchical structure is provided.

  Thus, in the hierarchical structure management device disclosed in the present application, the name of the management file and the name of each node in the old hierarchical structure are divided by morphological analysis to create word information, and all the nodes in the old hierarchical structure are related. For word information, for each hierarchical level, duplicate words that appear redundantly in the node name are extracted, the processing of the reconstruction means is executed for a word set other than the duplicate words, and duplicate words are extracted for all hierarchical levels And the correspondence information between the node and the file finally remaining after the processing of the reconstruction means is determined as the first new hierarchical structure, the duplication is eliminated for the node in which the duplicate word is included in the name. There is an effect that the hierarchical structure classified from one viewpoint can be automatically generated automatically.

  The hierarchical structure management device disclosed in the present application, when there is no other new hierarchical structure other than the first new hierarchical structure, for any one duplicate word extracted by the first hierarchical structure determining means, The correspondence information obtained by executing the processing of the reconfiguring means with only one overlapping word as a word set is determined as another new hierarchical structure after the second, and other than the first new hierarchical structure If there is a new hierarchical structure, the word set that constitutes the name of each node in the other new hierarchical structure is the word set obtained by adding the arbitrary one duplicate word for each other new hierarchical structure If the added duplicate word appears only once in the hierarchical structure reconstructed as a result of executing the processing of the reconstruction means sequentially, the correspondence information obtained by the processing of the reconstruction means is Other new hierarchies after 2 If the added duplicate word appears multiple times, the correspondence information obtained by executing the processing of the reconstructing means with only the added duplicate word as a word set is added to the second and subsequent new information. A process for determining a hierarchical structure is executed for all the duplicate words to provide an nth hierarchical structure determining means for determining the second and subsequent new hierarchical structures.

  As described above, in the hierarchical structure management device disclosed in the present application, for any one duplicate word extracted by the first hierarchical structure determining means, the word set constituting the name of each node in the new hierarchical structure When the processing of the reconstruction means is sequentially executed as a word set by adding any one overlapping word for each new hierarchical structure, and the added overlapping word appears only once in the reconstructed hierarchical structure The reconstructed hierarchical structure is determined as another new hierarchical structure after the second, and when the added duplicate word appears a plurality of times, only the added duplicate word is regarded as a word set and processing of the reconstruction means In addition to the first hierarchical structure obtained above, the second and subsequent hierarchical structures classified from another viewpoint are determined by determining the hierarchical structure obtained by executing Can be generated, like Achieved viewpoint from one old hierarchical structure are mixed, the effect of a plurality of hierarchical structure classified by different viewpoint can be easily generated automatically.

  In the hierarchical structure management device disclosed in the present application, the n-th hierarchical structure determination unit adds the arbitrary one duplicate word to a word set that constitutes the name of each node in the other new hierarchical structure. When the processing of the reconfiguring means is sequentially executed as a set, there are a plurality of other new hierarchical structures in which the added one duplicate word appears only once in the hierarchical structure reconfigured as a result of execution. When there is, the number of files associated with each root node in the plurality of other new hierarchical structures is compared, and the one new overlapping word is added to the other new hierarchical structure in which the number of the files is minimized. The correspondence information obtained by adding is determined as a new hierarchical structure.

  As described above, in the hierarchical structure management device disclosed in the present application, processing of the reconstruction means is performed by using any one duplicate word added to the word set constituting the name of each node in the new hierarchical structure as a word set. When there are multiple new hierarchical structures in which any one duplicate word appears only once, the number of files associated with each root node in the multiple new hierarchical structures is compared, and Since the correspondence information obtained by adding any one duplicate word to the new hierarchy structure with the smallest number of files is determined as the new hierarchy structure, a new candidate that can be classified as any one duplicate word is determined. Even if there are multiple hierarchies, the number of files associated with the root node is small, and the new hierarchies are efficiently organized to improve the accuracy of classification and manage files accurately. An effect that can be.

  In the hierarchical structure management device disclosed in the present application, when the n-th hierarchical structure determining unit compares the number of files associated with root nodes in the plurality of other new hierarchical structures, the minimum number of files is the same. When there are a plurality of other new hierarchical structures, the added duplicate word is stored as a reserved word, and the processing of the reconstructing means is completed for other duplicate words other than the duplicate word that has become the reserved word. After that, the correspondence information obtained by executing the processing of the reconstructing means for the reserved word again is determined as a new hierarchical structure.

  Thus, in the hierarchical structure management device disclosed in the present application, when comparing the number of files associated with the root node in a plurality of new hierarchical structures, there are a plurality of new hierarchical structures with the same minimum number of files. In some cases, the added duplicate word is stored as a reserved word, and the process of the reconstruction means is performed again for the reserved word after the process of the reconstruction means is finished for other duplicate words other than the duplicate word that has become the reserved word. In order to determine the correspondence information obtained by execution as a new hierarchical structure, the structure of the new hierarchical structure has been changed by executing the processing of the reconstruction means for other duplicate words once as a reserved word. If the processing of the reconstruction means is executed again for, there is a possibility that it can be classified into any one new hierarchical structure, so the new hierarchical structure is correctly classified and generated There is an effect that it is Rukoto.

  The hierarchical structure management device disclosed in the present application is a display control for displaying the first new hierarchical structure determined by the first hierarchical structure determining means and one or more new hierarchical structures determined by the nth hierarchical structure determining means. And a new hierarchy in which the target word is included for a target word composed of any one or a plurality of words among words constituting the names of all nodes in the first new hierarchical structure and the other new hierarchical structure For each candidate new hierarchical structure other than the structure, the processing of the reconstruction means was sequentially executed by adding the target word to the word constituting the name of the node in the candidate new hierarchical structure as a word set. As a result, when there is a candidate new hierarchical structure in which the target word appears only once after reconstruction, target word extraction means is provided for extracting the target word, and the display control means is configured to extract the target word. For target words extracted in unit, and displays with different display modes of the other words.

As described above, in the hierarchical structure management device disclosed in the present application, the new hierarchical structure is displayed, and among the words constituting the names of all the nodes, the target word is the target word including any one or a plurality of words. For each candidate new hierarchical structure that is not included, the processing of the reconstruction means is sequentially executed as a word set by adding the target word to the word that constitutes the name of the node in the candidate new hierarchical structure, and as a result of the reconstruction, When there is a candidate new hierarchical structure in which the target word appears only once, the target word is extracted and displayed differently from other words, so that the word is currently classified for the target word It is possible to visually recognize that it is possible to classify from the new hierarchical structure to another new hierarchical structure, and the user-friendliness can be improved.

  In the hierarchical structure management device disclosed in the present application, the display control means displays information indicating a candidate new hierarchical structure in which the target word appears only once in association with the target word.

  As described above, in the hierarchical structure management device disclosed in the present application, the information indicating the candidate new hierarchical structure in which the target word appears only once is displayed in association with the target word. It is possible to visually recognize that it is possible to classify from the new hierarchical structure that has been classified into another new hierarchical structure, and display the possible classification destinations together to further improve the user-friendliness. There is an effect that it can be improved.

  The hierarchical structure management device disclosed in the present application is a display control for displaying the first new hierarchical structure determined by the first hierarchical structure determining means and one or more new hierarchical structures determined by the nth hierarchical structure determining means. Means for displaying the same word in distinction from other words when the same word appears in the same hierarchical structure in the new hierarchical structure.

  In this way, in the hierarchical structure management device disclosed in the present application, a new hierarchical structure is displayed, and when the same word appears in the same hierarchical structure, the same word is displayed separately from other words. Therefore, even if there is a word classified in the same hierarchical structure without being classified correctly, it is possible to visually display the word and easily recognize it. .

  The hierarchical structure management device disclosed in the present application is a display control for displaying the first new hierarchical structure determined by the first hierarchical structure determining means and one or more new hierarchical structures determined by the nth hierarchical structure determining means. Means, a designated word consisting of any one or a plurality of words constituting the name of the node of any one of the displayed new hierarchical structures, and a destination new hierarchical structure for moving the designated words are designated In this case, the processing of the reconfiguring unit is executed by adding the designated word to the words constituting the new destination hierarchical structure as a word set, and the name of the node in the new hierarchical structure including the designated word And a movement control means for controlling the movement of the words between the new hierarchical structures by executing the processing of the reconstruction means as a set of words excluding the designated word from the words constituting

  As described above, in the hierarchical structure management device disclosed in the present application, a specified word composed of any one or a plurality of words that displays a new hierarchical structure and constitutes the name of a node of any one displayed new hierarchical structure. And the new destination hierarchical structure to which the designated word is to be moved, the processing of the reconstruction means is executed by using the word constituting the new destination hierarchical structure plus the designated word as a word set. At the same time, the word structure between the new hierarchical structures is controlled by executing the processing of the reconstruction means as a set of words that excludes the specified words from the words constituting the node names in the new hierarchical structure including the specified words. Therefore, it is possible to move a word constituting the node name from the currently classified hierarchical structure to another hierarchical structure, and to construct a new hierarchical structure. An effect that can be constructed.

The hierarchical structure management device disclosed in the present application is a display control for displaying the first new hierarchical structure determined by the first hierarchical structure determining means and one or more new hierarchical structures determined by the nth hierarchical structure determining means. Means and a designated word consisting of any one or more words constituting the name of any one of the displayed new hierarchical structure nodes, and deletion instruction information for deleting the designated word, The processing of the reconstructing means is performed by using the word constituting the name of the node in the new hierarchical structure including the inputted designated word, excluding the designated word as a word set, and the word from the new hierarchical structure Deletion control means for controlling deletion.

  As described above, in the hierarchical structure management device disclosed in the present application, a specified word composed of any one or a plurality of words that displays a new hierarchical structure and constitutes the name of a node of any one displayed new hierarchical structure. When the delete instruction information for deleting the specified word is input, the word constituting the node name in the new hierarchical structure including the input specified word is reconfigured as a word set. Since the processing of the means is executed and the deletion of words from the new hierarchical structure is controlled, an unnecessary hierarchical word can be deleted and an accurate hierarchical structure can be constructed by a high-level operation.

  The hierarchical structure management device disclosed in the present application is a display control for displaying the first new hierarchical structure determined by the first hierarchical structure determining means and one or more new hierarchical structures determined by the nth hierarchical structure determining means. And a plurality of words constituting the names of the displayed nodes of the new hierarchical structure are designated, and when the instruction information having the designated designated word as a synonym is input, the designated word is arbitrarily selected. Or synonym control means for controlling the synonyms by executing the processing of the reconstruction means by unifying information indicating the designated word into one information.

  As described above, in the hierarchical structure management device disclosed in the present application, a new hierarchical structure is displayed, a plurality of words constituting the name of the displayed node of the new hierarchical structure are designated, and the designated designated word is defined as a synonym. When the instruction information to be input is input, the designated word is unified into one arbitrary word, or the information indicating the designated word is unified into one information, the processing of the reconstruction means is executed, and the synonym control is performed. Therefore, even if different words have the same meaning (for example, 2010 and 2010), they can be processed as the same word, and the hierarchical structure is accurately constructed and simplified. There is an effect that can be.

  The hierarchical structure management device disclosed in the present application creates word information by dividing the name of the management file and the name of each node in the old hierarchical structure by morphological analysis, and assigns the word to each word in the word information. A word information generating means for assigning a word ID to be indicated; an old correspondence information display control means for displaying the correspondence information stored in the old correspondence information storage means; and names of nodes in a hierarchical structure in the displayed correspondence information. Among the words, when a plurality of words having the same name are designated and information indicating that the designated name is a word having a different meaning is input, And a polysemy control means for controlling polysemy with different word IDs.

  Thus, in the hierarchical structure management device disclosed in the present application, the name of the management file and the name of each node in the old hierarchical structure are divided by morphological analysis to create word information, and each word in the created word information is assigned to each word. , Assigning a word ID indicating a word, displaying the correspondence information (old hierarchical structure) stored in the old correspondence information storage means, and the same name among the words constituting the name of the node of the hierarchical structure in the displayed correspondence information When a plurality of words are specified and information indicating that the word with the specified name is a word having a different meaning is input, each word ID of the specified word with the same name is made different to make it ambiguous Because words are controlled, words that have different meanings even if they are the same word (for example, another person's “Suzuki” surname) can be processed as different words, and the hierarchical structure An effect that can be constructed to Make.

The hierarchical structure management device disclosed in the present application forms a plurality of hierarchical structures according to different classification standards, and for each of the formed hierarchical structures, for all the management files, the management file and the nodes in the hierarchical structures The correspondence information storage means for storing the correspondence information, the display control means for displaying each hierarchical structure stored in the correspondence information storage means, and the names of all nodes in the correspondence information stored in the correspondence information storage means Words constituting the names of nodes in the candidate new hierarchical structure with respect to each candidate new hierarchical structure other than the hierarchical structure including the target word with respect to a target word composed of any one or a plurality of words The processing of the reconstructing means is sequentially executed by adding the target word to the word set as a word set, and as a result of the execution, the target word is A target word extraction unit that extracts the target word when there is a candidate new hierarchical structure that appears only once, and the display control unit determines whether the target word extracted by the target word extraction unit The display mode is displayed differently.

It is a hardware block diagram of the hierarchical structure management apparatus which concerns on 1st Embodiment. It is a functional block diagram of the hierarchical structure management apparatus which concerns on 1st Embodiment. It is a flowchart which shows operation | movement of the reconstruction process part in the hierarchical structure management apparatus which concerns on 1st Embodiment. It is a figure which shows an example of the old hierarchical structure which the hierarchical structure management apparatus which concerns on 1st Embodiment manages. It is a 1st figure which shows the reconstruction process of the hierarchical structure management apparatus which concerns on 1st Embodiment. It is a 2nd figure which shows the reconstruction process of the hierarchical structure management apparatus which concerns on 1st Embodiment. It is a 3rd figure which shows the reconstruction process of the hierarchical structure management apparatus which concerns on 1st Embodiment. It is a flowchart which shows the whole process of the hierarchical structure management apparatus which concerns on 1st Embodiment. It is a flowchart which shows the word information generation process in the hierarchical structure management apparatus which concerns on 1st Embodiment. It is a figure which shows the word information produced | generated with the hierarchical structure management apparatus which concerns on 1st Embodiment. It is a flowchart which shows the 1st hierarchical structure production | generation process in the hierarchical structure management apparatus which concerns on 1st Embodiment. It is a 1st figure which shows the 1st hierarchical structure production | generation process in the hierarchical structure management apparatus which concerns on 1st Embodiment. It is a 2nd figure which shows the 1st hierarchical structure production | generation process in the hierarchical structure management apparatus which concerns on 1st Embodiment. It is a 3rd figure which shows the 1st hierarchical structure production | generation process in the hierarchical structure management apparatus which concerns on 1st Embodiment. It is a flowchart which shows the nth hierarchical structure production | generation process in the hierarchical structure management apparatus which concerns on 1st Embodiment. It is a 1st figure which shows the nth hierarchical structure production | generation process in the hierarchical structure management apparatus which concerns on 1st Embodiment. It is a 2nd figure which shows the nth hierarchical structure production | generation process in the hierarchical structure management apparatus which concerns on 1st Embodiment. It is a 3rd figure which shows the nth hierarchical structure production | generation process in the hierarchical structure management apparatus which concerns on 1st Embodiment. It is a figure which shows the new hierarchical structure produced | generated with the hierarchical structure management apparatus which concerns on 1st Embodiment. It is a functional block diagram of the hierarchical structure management apparatus which concerns on 2nd Embodiment. It is a 1st figure which shows the process of the check process part in the hierarchical structure management apparatus which concerns on 2nd Embodiment. It is a 2nd figure which shows the process of the check process part in the hierarchical structure management apparatus which concerns on 2nd Embodiment. It is a 3rd figure which shows the process of the check process part in the hierarchical structure management apparatus which concerns on 2nd Embodiment. It is a 1st figure which shows the process of the edit process part in the hierarchical structure management apparatus which concerns on 2nd Embodiment. It is a 2nd figure which shows the process of the edit process part in the hierarchical structure management apparatus which concerns on 2nd Embodiment. It is a 3rd figure which shows the process of the edit process part in the hierarchical structure management apparatus which concerns on 2nd Embodiment. It is a 4th figure which shows the process of the edit process part in the hierarchical structure management apparatus which concerns on 2nd Embodiment.

  Embodiments of the present invention will be described below. The present invention can be implemented in many different forms. Therefore, the present invention should not be construed based only on the description of the present embodiment. Also, the same reference numerals are given to the same elements throughout the present embodiment.

  In the following embodiments, the apparatus will be mainly described. However, as is apparent to those skilled in the art, the present invention can also be implemented as a method and a program for operating a computer. In addition, the present invention can be implemented in hardware, software, or hardware and software embodiments. The program can be recorded on any computer-readable medium such as a hard disk, CD-ROM, DVD-ROM, optical storage device, or magnetic storage device. Furthermore, the program can be recorded on another computer via a network.

(First embodiment of the present invention)
A hierarchical structure management apparatus according to the present embodiment will be described with reference to FIGS. The hierarchical structure management apparatus according to the present embodiment automatically classifies a plurality of files stored in an existing hierarchical structure (old hierarchical structure) in a computer, and sets a plurality of independent hierarchical structures (new hierarchical structure). ) And new hierarchy structure and file correspondence information is automatically generated.

  A list of path names of existing files is used as input data, and word information is constructed by morphological analysis of node names and file names. The old hierarchical structure is simplified by cutting out words appearing at a plurality of places from the old hierarchical structure holding the existing file. At this point, the first new hierarchical structure is formed. Thereafter, second and subsequent new hierarchical structures are constructed using the extracted words. At that time, a new hierarchical structure to which each word belongs is determined so that each word does not appear at a plurality of locations on the new hierarchical structure. Specifically, a category ID is assigned to each word, and one new hierarchical structure is formed with words having the same category ID. In parallel with this, the correspondence between the file and the new hierarchical structure is automatically determined based on the correspondence between the file and the old hierarchical structure.

  It is expected that existing file pathnames will contain almost complete classification criteria for the files used by the organization that created them. By using the hierarchical structure management apparatus according to the present embodiment, these classification criteria can be reconfigured using an independent new hierarchical structure. In general organizations, the business and files are closely related, so the reconstructed multidimensional tree is in line with the actual business of the organization and can be useful for reviewing the business flow and organizational reform. Be expected.

  FIG. 1 is a hardware configuration diagram of the hierarchical structure management apparatus according to the present embodiment. The hierarchical structure management apparatus 1 includes a CPU 11, a RAM 12, a ROM 13, a hard disk (HD) 14, a communication I / F 15, and an input / output I / F 16. The ROM 13 and the HD 14 store an operating system and various programs (for example, a hierarchical structure management program). The ROM 13 and the HD 14 are read into the RAM 12 as necessary, and are executed by the CPU 11. The communication I / F 15 is an interface for communicating with other devices (for example, a client terminal in the case of a file server). The input / output I / F 16 is an interface for receiving input from an input device such as a keyboard and a mouse, and outputting data to a printer, a display, or the like. USB, RS232C, or the like is used as the input / output I / F 16. Further, if necessary, a drive corresponding to a removable disk such as a magneto-optical disk, a floppy disk (registered trademark), a CD-R, or a DVD-R can be connected.

  FIG. 2 is a functional block diagram of the hierarchical structure management apparatus according to the present embodiment. The hierarchical structure management apparatus 1 includes a reconstruction processing unit 21, a word information generation unit 22, a first hierarchical structure generation unit 23, an nth hierarchical structure generation unit 24, a correspondence information generation unit 25, a display control unit 26, a management file 27, A correspondence information unit 28 and a display screen 29 are provided.

  When any one or a plurality of word sets are input, the reconstruction processing unit 21 selects words that are not included in the input word set from the names of the nodes in the old hierarchical structure stored in the correspondence information 28. By deleting, a hierarchical structure including only the input word set and a correspondence between the hierarchical structure and the file are generated. As for the empty node whose name is empty as a result of the deletion of the word, the empty node is deleted, and the nodes with the same name among the sibling nodes that have a common parent are integrated. 28 is reconfigured. When deleting an empty node, the child node and leaf node of the empty node are associated with the nearest ancestor node. Details of this reconstruction processing will be described later with reference to FIGS.

  The word information generation unit 22 receives the name of the management file 27 and the name of each node in the old hierarchical structure associated with the correspondence information 28, and creates word information using morphological analysis and post-similarity determination processing. The word information is a one-dimensional array having words as elements, and the number of elements matches the number of words. The element of the one-dimensional array is a structure having three fields of word, word ID, and category ID. Details of the word information generation processing will be described later with reference to FIGS. 9 and 10.

  The first hierarchical structure generation unit 23 receives word information relating to all nodes in the hierarchical structure associated with the correspondence information 28 as input, and extracts duplicate words that appear redundantly in the node names for each hierarchical level. When a duplicate word is extracted, a reconstruction process is executed for a word set other than the duplicate word in the word information, and finally, after the duplicate word extraction and the reconstruction process are executed for all hierarchical levels. The correspondence information between the remaining nodes and files is determined as the first new hierarchical structure. The details of the first new hierarchical structure generation process will be described later with reference to FIGS.

The n-th hierarchical structure generation unit 24 receives the duplicate words extracted by the first hierarchical structure generation unit 23 as input, and classifies each duplicate word into a category (a plurality of new hierarchical structures). At that time, the same word is prevented from appearing more than once in the new hierarchical structure constituting the category. In other words, the word set that constitutes the name of each node in the new hierarchical structure is added with any one duplicate word for each new hierarchical structure, and the reconstruction process is sequentially executed as a word set, and the result is reconstructed. If any one added duplicate word appears in the hierarchical structure once, it is classified into that category, and if any one added duplicate word appears multiple times, any one Are classified into a new category. By being classified into categories, correspondence information of a new hierarchical structure is generated based on the categories.

When an operation such as addition, deletion, or movement of a file or a node is performed, the correspondence information generation unit 25 generates correspondence information of the correspondence information unit 28 corresponding to the operation information.
The display control unit 26 displays the old hierarchical structure and the new hierarchical structure on the display screen 29 based on the correspondence information in the correspondence information unit 28.

The management file 27 stores all files managed in the computer.
The correspondence information unit 28 stores correspondence information indicating correspondence relationships between nodes in the old hierarchical structure and the new hierarchical structure, and between nodes and files. The word information generated by the word information generating unit 22 may also be included in the correspondence information unit 28.

  Note that the word information generation unit 22, the first hierarchical structure generation unit 23, and the nth hierarchical structure generation unit 24 are not essential components, and may be configured as necessary.

  Next, the operation of each processing unit will be described. FIG. 3 is a flowchart showing the operation of the reconstruction processing unit in the hierarchical structure management apparatus according to the present embodiment. First, the correspondence information of the old hierarchical structure is copied to create a working hierarchical structure (S31). Thereafter, the process is executed on the hierarchical structure for work. FIG. 4 is an example of the old hierarchical structure. The following processing will be specifically described with respect to the hierarchical structure of FIG.

  An arbitrary word set is input to the reconstruction processing unit 21 (S31), and words not included in the word set are deleted from the working hierarchical structure (S32). Here, if {Education, Research, Watanabe, Suzuki, Sato, 2008} is input as a word set, the colored nodes in FIG. 5 that are not included in the word set are deleted, and FIG. The hierarchical structure shown is formed.

  It should be noted that the deleted word and the node name do not necessarily match completely, and the deleted word may be included in a part of the node name. In that case, not all the names of the nodes may be deleted, but only the words in the names may be deleted.

  The following processing is sequentially performed for each node in order of increasing distance from the root node of the working hierarchical structure. A sibling node in which each node to be processed is a target node, and a word or a set of words constituting the name of the target node (when the name is composed of a plurality of words) has a common parent with the target node When the word or the word set constituting the name matches, the target node and the sibling node are integrated (S34). At this time, a child node of the target node and a child node of a sibling node integrated with the target node are all child nodes of the integrated node. If the name of the target node becomes empty due to the deletion of the word (the name of the target node does not include a word), the target node is deleted (S35) and the child of the target node The node and the leaf node are directly associated with the nearest ancestor node (for example, the parent node) of the target node. Similarly, a node that becomes a new sibling node by deleting the target node is also subject to integration.

  The hierarchical structure generated as a result of the above processing is shown in FIG. A hierarchical structure formed only of words included in the input word set is newly generated in a simplified form, and correspondence between nodes and correspondence between files and nodes are also correspondences in the old hierarchical structure. It is generated without contradicting (hierarchical structure in FIG. 4).

That is, the processing of the reconstruction processing unit 21 can generate a new hierarchical structure in view of any one or a plurality of input word sets, and the user can select an arbitrary viewpoint from the existing old hierarchical structure. Customize and generate a new hierarchical structure based on, making it easy to use.

  FIG. 8 is a flowchart showing overall processing of the hierarchical structure management apparatus according to the present embodiment. First, the word information generation part 22 performs a word information generation process (S81). This word information generation process will be described in detail.

  FIG. 9 is a flowchart showing word information generation processing in the hierarchical structure management apparatus according to the present embodiment. First, the initial state of word information is emptied (S91). Each node (node name and file name) constituting the old hierarchical structure is decomposed into a set of words using a morphological analysis algorithm (S92). For example, in the case of the node name “specification_May20”, it is decomposed into “specification, May, 20”. Here, the words to be decomposed may be limited to independent words. It is determined whether or not the same word has already been registered, and if it is not registered, it is registered in the word information and given a unique word ID (S93). At this time, the category ID of each word is undefined.

  FIG. 10 shows word information generated based on the old hierarchical structure of FIG. In this word information, even if the character strings representing the words are different, if the meanings are the same (synonyms), the same word ID is given, and even if the word faces are the same, the meanings are different ( Different word IDs are given to the polysemy. In the processing of FIG. 9, different word IDs are assigned if the character strings of the words are different. Details of the processing of the above synonyms and polysemy will be described later in the second embodiment.

  The category ID in FIG. 10 is for classifying different words for each viewpoint, and the same category ID is assigned to words belonging to the same viewpoint. That is, words having the same category ID gather to form one new hierarchical structure. Each word belongs to only one category. There may be a word that does not belong to the category, but the word does not appear in the new hierarchical structure. Further assume that no words belong to more than one category.

  At the time of the processing of FIG. 9, this category ID is undefined, but is assigned by the processing of the first hierarchical structure generating unit 23 and the nth hierarchical structure generating unit 24 described later. By constructing such word information, it is possible to easily reconfigure the correspondence between nodes and the correspondence between nodes and files in the new hierarchical structure. That is, the correspondence information can be calculated by obtaining a set of words belonging to each category in the word information and executing the reconstruction process of the old hierarchical structure using the set.

  Returning to FIG. 8, the first hierarchical structure generating unit 23 performs the first hierarchical structure generating process (S82). Here, it is a process of cutting out and simplifying all the words (duplicated words) that appear redundantly from the old hierarchical structure, and finally setting the remaining nodes and files to the first new hierarchical structure. The first hierarchical structure generation process will be described in detail.

  FIG. 11 is a flowchart showing a first hierarchical structure generation process in the hierarchical structure management apparatus according to the present embodiment. First, the word set W is set as all the words registered in the word information generated by the word information generation unit 22 (S111), and the work hierarchical structure to be processed is set as the old hierarchical structure (S112). . The initial value of the duplicate word list X cut out from the old hierarchical structure is emptied (S113). The following processing is performed in the order of the distance from the root node for each node constituting the working hierarchical structure.

When each word w constituting the node of the working hierarchical structure appears twice or more in the hierarchical structure, the word w is deleted from the word set W (S114). This process is shown in FIG. Here, processing is performed on each hierarchical level in the order from the root node. In FIG. 12, when a word that appears redundantly is extracted from the names of the four nodes at the first hierarchical level, the word w {Suzuki, Sato} is extracted. This w {Suzuki, Sato} is deleted from the word set W.

  Using the new word set W from which the word w has been deleted, a reconstruction process is performed on the working hierarchical structure (S115). That is, with the word set excluding w {Suzuki, Sato} as input, the reconstruction processing unit 21 executes the reconstruction processing described in FIG. 3 for the working hierarchical structure. The word w {Suzuki, Sato} is added to the duplicate word list X (S116). FIG. 13 shows the result of executing the reconstruction process in S115. In FIG. 13, the word w {Suzuki, Sato} is deleted from the working hierarchical structure, and sibling nodes with the same name are integrated and simplified.

  Similarly, in the hierarchical structure of FIG. 13, words that appear redundantly among the words at the first hierarchical level are extracted and deleted from the word set W. Specifically, the word w {education, research, 2008} is deleted from the word set W and added to the duplicate word list X. Then, reconstruction processing is performed using the word set W. The result is shown in FIG.

  In FIG. 14A, the word w {education, research, 2008} is further deleted and simplified as compared with FIG. In FIG. 14A, since there are no duplicate words in the first hierarchy level, duplicate words at the second hierarchy level are extracted. Here, the word w {Watanabe} is extracted, deleted from the word set W, and added to the duplicate word list X. Then, reconstruction processing is performed using the word set W. The result is shown in FIG. Duplicate words are extracted at all hierarchical levels, and when there are no duplicate words, category ID = 1 is set for all words finally included in the word set W (S117). That is, the first new hierarchical structure as shown in FIG. 14B is determined.

  Although FIG. 12, FIG. 13 and FIG. 14A do not show the file, it is assumed that the file is actually associated with each node according to the correspondence relationship of FIG. Further, as shown in FIG. 14B, there may be a node that is difficult to say, such as “social contribution”. For these nodes, the function described in the second embodiment is provided. Can be eliminated.

  In this way, the above processing makes it possible to eliminate the duplication of nodes including duplicate words in the name and make them exclusive, and to easily automatically generate a hierarchical structure classified from one viewpoint.

  Returning to FIG. 8, the n-th hierarchical structure generation unit 24 performs the n-th hierarchical structure generation process (S83). Here, one category ID is assigned to each word. Since category ID = 1 (first new hierarchical structure) is determined in the process of S82, category IDs after 2 (second and subsequent new hierarchical structures) are determined here. The nth hierarchical structure generation process will be described in detail.

  FIG. 15 is a flowchart showing the n-th hierarchical structure generation process in the hierarchical structure management apparatus according to the present embodiment. First, an existing category C is initialized to an empty set (S151), and a reserved list S for registering words to be reserved is initialized to an empty list (S152). For the word w registered in the duplicate word list X, the following processing is performed in the order added to the duplicate word list X.

The word w is added to the existing category C, and the reconstruction process is performed on the working hierarchical structure with the word set belonging to the category C (S153). In the initial state, category C is an empty set (no new hierarchical structure after the second is formed at all), so a hierarchical structure composed only of the duplicate words added first to the duplicate word list X is formed. Is done. If the word is already classified into category C (at least a part of the second and subsequent new hierarchical structures is formed) instead of the initial state, reconstruction processing is performed on the category C word set. .

  As a result of the reconstruction process, if the word w appears only once in the hierarchical structure, the word w can be added to the category C (S154). When the word w can be added to the category C, the number of files associated with the root node of the hierarchical structure is calculated as a result of the reconstruction process (S155). This is because the number of files calculated here is used in later processing. If there is only one category candidate to which the word w can be added, the word w is classified into that category (S156). If there is no category candidate to which the word w can be added, the word w is classified into a new category (S157). Here, these processes will be specifically described with reference to the following Table 1 and FIG. Table 1 below shows a list X of duplicate words obtained by the process of FIG.

  FIG. 16 (A) assumes that “Suzuki” first added to the duplicate word list X is classified into the category C with the category C being empty, and then second to the duplicate word list X. A determination as to whether or not the added “Sato” can be added to the category C (specifically, {Suzuki, Sato} performs a reconfiguration process on the working hierarchical structure, and the result is “Sato”. In this case, it is determined that it can be added, and “Suzuki” and “Sato” are classified into the same category C.

  In this state, it is determined whether or not “2008” added third to the duplicate word list X can be added to the category C. As a result of performing the reconstruction process on the working hierarchical structure in {Suzuki, Sato, 2008}, “2008” appears multiple times as shown in FIG. Therefore, since there is no category to which “2008” can be added, “2008” is classified into a new category. That is, as shown in FIG. 16A, a category of {Suzuki, Sato} and a category of {2008} are formed. Similarly, the results of processing “education” and “research” in succession are shown in FIG. At this stage, three categories are formed. Thereafter, the process is repeated for the duplicate words in Table 1 above.

  It should be noted that the duplicate word list X is not a simple set, but is preferably defined by a FIFO (First-in First-out) queue that holds the words in order. In addition, the duplicate word list X and the word set W need only hold the word ID.

Returning to FIG. 15, when there are a plurality of category candidates to which the word w can be added, the word w is classified into the category candidates with the smallest number of files associated with the root node calculated in S155 (S158). . Here, the processing of S158 will be described with reference to FIGS. FIG. 17 shows three new hierarchical structures composed of words classified into three categories ({Suzuki, Sato}, {2008}, {Education, Research}). In this state, “Watanabe” is shown. Classify. When the word “Watanabe” is added for each category and reconfiguration processing is performed, category candidates in which “Watanabe” appears only once are {Suzuki, Sato} and {2008}. In this case, since there are two category candidates, the word “Watanabe” is classified into a category candidate having the smallest number of files associated with the root node as a result of the reconstruction process.

  FIG. 18A shows the result of performing reconstruction processing using {Suzuki, Sato, Watanabe} as a word set, and FIG. 18B shows the reconstruction processing using {2008, Watanabe} as a word set. The results are shown. As is clear from FIG. 18, in the case of {Suzuki, Sato, Watanabe}, the number of files associated with the root node is 0, and in the case of {2008, Watanabe}, the file is associated with the root node. Since the number of files is 11, the word “Watanabe” is classified into the category {Suzuki, Sato, Watanabe}. In other words, the fact that the file is associated with the root node means that it is an unorganized file that does not belong to any node, so the hierarchical structure with fewer unorganized files is a more organized hierarchy. It can be determined that the word is structured, and processing is performed so as to classify words into a more organized hierarchical structure.

  Returning to FIG. 15, when there are a plurality of category candidates to which the word w can be added and there are a plurality of category candidates with the smallest number of files corresponding to the root node, the classification of the word w is temporarily put on hold in the hold list S. The word w is added (S159). Each time one or a plurality of other words w registered in the duplicate word list X are classified, or after all other words w are classified, the reserved list for the word ws registered in the reserved list S The same processing as S153 to S158 is performed in the order added to S (S160).

  That is, even if the word w cannot be classified into the category and is registered in the holding list S, the word ws registered in the holding list S can be classified by classifying other words into the category. There is a possibility. For this reason, after processing another word w first, the word ws registered in the hold list S is processed again.

  When there are a plurality of category candidates that can be added to the word ws added to the hold list S and there are a plurality of category candidates that minimize the number of files corresponding to the root node, the word ws is randomly assigned to one of the category candidates. Classify (S161).

  Through the above processing, category IDs are assigned to all the words w, and a plurality of new hierarchical structures are formed. The result is shown in FIG. By executing the above processing from the one-dimensional old hierarchical structure shown in FIG. 4, a new hierarchical structure based on five different classification criteria is finally formed as shown in FIG.

  If there are a plurality of category candidates to which the word w can be added at the stage of processing in S158, the category candidates may be randomly classified without comparing the number of files associated with the root node. In that case, it is not necessary to perform the process in step S155 or the process related to the number of files associated with the root node.

  Further, the word w may be randomly classified into the category candidates without adding the word w to the holding list S in the process of S158. In that case, the processing related to the holding list S and the holding list S is unnecessary.

  Furthermore, as the data structure for holding the category, it is desirable to use a list (or set) as described above, and its elements are a word ID and a category ID. Further, since it has been confirmed that the word w registered in the duplicate word list X appears in the category of category ID = 1 set in FIG. It shall not be included.

  Furthermore, for each word included in the duplicate word list X and the pending list S, it is determined whether or not it can be added for each existing category, but in order to hold the result, a one-dimensional array is defined. May be. In that case, the number of elements in the array matches the number of categories. In addition, the elements of the array are a Boolean type value indicating whether or not the word w (or word ws) can be added to the category, and a tree reconstructed by adding the word w (or word ws) to the category C. A structure having two fields of the number of files corresponding to the root node may be used.

  Returning to FIG. 8, a process for refining the new hierarchical structure is performed on the new hierarchical structure generated in the processes of S81 to S83 (S84). This hierarchical structure refinement process will be described in detail in the second embodiment.

When a new node or file is added to the old hierarchical structure, it is necessary to reconstruct word information. To that end, it is possible to cope with the above changes by adding the following changes. (1) The process of emptying the initial state of the word information in S91 in FIG. 9 is changed to a process of setting the already constructed word information to the initial state. (2) The process of setting the word set W in S111 of FIG. 11 as all the words registered in the word information, the category ID is 1 or undefined among the words registered in the word information. In addition to the processing to be assumed, the processing in which the processing target hierarchical structure in S112 is changed to the old hierarchical structure, and the processing target hierarchical structure is reconfigured using the word set W. Change to processing. (3) The process of initializing the existing category in S151 of FIG. 15 to the empty set is changed to a process of setting the category set registered in the word information as an existing category.
By changing to the above processing, it is possible to cope with a case where a new node or file is added to the old hierarchical structure.

(Second embodiment of the present invention)
The hierarchical structure management apparatus according to the present embodiment will be described with reference to FIGS. The hierarchical structure management apparatus according to the present embodiment has a function of checking and editing a new hierarchical structure by high-level operation when a new hierarchical structure is formed based on a plurality of classification criteria. The target for high-level operation may be a plurality of new hierarchical structures formed by the hierarchical structure management apparatus according to the first embodiment, or a plurality of new levels based on different classification criteria formed by other methods. It may be a hierarchical structure.

  In the present embodiment, descriptions overlapping with those in the first embodiment are omitted.

  FIG. 20 is a functional block diagram of the hierarchical structure management apparatus according to this embodiment. The difference from the case of FIG. 2 is that a check processing unit 201 and an editing processing unit 202 are newly provided. As described above, in the present embodiment, since it is not always necessary to process a plurality of new hierarchical structures formed by the hierarchical structure management apparatus according to the first embodiment, the word information generation unit 22, the first The 1-layer structure generation unit 23 and the n-th layer structure generation unit 24 may be omitted.

  The check processing unit 201 extracts words that have a plurality of category candidates that can be added and words that appear redundantly in one new hierarchical structure. Further, when a word or word set designated by the user (for example, any node in any new hierarchical structure) is designated, category candidates to which the designated word or word set can be added are extracted. Similar to the case of the first embodiment, the categories that can be added are added when the word constituting each category is reconfigured with a word set obtained by adding a specified word or a word set. When a word or word set appears only once, it can be added to the category.

The edit processing unit 202 controls the movement of a word or a word set between a plurality of new hierarchical structures, deletes the specified word or word set, sets the specified word or word set as a synonym, and specifies the specified word Alternatively, processing such as designating a word set as a polysemy is performed to control editing of the new hierarchical structure.

  The display control unit 26 displays the old hierarchical structure and the new hierarchical structure on the display screen 29 based on the management file 27 and the correspondence information unit 28, and reflects the processing result of the check processing unit 201 and the processing result of the editing processing unit 202. Thus, the display of the old hierarchical structure and the new hierarchical structure of the display screen 29 is controlled.

  Next, the processing of the check processing unit 201 will be described in detail. As a first function of the check processing unit 201 and the display control unit 26, a word having a plurality of category candidates that can be added is displayed separately from words that are not. In the designated new hierarchical structure, a word that appears only once even when moved to another new hierarchical structure is displayed separately from other words. An example is shown in FIG. As shown in FIG. 21, a large number of nodes are displayed in different modes.

  For example, “2010”, “2009”, “before 2007”, “2007”, “before 2006” are shown in the same display mode, and the display mode is category ID = 3. This means that “2010”, “2009”, “before 2007”, “2007”, and “before 2006” can be added to category ID = 3. That is, for each word of “2010”, “2009”, “before 2007”, “2007”, and “before 2006”, when the reconfiguration process is performed in addition to the category ID = 3, each word is once It will only appear. Similarly, “social contribution” and “lecture material” can be added to category ID = 4, “tex” can be added to category ID = 5, “2008” can be added to category ID = 1, and “experimental data” can be added to category ID = 2. It can be easily confirmed visually.

  As a second function of the check processing unit 201 and the display control unit 26, words that appear redundantly in one new hierarchical structure are distinguished and displayed. When the same word appears twice or more in one new hierarchical structure, the corresponding word is distinguished and displayed. This provides the user with a clue to consider whether the word should be moved to another new hierarchical structure.

  When a new hierarchical structure is created according to the first embodiment, the same word does not appear more than once in one new hierarchical structure immediately after creation, but when a file or node is added after that, When the new hierarchical structure is not created according to the first embodiment, the same word may appear twice or more in one new hierarchical structure, and in this case, this function is effective.

  FIG. 22 shows an example of display. FIG. 22 shows one new hierarchical structure among a plurality of new hierarchical structures. As shown in FIG. 22, the words “education”, “research”, and “2008” appearing in duplicate are distinguished and displayed in relation to the overlapping word and in relation to the non-overlapping word. Has been. As a result, words that appear twice or more in one new hierarchical structure can be easily visually confirmed. In addition, when the user selects one overlapping word, a function may be provided in which words that match the word are selected in a lump.

  As a third function of the check processing unit 201 and the display control unit 26, when a word or word set designated by the user is designated, category candidates to which the designated word or word set can be added are extracted. And displayed in association with the designated word or word set. That is, a list of categories in which the designated word or word set does not appear redundantly is displayed in the vicinity of the designated word or word set. By this processing, the user can confirm the category candidates that can be added before moving the designated word or word set to another new hierarchical structure.

  FIG. 23 shows an example of display. FIG. 23 shows that the category candidate that can be added is category ID = 4 when the word set of “lecture material” is designated in the new hierarchical structure of category ID = 2. That is, when the word set of “lecture material” is added to category ID = 4 and the reconstruction process is performed, it means that “lecture material” appears only once.

  Next, the processing of the edit processing unit 202 will be described in detail. As a first function of the edit processing unit 202, the movement of a word or a word set between a plurality of new hierarchical structures is controlled. The user designates one or more words in any of a plurality of new hierarchical structures. After that, the new hierarchical structure of the destination is specified. The edit processing unit 202 performs a reconstruction process by substituting the category ID corresponding to the new hierarchical structure of the movement destination into the category ID of the designated word. By doing so, the designated word is excluded from the category ID of the movement source, added to the category ID of the movement destination, and the movement of the word between the new hierarchical structures is realized.

  24 and 25 show an example of the movement process. As shown in FIG. 24, for example, “social contribution” and “lecture material” can be added to category ID = 4, “tex” can be added to category ID = 5, and “2008” is category ID = 1. And move those words to a category where they can be added. The result of the movement is shown in FIG. In FIG. 24, the user only has to specify the target word and the destination category, and the new hierarchical structure after the movement is automatically calculated by the reconstruction process as shown in FIG. Further, as a result of generating a new hierarchical structure in the first embodiment, even if some of the classifications are not valid, this moving operation allows accurate classification without any contradiction as shown in FIG. A new hierarchical structure can be generated. At this time, the category ID = 3 is deleted because the node becomes empty. After the movement, the word information is updated as shown in FIG.

  In this way, the user can move a word only by specifying a destination category (a new hierarchical structure corresponding to the category), and does not need to specify a hierarchical level. The designation of the new hierarchical structure of the movement source and the designation of the word set in the new hierarchical structure are left to the user, and all new hierarchical structures and words can be moved.

  As a second function of the edit processing unit 202, the designated word or word set is deleted. In this process, words that are included in the node name or file name but are unnecessary as a classification criterion are deleted, and the ease of understanding the new hierarchical structure is improved. A user designates one or more words in any one of a plurality of new hierarchical structures. The edit processing unit 202 sets the word ID and category ID of the specified word to an invalid value in the word information. Thereafter, the designated word is deleted by performing a reconstruction process in a state where the designated word is removed.

  As a third function of the edit processing unit 202, a plurality of designated words or word sets are set as synonyms having the same meaning. Here, words having different meanings but having the same meaning are set as synonyms. The user designates a plurality of words or word sets in any one of the plurality of new hierarchical structures. At that time, the user can also specify a representative value used when displaying the word. For example, when “2010” and “2010” are set as synonyms, and the representative value for display is designated as “2010”, the new hierarchical structure is “2010”. Only the word appears, and the indication “2010” does not appear in the new hierarchical structure.

The edit processing unit 202 updates word information to match word IDs of a plurality of designated words or word sets in accordance with a user operation. Then, a new hierarchical structure is constructed by performing reconstruction processing using new word information. With this process, words with the same meaning are expected to have a sibling relationship in the hierarchical structure, so a new hierarchical structure is organized by combining the words with the same meaning into one. can do. It should be noted that the representative value set above may be edited such as changing or switching as necessary.

  As a fourth function of the edit processing unit 202, a designated word or word set is set as a polysemy having a different meaning. Here, words that have the same wording but different meanings are set as ambiguous words. The setting of polysemy needs to be given different word IDs before building a new hierarchical structure. Therefore, the display control unit 26 displays the old hierarchical structure, and the user designates using the displayed old hierarchical structure. The user designates a plurality of words having the same wording and different meanings, and performs an operation to the effect that they are ambiguous words. The edit processing unit 202 changes the word IDs of a plurality of designated words. An example is shown in FIG.

  In FIG. 27, “Suzuki” appears in four places, and it is assumed that Suzuki A, Suzuki B, and Suzuki C mean the same person, and Suzuki D means a different person. First, as a first method, the user designates Suzuki A and Suzuki D and sets a polysemy. Similarly, polynyms are set for Suzuki B and Suzuki D, and Suzuki C and Suzuki D. Since Suzuki A, B, and C have the same meaning, the synonyms described above are set for these three words. By doing so, the same word ID is given to Suzuki A, B, and C, and a different word ID is given only to Suzuki D, so that a new hierarchical structure can be constructed in consideration of polysemy.

  As a second method, the user designates Suzuki C and Suzuki D and sets a polysemy. Similarly, 2010F and Suzuki D are specified to set a polysemy. When 2010F is specified, the edit processing unit 202 sets all “Suzuki” subordinate to 2010F to the same word ID, and sets a polysemy in relation to Suzuki D. Then, the synonyms described above for 2010F and Suzuki C are set. By doing so, the same word ID is given to Suzuki A, B, and C, and a different word ID is given only to Suzuki D, so that a new hierarchical structure can be constructed in consideration of polysemy. Compared with the first method, the operation for setting a polysemy can be simplified.

  As a third method, the user designates Suzuki D and root node (root) F and sets a polysemy. The edit processing unit 202 sets all “Suzuki” subordinates except the directly designated Suzuki D to the same word ID, and sets a polysemy in relation to the directly designated Suzuki D. By doing so, the same word ID is given to Suzuki A, B, and C, and a different word ID is given only to Suzuki D, so that a new hierarchical structure can be constructed in consideration of polysemy. Compared to the second method, it is possible to set polysemy with a further simplified operation.

In the case of the first method, the system side may perform control so that only one word having the same wording as the designated word can be designated after designating one word in the operation for designating a polysemy.
The new hierarchical structure can be refined by the processing of the check processing unit 201, the editing processing unit 202, and the display control unit 26 as described above.

(Other embodiments)
The following additional notes will be given with respect to the above embodiments.
(Supplementary Note 1) File storage means for storing a plurality of management files to be managed, correspondence information associating the management file with each node in the existing old hierarchical structure, and the management file newly generated When correspondence information storage means for storing correspondence information in association with each node in the new hierarchical structure and a word set consisting of any one or a plurality of words are input, from the name of the node in the correspondence information Delete a word that is not included in the input word set, delete the empty node whose empty name is the result of deletion, and delete the same empty node among sibling nodes that have a common parent Reconfiguring means for reconfiguring the correspondence information by integrating the nodes having names, and when the reconfiguring means deletes the empty node, Hierarchical Management and wherein the associating the nearest ancestor node for the leaf node.

  (Supplementary note 2) In the hierarchical structure management apparatus according to supplementary note 1, a plurality of hierarchical structures are formed according to different classification standards, and the management file and each of the management files for each of the formed hierarchical structures Correspondence information storage means for storing correspondence information with nodes in the hierarchical structure, display control means for displaying each hierarchical structure stored in the correspondence information storage means, and all the correspondence information in the correspondence information stored in the correspondence information storage means For each target new hierarchical structure other than the hierarchical structure that includes the target word, for each target new hierarchical structure that includes the target word, among the words that constitute the node name, the name of the node in the candidate new hierarchical structure The processing of the reconstructing means is sequentially executed by adding the target word to the words constituting the word set as a word set. A target word extraction means for extracting the target word when there is a candidate new hierarchical structure in which the recording target word appears only once, and the display control means for the target word extracted by the target word extraction means, A hierarchical structure management apparatus, characterized in that the display mode is different from that of other words.

  (Supplementary note 3) In the hierarchical structure management device according to supplementary note 2, the display control means uses the arbitrary one target word to indicate information indicating a candidate new hierarchical structure in which the arbitrary one target word appears only once. A hierarchical structure management apparatus characterized by being displayed in association with each other.

  (Supplementary note 4) In the hierarchical structure management apparatus according to supplementary note 1, a plurality of hierarchical structures are formed with different classification criteria, and the management file and each of the management files for each of the formed hierarchical structures Corresponding information storage means for storing correspondence information with nodes in a hierarchical structure; and display control means for displaying each hierarchical structure stored in the correspondence information storage means, wherein the display control means A hierarchical structure management apparatus, wherein, in a hierarchical structure, when the same word appears in the same hierarchical structure, the same word is displayed separately from other words.

  (Supplementary Note 5) In the hierarchical structure management device according to Supplementary Note 1, a plurality of hierarchical structures are formed with different classification criteria, and the management file and each of the management files for each of the formed hierarchical structures Correspondence information storage means for storing correspondence information with nodes in the hierarchical structure, display control means for displaying each hierarchical structure stored in the correspondence information storage means, and names of the nodes of any one of the displayed hierarchical structures The specified word is added to the words constituting the destination hierarchical structure when the designated word consisting of any one or more words constituting the destination and the destination hierarchical structure for moving the designated word are designated The processing of the reconstructing means is executed as a word set, and the designation is made from the words constituting the names of the nodes in the hierarchical structure including the designated word. Word said by performing the processing of reconstruction means those excluded as word set the hierarchy management apparatus characterized by comprising a movement control means for controlling the movement of words between hierarchy.

(Supplementary note 6) In the hierarchical structure management device according to supplementary note 1, a plurality of hierarchical structures are formed according to different classification criteria, and the management file and each of the management files are defined for each of the formed hierarchical structures. Correspondence information storage means for storing correspondence information with nodes in the hierarchical structure, display control means for displaying each hierarchical structure stored in the correspondence information storage means, and names of the nodes of any one of the displayed hierarchical structures When a specified word consisting of any one or a plurality of words that constitutes and deletion instruction information that deletes the specified word is input, the name of the node in the hierarchical structure that includes the input specified word The processing of the reconstructing means is executed as a word set from the words constituting the word excluding the designated word, and the deletion for controlling the deletion of the word from the hierarchical structure is controlled. Hierarchical Management device characterized by a control unit.

  (Supplementary note 7) In the hierarchical structure management device according to supplementary note 1, a plurality of hierarchical structures are formed according to different classification criteria, and the management file and each of the management files are defined for each of the formed hierarchical structures. Correspondence information storage means for storing correspondence information with nodes in the hierarchical structure, display control means for displaying each hierarchical structure stored in the correspondence information storage means, and words constituting names of the nodes of the displayed hierarchical structure Is specified, and the specified information is unified to any one word, or the information indicating the specified word is unified to one information, when the instruction information having the designated designated word as a synonym is input And a synonym control unit that executes processing of the reconfiguration unit and controls synonyms.

  (Supplementary note 8) In the hierarchical structure management apparatus according to supplementary note 1, the name of the management file and the name of each node in the old hierarchical structure are divided by morphological analysis to create word information, and each word in the word information A word information generation means for assigning a word ID indicating the word, an old correspondence information display control means for displaying the correspondence information stored in the old correspondence information storage means, and a node of a hierarchical structure in the displayed correspondence information When a plurality of words having the same name are designated and information indicating that the designated name is a word having a different meaning is input, the designated name is designated. A hierarchical structure management device, comprising: a polysemy control means for controlling polysemy by differentiating each word ID of a word having the same name.

  Although the present invention has been described with the above embodiments, the technical scope of the present invention is not limited to the scope described in the embodiments, and various modifications or improvements can be added to these embodiments. . And embodiment which added such a change or improvement is also contained in the technical scope of the present invention. This is apparent from the claims and the means for solving the problems.

1 Hierarchical Structure Generation Device 11 CPU
12 RAM
13 ROM
14 HD
15 Communication I / F
16 Input / output I / F
21 Reconfiguration Processing Unit 22 Word Information Generation Unit 23 First Hierarchical Structure Generation Unit 24 nth Hierarchical Structure Generation Unit 25 Corresponding Information Generation Unit 26 Display Control Unit 27 Management File 28 Corresponding Information Unit 29 Display Screen 201 Check Processing Unit 202 Editing Process Part

Claims (13)

A file storage means for storing a plurality of management files to be managed;
Correspondence information storage means for storing correspondence information in which the management file is associated with each node in the existing old hierarchical structure, and correspondence information in which the management file is associated with each node in the newly generated new hierarchical structure When,
When a word set consisting of one or more arbitrary words is input, words not included in the input word set are deleted from the names of the nodes in the correspondence information, and the deleted result name becomes empty The empty node is deleted, and among the sibling nodes having a common parent, reconstructing means for reconfiguring the correspondence information by integrating the nodes having the same name,
When the reconfiguration unit deletes the empty node, the child node and the leaf node of the empty node are associated with the nearest ancestor node. In the hierarchical structure management apparatus according to claim 1,
A word information generating means for dividing the name of the management file and the name of each node in the old hierarchical structure by morphological analysis to create word information;
For the word information related to all the nodes in the old hierarchical structure, for each hierarchical level, a duplicate word that appears redundantly in the name of the node is extracted, and when the duplicate word is extracted, the word information Among the word sets other than the duplicate word, the process of the reconstruction means is executed, and the node that finally remains after the extraction of the duplicate word and the process of the reconstruction means are executed for all hierarchical levels A hierarchical structure management apparatus comprising first hierarchical structure determining means for determining correspondence information with a file as a first new hierarchical structure. In the hierarchical structure management apparatus according to claim 2,
If there is no other new hierarchical structure except the first new hierarchical structure for any one duplicated word extracted by the first hierarchical structure determining means, only that arbitrary duplicated word is used as a word set. , Determining the correspondence information obtained by executing the processing of the reconfiguration means as another new hierarchical structure after the second,
When there is another new hierarchical structure excluding the first new hierarchical structure, the arbitrary one of the other new hierarchical structures is added to the word set constituting the name of each node in the other new hierarchical structure. The processing of the reconstructing means is sequentially executed with a word set added with duplicate words, and when the added duplicate word appears only once in the reconstructed hierarchical structure as a result of the execution, the reconstruction is performed The correspondence information obtained by the processing of the means is determined as another new hierarchical structure after the second,
When the added duplicate word appears a plurality of times, only the added duplicate word is taken as a word set, and correspondence information obtained by executing the processing of the reconstruction means is taken as another new hierarchical structure after the second A hierarchical structure management apparatus comprising n-th hierarchical structure determining means for determining a second and subsequent new hierarchical structures by executing a process for determining all the duplicate words. In the hierarchical structure management apparatus according to claim 3,
The n-th hierarchical structure determining means is
Results obtained when the processing of the reconstructing means is sequentially executed by adding the arbitrary one duplicate word to the word set constituting the name of each node in the other new hierarchical structure as a word set When there are a plurality of other new hierarchical structures in which the added arbitrary duplicate word appears only once in the reconstructed hierarchical structure, it is associated with each root node in the plurality of other new hierarchical structures The number of files to be compared is compared, and correspondence information obtained by adding the arbitrary one duplicate word to the other new hierarchical structure in which the number of the files is minimized is determined as a new hierarchical structure. Hierarchical structure management device. In the hierarchical structure management apparatus according to claim 4,
The n-th hierarchical structure determining means is
When there are a plurality of other new hierarchical structures having the same minimum number of files when comparing the number of files associated with the root node in the plurality of other new hierarchical structures, the added duplicate word Is obtained as a reserved word, and after the processing of the reconstructing means is completed for other duplicate words other than the duplicate word that has become the reserved word, the processing of the reconfiguring means is again executed for the reserved word. A hierarchical structure management device characterized in that the correspondence information is determined as a new hierarchical structure. In the hierarchical structure management device according to any one of claims 3 to 5,
Display control means for displaying the first new hierarchical structure determined by the first hierarchical structure determining means and the one or more new hierarchical structures determined by the nth hierarchical structure determining means;
Among the words constituting the names of all nodes in the first new hierarchical structure and the other new hierarchical structures, the target word consisting of any one or more words other than the new hierarchical structure including the target word For each candidate new hierarchical structure, the processing of the reconstruction means is sequentially executed by adding the target word to the word constituting the name of the node in the candidate new hierarchical structure as a word set. When there is a candidate new hierarchical structure in which the target word appears only once after configuration, the target word extraction means for extracting the target word,
The hierarchical structure management apparatus, wherein the display control means displays the target word extracted by the target word extraction means in a different display mode from other words. In the hierarchical structure management apparatus according to claim 6,
The hierarchical structure management apparatus, wherein the display control means displays information indicating a candidate new hierarchical structure in which the target word appears only once in association with the target word. The hierarchical structure management device according to any one of claims 3 to 7,
Display control means for displaying the first new hierarchical structure determined by the first hierarchical structure determining means and one or more new hierarchical structures determined by the nth hierarchical structure determining means;
The display control means is
In the new hierarchical structure, when the same word appears in the same hierarchical structure, the same word is distinguished from other words and displayed. The hierarchical structure management device according to any one of claims 3 to 8,
Display control means for displaying the first new hierarchical structure determined by the first hierarchical structure determining means and the one or more new hierarchical structures determined by the nth hierarchical structure determining means;
When a designated word consisting of any one or a plurality of words constituting the name of the node of any one of the displayed new hierarchical structures and a destination new hierarchical structure for moving the designated words are designated,
The processing of the reconstructing means is executed with the word constituting the destination new hierarchical structure added with the designated word as a word set, and the name of the node in the new hierarchical structure including the designated word is constructed Hierarchical structure management comprising: a movement control means for controlling the movement of words between new hierarchical structures by executing the processing of the reconstructing means as a set of words excluding the designated word from words apparatus. The hierarchical structure management device according to any one of claims 3 to 9,
Display control means for displaying the first new hierarchical structure determined by the first hierarchical structure determining means and the one or more new hierarchical structures determined by the nth hierarchical structure determining means;
When a designated word consisting of any one or a plurality of words constituting the name of any one of the displayed new hierarchical structure nodes and deletion instruction information for deleting the designated word are inputted, Control the deletion of words from the new hierarchical structure by executing the processing of the reconstruction means as a set of words excluding the designated words from the words constituting the node names in the new hierarchical structure including the designated words A hierarchical structure management apparatus comprising: deletion control means for performing The hierarchical structure management device according to any one of claims 3 to 10,
Display control means for displaying the first new hierarchical structure determined by the first hierarchical structure determining means and the one or more new hierarchical structures determined by the nth hierarchical structure determining means;
When a plurality of words constituting the name of the displayed node of the new hierarchical structure are designated and instruction information having the designated designated word as a synonym is input, the designated word is changed to any one word. A hierarchical structure management device comprising: synonym control means for unifying or unifying information indicating the designated word into one information, executing processing of the reconstruction means, and controlling synonyms . The hierarchical structure management device according to any one of claims 3 to 10,
A word information generating means for creating word information by dividing the name of the management file and the name of each node in the old hierarchical structure by morphological analysis, and allocating a word ID indicating the word to each word in the word information; ,
And the old correspondence information display control means for displaying the corresponding information stored prior SL-enabled information storage means,
Information indicating that a plurality of words having the same name are designated among the words constituting the names of the nodes of the hierarchical structure in the displayed correspondence information, and the words having the designated names are words having different meanings. A hierarchical structure management device, comprising: a polysemy control means for controlling polysemy by differentiating each word ID of the designated word having the same name.
In the hierarchical structure management apparatus according to claim 1,
Corresponding information storage means for forming a plurality of hierarchical structures with different classification criteria and storing correspondence information between the management files and nodes in each hierarchical structure for all the management files for each of the formed hierarchical structures When,
Display control means for displaying each hierarchical structure stored in the correspondence information storage means;
Among the words constituting the names of all the nodes in the correspondence information stored in the correspondence information storage means, for each candidate word composed of any one or more words, each candidate new one other than the hierarchical structure including the target word For the hierarchical structure, the processing of the reconstructing means is sequentially executed as a word set by adding the target word to the word constituting the name of the node in the candidate new hierarchical structure. A target word extraction unit that extracts the target word when there is a candidate new hierarchical structure in which the target word appears only once;
The hierarchical structure management apparatus, wherein the display control means displays the target word extracted by the target word extraction means in a different display mode from other words.