JP5900486B2 - Related specification mapping system, related specification mapping method and program - Google Patents

Description

  The present invention relates to a related specification associating system, a related specification associating method, and a related specification associating program for associating related specifications with each other, and in particular, a related specification associating system and related specifications associating specifications included in different texts. The present invention relates to a correlation method and a related specification correlation program.

  In the development of systems and software, specifications within each process and between processes such as requirements, design, and testing are determined.

  Various techniques for managing the influence of changing part of software or changing specifications have been proposed (see, for example, Patent Documents 1 and 2).

  Patent Literature 1 describes a software development system for preventing the influence of a modification from extending over a wide range when a part of the software is modified, and extracting the range in which the modification affects the other. Yes. In the software development system described in Patent Document 1, the first keyword and the second keyword are attached to the document of each process, the document in the own process is stored based on the first keyword, and the second keyword is stored in the second keyword. Based on the document, the document of the higher-order process that is affected by the modification to the document in the own process is stored. In addition, the software development system described in Patent Document 1 stores documents in each process for each process, and defines a relationship between a document in each process and a document in a lower process of the process. Then, when a document in any process is modified, the document in the upper process is searched based on the second keyword assigned to the document.

  Patent Document 2 describes an influence range extraction system that extracts an influence range on a check list and source code caused by a specification change when the specification change occurs in software development. In the system described in Patent Document 2, a data item dictionary in which item names for each program language corresponding to data items are registered is used.

  Further, Patent Document 3 describes a data processing device for making it possible to use a document from which ambiguous description is excluded. In the system described in Patent Document 3, a state definition word used as a definition word having a common meaning in other documents is described in a document. Then, the keyword search unit performs a hierarchical keyword search using the state definition word as a keyword. In Patent Document 3, a link is defined by connecting an upper layer keyword and a lower layer keyword.

  Patent Document 4 describes that data in which character data and dot map image data are mixed is converted into compressed dot map data having a predetermined size or less.

  Japanese Patent Application Laid-Open No. H10-228667 describes inline expansion of a function call instruction of an instruction sequence whose execution count exceeds a threshold value.

JP 2010-39551 (paragraphs 0010, 0011) JP 2007-128123 A (paragraph 0006) JP 2010-108346 A (paragraphs 0011, 0021, 0025, 0031) JP 2006-120145 A (paragraphs 0013, 0033, 0039) JP 2007-18254 (paragraphs 0002, 0003, 0016, 0044)

  By associating development products in and between processes in the development of systems and software, it can be used for detection of reflection omission in the subsequent process of specifications and confirmation of consistency. It can also be used to understand the scope of influence when changing specifications. More specifically, by associating related specifications between texts containing multiple specifications, as described above, detection of reflection reflection to the subsequent process of specifications, consistency check, and specification change It can be used to understand the scope of influence.

  The invention described in Patent Document 1 can prevent the influence of the modification from extending over a wide range when a part of the software is modified, and can extract the range in which the modification affects the other. However, the task of attaching keywords to the document had to be performed in advance. In addition, it is necessary to define a relationship between a document in each process and a document in a lower process of the process. Since these operations are performed manually, these operations take time.

  Even in the invention described in Patent Document 2, it is possible to extract a range of influence on a source code or the like caused by a specification change when the specification change occurs in software development. However, it has been necessary to prepare a data item dictionary in which item names for each programming language corresponding to the data items are registered. Since this work is also done manually, this work took time.

  In the invention described in Patent Document 3, it is necessary to define a state definition word in a document. However, since this operation is also performed manually, this operation takes time.

  To provide a related specification associating system, a related specification associating method, and a related specification associating program capable of associating related specifications between texts including a plurality of specifications with a minimum of manual work. With the goal.

Related specification associating system according to the present invention selects two text or et two specifications set, based on common string that appears in common to the two specifications, the presence or absence of association of the two specifications and relevant presence determining means for determining, if it is determined to be relevant to the two specifications, the two specifications of each lossless compressed size, lossless compression of concatenated string of text of two specifications based on the size, the relevance calculating means for calculating a relevance value, characterized in that it comprises a determining specifications correlating means whether associating two specifications.

The related specifications associating method according to the present invention selects two sets of text or et two specifications, based on common string that appears in common to the two specifications, the relevance of the two specifications determines the presence of, if it is determined to be relevant to the two specifications, based on each of the lossless compression size two specifications, a lossless compression size of the string of string concatenation of two specifications Te, it calculates a relevance value, and wherein the determining whether associating two specifications.

The related specifications correspondence program according to the present invention, the computer selects two sets of text or et two specifications, based on common string that appears in common to the two specifications, two specifications that character relevance relevance existence determining process determines the presence or absence of, if it is determined that there is a relationship between the two specifications, the two specifications of each lossless compressed size, the concatenation of the strings of the two specifications based on the lossless compression column size, relevance calculating process of calculating a relevance value, and, characterized in that to execute a determining specifications associating process whether associating two specifications.

  According to the present invention, manual specifications can be reduced as much as possible, and related specifications can be associated with each other between texts including a plurality of specifications.

It is a block diagram which shows the example of the related specification matching system of this invention. It is explanatory drawing which shows the example of the specification described in the text. It is explanatory drawing which shows the example of the specification described by the figure. It is explanatory drawing which shows the example of the initial specification which an initial specification memory | storage part memorize | stores. It is explanatory drawing which shows the example of the information which a conversion word memory | storage part memorize | stores. It is explanatory drawing which shows the example of the conversion specification which a conversion specification memory | storage part memorize | stores. It is explanatory drawing which shows the example of the link memorize | stored in the link memory | storage part. It is a flowchart which shows the example of the process progress of this embodiment. It is explanatory drawing which shows the example of the link production | generation result by a link production | generation part typically. It is a flowchart which shows the detail of step A2, A3. It is a flowchart which shows the example of the complementation operation | movement by a link complementation part. It is explanatory drawing which shows the example of the link memorize | stored in the link memory | storage part. It is explanatory drawing which shows the example of the link after a complementation typically. It is explanatory drawing which shows the example of the link memorize | stored in the link memory | storage part at the time of the completion | finish of a process. It is explanatory drawing which shows the example of the minimum structure of the related specification matching system of this invention.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
The related specification associating system of the present invention associates related specifications between texts including a plurality of specifications. This text may be in the same document or in a different document. In the following description, a case in which related specifications are associated with each other between documents including a plurality of specifications will be described as an example, and a document ID is used as an ID for identifying text. When matching between different texts in the same document, a text ID for identifying each text may be used. The specifications include, for example, requirements required for the development target, specifications regarding design elements, specifications regarding test items, and the like, but specifications regarding other matters may also be used.

  Hereinafter, association between related specifications between texts including a plurality of specifications may be referred to as a traceability link (or simply a link). In addition, performing such association may be described as “generating a traceability link (or simply link)”.

  FIG. 1 is a block diagram showing an example of a related specification association system of the present invention. The related specification associating system of the present invention includes a storage unit 11 that stores information and a calculation unit 12 that operates under program control.

  The storage unit 11 includes an initial specification storage unit 100, a conversion phrase storage unit 101, a conversion specification storage unit 102, and a link storage unit 103.

  The initial specification storage unit 100 stores in advance text indicating the content of the specification as an initial specification.

  In general, specifications are described in text or in a diagram such as a business flow diagram. FIG. 2 is an explanatory diagram showing an example of specifications described in text. In the example shown in FIG. 2, one specification is shown in one line. FIG. 3 is an explanatory diagram illustrating an example of specifications described in the figure. In FIG. 3, the specifications described in the business flow diagram are shown, and one specification is represented by one box or a plurality of boxes and arrows between the boxes.

  Since the initial specification storage unit 100 stores text representing the contents of the specification, the specification illustrated in FIG. 2 need not be rewritten into the text format and may be stored in the initial specification storage unit 100. On the other hand, the specification illustrated in FIG. 3 may be stored in the initial specification storage unit 100 after being rewritten into a text format.

  FIG. 4 is an explanatory diagram illustrating an example of initial specifications stored in the initial specification storage unit 100. Each specification is assigned a text ID (document ID in this example) that identifies the text (in this example, a document) to which the specification belongs. Also, each specification is assigned a number that is a specification ID in the order in which it appears in the document. It should be noted that in the diagram showing the specifications (for example, see FIG. 3), the individual specifications are generally described in order from the upper left to the lower right. Therefore, when individual specifications shown in such a diagram are converted to text format, the specification shown in the upper left in the original diagram is assigned a smaller number as the specification ID, and is displayed in the lower right in the original diagram. A larger number may be assigned as a specification ID for a specification that has been made.

  Hereinafter, a specification whose specification ID is x is simply referred to as a specification x.

  In FIG. 4, the initial specification of a document whose document ID is 0 (hereinafter referred to as document 0) is described as a specification by extracting a character string indicating an actor and specification contents from the business flow diagram shown in FIG. . Specifically, the actor in the workflow diagram and the character string representing the specification content in the workflow diagram are described by being separated by a predetermined symbol (“:” in this example). The branch condition is described by “→”.

  In FIG. 4, the initial specification of a document whose document ID is 1 (hereinafter referred to as document 1) is the same as the specification shown in FIG.

  In the following, a case where a link is generated between the specification originally described in the drawing (specification of document 0) and the specification expressed in the text format (specification of document 1) will be described as an example. A link may be generated between the specified specifications, or a link may be generated between the specifications expressed in the text format.

  The conversion phrase storage unit 101 stores in advance a conversion phrase for converting a character string described as an initial specification (see FIG. 4). Specifically, the converted phrase storage unit 101 stores the phrase before conversion and the converted phrase in association with each other. For example, a word before conversion and a word after conversion are associated with each other so that synonymous words described in the initial specification are unified with the same word. Note that the phrase also includes punctuation marks and symbols used in the initial specification (in this example, “:” and “→”). If the word after conversion is not defined, it means that the word before conversion is deleted.

  FIG. 5 shows an example of information stored in the conversion word storage unit 101. In FIG. 5, information for one line represents a correspondence relationship between one or more pre-conversion words and one post-conversion word / phrase. In the example shown in FIG. 5, “ordering” is converted to “ordering” and the like. In the example shown in FIG. 5, the symbol “,” is used to delimit a plurality of pre-conversion words that are converted to the same word. For example, each of the words “do”, “be done”, “do”, and “to be” written with “,” as a word before conversion is a punctuation mark “.”. "Is converted to" ".

  It can be said that the information stored in the conversion phrase storage unit 101 is a conversion rule for converting a phrase in the initial specification.

  A combination of words before and after conversion may be added using a general-purpose synonym dictionary. For example, in WordNet, which is a general-purpose synonym dictionary, terms such as “order” and “ordering” are synonymous sets. Therefore, any of the terms belonging to the synonym set is used as a post-conversion term, and the synonym set is included in the synonym set. What is necessary is just to define the other words to which it belongs as a word before conversion.

  In the example shown in FIG. 5, the converted word / phrase corresponding to “Here” is blank and is not defined. This means that the phrase “here” in the initial specification is deleted. Thus, when deleting a phrase, it is sufficient that the phrase is defined as a phrase before conversion and the phrase after conversion is not defined.

  The conversion phrase storage unit 101 uses a general-purpose synonym dictionary regardless of what initial specification is used, stores the correspondence between words before and after conversion, and before and after conversion for simplification of expression. What is necessary is just to memorize | store the correspondence of a phrase. Then, synonyms specific to the industry / business such as a business glossary may be added. In this way, if the phrases to be stored in the conversion phrase storage unit 101 are determined, the storage contents of the conversion phrase storage unit 101 can be prepared without manpower and time.

  The conversion specification storage unit 102 stores a result of conversion of the initial specification stored in the initial specification storage unit 100 using the conversion rule stored in the conversion word storage unit 101 (hereinafter referred to as conversion specification). FIG. 6 is an explanatory diagram illustrating an example of conversion specifications stored in the conversion specification storage unit 102. FIG. 6 illustrates a conversion specification obtained by converting the initial specification shown in FIG. 4 according to the conversion rule shown in FIG.

  The link storage unit 103 stores a link between specifications. That is, the link storage unit 103 stores a correspondence relationship between related specifications between texts (in this example, between documents) including a plurality of specifications. FIG. 7 is an explanatory diagram illustrating an example of a link stored in the link storage unit 103. In FIG. 7, for example, between “specification 0” of “document 0” and “specification 0” of “document 1”, “specification 2” of “document 0”, and “specification 3” of “document 1”. It is shown that a link is generated between the two. In the example shown in FIG. 7, the link reliability is stored for each link. The link reliability represents the reliability related to the appropriateness of the link (association between specifications). For example, when a link is generated in the link generation unit 107, information indicating that the link reliability is high (“◯” in FIG. 7) is stored, and the link complementing unit 108 estimates the link candidate The link generation unit 107 may store information (for example, “Δ”, not shown in FIG. 7) indicating that the link reliability is lower than that of the link generated.

  The calculation unit 12 includes a phrase conversion unit 104, an association presence / absence determination unit 105, an association calculation unit 106, a link generation unit 107, and a link complementation unit 108.

  The phrase conversion unit 104 converts the initial specification stored in the initial specification storage unit 100 according to the conversion rule stored in the conversion word storage unit 101. That is, if there is a character string corresponding to the word before conversion stored in the conversion word storage unit 101 in the initial specification, the word conversion unit 104 converts the character string into the conversion corresponding to the word before conversion. Convert to a later phrase. The phrase conversion unit 104 causes the conversion specification storage unit 102 to store the result of converting the initial specification (conversion specification). When the phrase conversion unit 104 converts the initial specification shown in FIG. 4 according to the conversion rule shown in FIG. 5, the conversion specification shown in FIG. 6 is obtained.

  The relevance determination unit 105 extracts a common partial sequence for each of two conversion specifications having different document IDs from the conversion specification stored in the conversion specification storage unit 102, and based on the common partial sequence, The presence / absence of relevance between the two conversion specifications is determined. Here, the common partial sequence is a character string having a predetermined length, and is a character string that appears in common in the two specifications. Conditions such as not including hiragana in the common subsequence may be defined. This condition is an example, and other conditions may be set. An exception may be specified together with a condition. For example, in addition to the condition that hiragana characters are not included in the common subsequence, it may be possible to specify an exception that a specific hiragana such as “none” is extracted as “common subsequence”. Or the conditions regarding a common partial row | line | column do not need to be defined.

  The relevance determination unit 105 may select two conversion specification groups belonging to different documents from the two documents (document 0 and document 1 in FIG. 6) that have been converted according to the conversion rule. Then, the relevance determination unit 105 may extract a common partial sequence for each selected set, and determine whether there is relevance between the two conversion specifications based on the common partial sequence.

  The relevancy calculation unit 106 calculates a value indicating the strength of relevance between specifications (hereinafter referred to as relevance value) for each of two conversion specifications having different document IDs.

  The determination process of the presence / absence of relevance by the relevance presence / absence determination unit 105 and the relevance value calculation process by the relevance calculation unit 106 will be described later.

  The link generation unit 107 generates a link between conversion specifications having different document IDs based on the relevance value between the conversion specifications calculated for the conversion specifications stored in the conversion specification storage unit 102. That is, the link generation unit 107 associates conversion specifications with different document IDs. For example, if the relevance value between two conversion specifications is equal to or greater than a predetermined first threshold, the link generation unit 107 estimates that a link should be generated for the two conversion specifications, and generates a link. . The first threshold may be set, for example, by a user of the related specification association system inputting via an input means (not shown) such as a keyboard. For example, the default value of the first threshold value may be stored in advance in the related specification association system. These are examples, and the first threshold setting method is not limited to these methods. The first threshold value may be set to a value such as 0.6, for example, but this value is also an example, and the first threshold value may be a value other than 0.6.

  The link complementing unit 108 pays attention to the order of the conversion specifications in the document with respect to the conversion specifications for which no link has been generated by the link generation unit 107, and the conversion specifications whose order precedes the order of the conversion specifications. Whether or not to complement the link is determined based on the link generation status and the link generation status of the conversion specifications whose order is later than the order of the conversion specifications. Then, the link complementation unit 108 complements the link between the conversion specifications determined to complement the link. Here, complementing a link means generating a link in a conversion specification for which a link has not yet been generated after the link generation processing by the link generation unit 107. In other words, the conversion specifications that are not associated with other conversion specifications are associated with other conversion specifications.

  The order of specifications in the document is represented by a specification ID. For example, in the conversion specification illustrated in FIG. 6, when focusing on the specification 1 of the document 0, the conversion specification whose order precedes the specification 1 is the specification 0 of the document 0 and the order is higher than the specification 1. The conversion specifications that follow are specification 2, specification 3, and specification 4 of document 0. Note that the specification order is not determined between specifications with different document IDs. Therefore, when attention is focused on the specification 1 of the document 0, the link complementation unit 108 determines whether the document 0 is based on the link generation status of the specification 0 of the document 0, the link generation status of the specification 2, the specification 3, and the specification 4 of the document 0. It is determined whether or not the link is complemented between the specification 1 of 0 and the specification of the document 1.

Next, the operation will be described.
FIG. 8 is a flowchart illustrating an example of processing progress of the present embodiment. First, the phrase conversion unit 104 reads the initial specifications stored in the initial specification storage unit 100 one by one, and converts the read initial specifications according to the conversion rules stored in the conversion word storage unit 101. That is, if there is a character string corresponding to the word before conversion stored in the conversion word storage unit 101 in the read initial specification, the character string is converted into a word after conversion corresponding to the word before conversion. To do. For example, if there is a character string “Purchase” in the read initial specification, the character string is converted to “Order”, or if there is a character string “Accept” in the read initial specification, the character string is Or convert it to “acceptance”. The phrase conversion unit 104 stores the conversion specification obtained by this conversion in the conversion specification storage unit 102. The phrase conversion unit 104 performs this process on all initial specifications stored in the initial specification storage unit 100 (step A1).

  Next, the relevance presence / absence determination unit 105 determines the presence / absence of relevance between the two conversion specifications for each of the two conversion specifications having different document IDs with respect to the conversion specification stored in the conversion specification storage unit 102. (Step A2). Next, the relevancy calculation unit 106 calculates a relevance value for each of two conversion specifications having different document IDs (step A3). Details of the processes of steps A2 and A3 will be described later.

  Next, the link generation unit 107 determines whether or not all the relevance values calculated for each of two conversion specifications having different document IDs are compared with the first threshold (step A5 described later). Determine (step A4). If the comparison with the first threshold is not completed for all relevance values (No in step A4), the link generation unit 107 calculates each relevance value calculated for each of two conversion specifications having different document IDs. Among them, one unselected relevance value is selected, and it is determined whether or not the relevance value is equal to or greater than a first threshold value (step A5). If the selected relevance value is greater than or equal to the first threshold value (Yes in step A5), the link generation unit 107 generates a link between the two conversion specifications from which the relevance value is calculated, The link information is stored in the link storage unit 103 (step A6). In step A6, the link generation unit 107 stores “◯” as the link reliability of the generated link. On the other hand, if the selected relevance value is less than the first threshold value (No in step A5), the link generation unit 107 repeats the processing after step A4.

  FIG. 9 is an explanatory diagram schematically illustrating an example of a link generation result (specification matching result) by the link generation unit 107. Since the specification 0 of the document 0 and the specification 0 of the document 1 have almost the same character strings and are highly related, the link generation unit 107 has the specifications 0 of the document 0 and the specifications of the document 1 as shown in FIG. Create a link with 0. In addition, regarding the specification 1 of the document 1, since the specification including the highly relevant character string does not exist in the document 0, the link generation unit 107 generates a link related to the specification 1 of the document 1 as shown in FIG. do not do.

  When the link is generated as illustrated in FIG. 9, the link generation unit 107 stores the information illustrated in FIG. 7 in the link storage unit 103. For example, the link generation unit 107 has generated a link between the specification 0 of the document 0 and the specification 0 of the document 1, so that (the specification 0 of the document 0 and the specification 0 of the document 1 are associated) Is stored in the link storage unit 103 (see FIG. 7).

  After step A6, the link generation unit 107 repeats the processing after step A4. If it is determined in step A4 that the comparison with the first threshold has been completed for all relevance values (Yes in step A4), the link complementing unit 108 generates a link among the conversion specifications in the document 0. It is determined whether or not a link specification is to be complemented with a conversion specification in document 1 for a conversion specification that has not been completed, and the link is supplemented between the conversion specifications that have been determined to be supplemented (step A7). Similarly, the link complementation unit 108 determines whether or not to complement a link with a conversion specification in the document 0 for a conversion specification in which no link is generated among the conversion specifications in the document 1. The link is complemented between the conversion specifications determined to complement the link (step A8).

  In step A8, the same processing as in step A7 may be executed by switching the document ID of interest.

  Next, the process of step A2 by the relevance presence / absence determination unit 105 and step AS3 by the relevance calculation unit 106 will be described. FIG. 10 is a flowchart showing details of steps A2 and A3. In the description of FIG. 10, two conversion specifications having different document IDs are referred to as specifications A and B. In FIG. 10, steps A11 and A12 correspond to step A2 shown in FIG. 8, and processing subsequent to step A12 corresponds to step A3 shown in FIG.

  The relevance determination unit 105 extracts a common partial sequence from the specifications A and B (step A11). That is, from the character string representing the specification A and the character string representing the specification B, a character string that appears in common in the specifications A and B and has a predetermined length is extracted.

  Here, a case where the length of the common subsequence is set to “2” in advance will be described as an example. An example in which a common partial sequence is extracted from specification 0 of document 0 and specification 0 of document 1 shown in FIG. 6 will be described. In this case, a character string that appears in common in specification 0 of document 0 and specification 0 of document 1 and has a length of “2” may be extracted as a common partial sequence. In this example, “customer”, “customer”, “gai”, “sales”, “to business”, “to note”, “order”, “sentence table”, “in table”, “in note” , “Order” corresponds to the common partial sequence, and these common partial sequences may be extracted.

  Further, as already described, a condition may be predetermined for the common subsequence. In that case, the relevance determination unit 105 may extract a common partial sequence that satisfies the condition. For example, it is assumed that the length of the partial subsequence is defined as “2” and the condition that hiragana is not included in the common subsequence is defined. At this time, when the common partial sequence is extracted from the specification 0 of the document 0 and the specification 0 of the document 1 shown in FIG. 6, the relevance determination unit 105 displays “customer”, “sales”, “order”, ”And“ order ”are extracted.

  Further, the condition relating to the common partial sequence is not limited to the condition “not including hiragana”, and may be other conditions. For example, a condition that “the same common partial sequence is not extracted twice” may be defined. Suppose that the common partial sequence does not include hiragana and that the common partial sequence is not duplicated. At this time, when the common partial sequence is extracted from the specification 0 of the document 0 and the specification 0 of the document 1 shown in FIG. 6, the relevance determination unit 105 reads “customer”, “sales”, “order”, “sentence table”. "Will be extracted. Under the condition of “Do not include hiragana”, “Order” will be extracted twice, but if you add the condition of “Do not extract the same common subsequence”, “Order” will be Only one will be extracted.

  Further, as already described, an exception may be specified together with a condition. For example, in addition to the condition that hiragana characters are not included in the common subsequence, it may be possible to specify an exception that a specific hiragana such as “none” is extracted as “common subsequence”. The character string “None” illustrated here is a character string necessary for expressing negation in the specification. By specifying an exception, a character string such as “None” can be extracted as a common substring. it can.

  The conditions related to the common partial sequence and the exceptions may be set by the user of the related specification association system by inputting via an input means (not shown) such as a keyboard. Further, for example, conditions and exceptions may be stored in the related specification association system in advance. These are merely examples, and the method of setting conditions and exceptions is not limited to these methods.

  In the following description, a condition is set that the common substring does not include hiragana and that the common substring does not overlap, and an exception is given that “no” is allowed as a common substring. And

  After extracting the common partial sequence of the two specifications A and B in step A11, the relevance determination unit 105 satisfies the condition for determining that the extracted common partial sequence is related to the two specifications. It is determined whether or not (step A12). Hereinafter, this condition is referred to as a relevance condition.

  As an example of the relevance condition, “the number of extracted common subsequences is equal to or greater than a predetermined number” or the like can be cited. In this case, the relevance determination unit 105 determines that the two specifications are related if the number of extracted common partial sequences is equal to or greater than a predetermined number, and the number of extracted common partial sequences is predetermined. If it is less than the number, it is determined that the two specifications are not related.

  Alternatively, a condition that “the ratio of the number of extracted common partial strings to the sum of the lengths of the character strings of the two specifications A and B is equal to or greater than a predetermined ratio” may be used as the relevance condition. . In this case, if the ratio of the number of extracted common subsequences to the sum of the lengths of the character strings of the two specifications A and B is greater than or equal to a predetermined ratio, Is determined to be related. On the other hand, if the ratio of the number of extracted common subsequences to the sum of the lengths of the character strings of the two specifications A and B is less than a predetermined ratio, the relevance determination unit 105 determines that the two specifications It is determined that there is no relevance.

  Or, “When the length of the character string of the specification A is p, the length of the character string of the specification B is q, and the length of the character string of the common partial sequence is r, (the extracted common partial sequence The condition that “number) / {(p−r) × (q−r)} is equal to or greater than a predetermined value” may be used as the relevance condition.

  The relevance condition may be set by the user of the related specification associating system inputting via an input means (not shown) such as a keyboard. For example, the relevance condition may be stored in advance in the related specification association system. These are examples, and the setting method of the relevance condition is not limited to these methods.

  When the relevance condition is satisfied, the relevance presence / absence determination unit 105 determines that there is a relevance between the specifications A and B, and when the relevance condition is not satisfied, the relevance is determined between the specifications A and B. It is determined that there is no relevance.

  When calculating the relationship value between the two specifications A and B in step A3, if it is determined that there is a relationship between the specifications A and B (in step A12, it is determined that the relationship is satisfied) If so, the relevancy calculation unit 106 calculates the lossless compression size of the character string of the specification A (step A13). Here, the lossless compression size is a size when a character string of a specification is compressed using a lossless compression algorithm. The lossless compression algorithm is a compression algorithm in which a character string before compression completely matches a character string after being compressed and further expanded. Examples of lossless compression algorithms include zip and gzip. What type of lossless compression algorithm is used may be specified by the user of the related specification associating system via input means (not shown) such as a keyboard. Further, for example, what type of lossless compression algorithm is used may be stored in advance in the related specification association system. The method for designating which lossless compression algorithm is used is not limited to these, and other methods may be used. In the present embodiment, a case where gzip is used as the lossless compression algorithm is taken as an example.

  After step A13, the relevance calculation unit 106 calculates the lossless compression size of the character string of the specification B (step A14). Further, the relevancy calculation unit 106 calculates the lossless compression size of the character string obtained by concatenating the character string of the specification A and the character string of the specification B (step A15). Hereinafter, this value is referred to as a concatenated lossless compression size of specifications A and B. The relevancy calculation unit 106 uses the same lossless compression algorithm in steps A13, A14, and A15.

  A character string obtained by concatenating two specification character strings is a character string in which one specification character string is connected to the other specification character string. For example, the character string in which the character string of specification 0 of document 0 and the character string of specification 0 of document 1 shown in FIG. 6 are concatenated is “customer orders to sales and order customer orders to sales. "

  After step A15, the relevance calculation unit 106 calculates a relevance value between specifications A and B using each lossless compression size calculated in steps A13, A14, and A15 (step A16). In step A16, the relevance calculation unit 106 may calculate the relevance value between the specifications A and B by the following calculation method, for example. Let C (A) be the lossless compression size of the specification A calculated in step A13. In addition, the reversible compression size of the specification B calculated in step A14 is C (B). Further, the concatenated lossless compression size of the specifications A and B calculated in step A15 is defined as C (AB). The relevancy calculation unit 106 may calculate the relevance value between the specifications A and B, for example, by calculating the following equation (1). The relevance value between specifications A and B is represented by s.

s = 1-C (AB) / {C (A) + C (B)} Formula (1)

  The calculation method of the relationship value between the specifications A and B is not limited to the method using the formula (1). For example, min {α, β} represents the smaller value of α and β. max {α, β} represents the larger value of α and β. At this time, the relevance calculation unit 106 may calculate the relevance value s between the specifications A and B by performing the calculation of Expression (2) shown below.

s =
1- [C (AB) -min {C (A), C (B)}] / max {C (A), C (B)} Equation (2)

  The relevance value is generally a numerical value from 0 to 1. The closer the relevance value is to 1, the more the specification A and the specification B are related.

  Here, the higher the redundancy, the higher the compression rate and the smaller the compression size. Therefore, if the specifications A and B contain many identical character strings and have high redundancy, the concatenated lossless compression size C (A, B) becomes small and the relevance value is calculated as a large value. Hereinafter, in the present embodiment, a case where the relevance value between the specifications A and B is calculated using Expression (2) will be described as an example.

  The storage unit 11 may include a relevance value storage unit (not shown) that stores a relevance value for each combination of two specifications. Then, the relevancy calculation unit 106 may store the relevance value storage unit in the relevance value storage unit by combining the calculated relevance value and the two specifications from which the calculation is made. . Alternatively, the relevance calculation unit 106 may store the calculated relevance value in the link storage unit 103. At this time, the relevancy calculation unit 106 may represent a combination of different document specifications based on the document ID and the specification ID, and may store a relevance value (not shown in FIG. 7) regarding the combination of the specifications. . In Step A17 described below, since it is determined that there is no relationship between the specifications, for example, “x” may be stored as the link reliability for the combination of the two specifications. “X” means that there is no relationship between specifications.

  In step A3, when calculating the relevance value between the two specifications A and B, the relevance calculating unit 106 determines that there is no relevance between the specifications A and B (step A12). If it is determined that the relevance condition is not satisfied), the relevance value between the specifications A and B is set to 0 (step A17).

  When it is determined that the relevance condition is not satisfied for the specifications A and B, the relevance value between the specifications A and B is set to 0. In this case, when the relevance value is calculated using the compression size, This is because the following problems arise. In the lossless compression algorithm, a character string representing a specification is converted into a binary symbol string and then compressed. At this time, for example, if the character “A” is converted to “1010010010100010” and the character “I” is converted to “1010010010100100”, “A” and “I” are completely different characters in the original character string. In spite of this, since the first 13 characters in the binary symbol string are the same, when calculating the relevance value using the compressed size, “a” and “i” are related to some extent. (That is, a relevance value that is somewhat high is calculated). As described above, when the compression size is used for calculating the relevance value between character strings that are determined to have no relevance, there is a problem that the relevance value may be calculated to be high. In order to prevent such a problem, when it is determined that the relevance condition is not satisfied for the specifications A and B, the relationship value of the specifications A and B is set to 0 without using the compression size.

  Moreover, even if the character string of two specifications includes the same character string, the same character string is a character string consisting of characters included in many texts (documents) such as hiragana. Therefore, it is not preferable to make the relevance value of the two specifications high. On the other hand, if the relevance value is simply calculated except for hiragana, for example, “customer to sales” and “customer to sales” cannot be distinguished.

  In the present embodiment, the relevance presence / absence determination unit 105 determines the relevance of the two specifications based on the common partial sequence in step A12. When it is determined that there are few common partial sequences and the two specifications are not related, the relationship calculation unit 106 sets the relationship value of the two specifications to 0 without using the compressed size. . Further, the relevance determination unit 105 performs steps A11 and A12, so that it is determined that a combination such as “customer to sales” and “customer to sales” is also relevant even if hiragana is excluded. be able to. And the relevance calculation part 106 can calculate the relevance value including hiragana by performing the processing of steps A13 to A16. In the present embodiment, the occurrence of the above problems can be prevented by the processing of steps A2 and A3 (specifically, steps A11 to A17).

  Next, the complementing operation by the link complementing unit 108 will be described. FIG. 11 is a flowchart illustrating an example of a complement operation performed by the link complement unit 108. Here, a case where the document 0 illustrated in FIG. 9 is complemented will be described as an example. More specifically, an example will be described in which the process proceeds to step A7 (see FIG. 8) after the link illustrated in FIG. 9 is generated.

In the following description, the specification ID of the document of interest (document 0 in this example) is represented by the symbol “i”, and the specification ID of the other document (document 1 in this example) is represented by the symbol “j”. Represent. Further, the number of specifications belonging to the document of interest is N _0, and the number of specifications belonging to the other document is N ₁ . Furthermore, it is assumed that a constant k which is a positive integer is predetermined. The constant k may be set, for example, by a user of the related specification associating system inputting via an input unit (not shown) such as a keyboard. Further, for example, the constant k may be stored in advance in the related specification association system. These are examples, and the method of determining the constant k is not limited to these methods. k is used to determine a range of specifications before the specification and a range of specifications after the specification with a certain specification as a reference.

  First, the link complementation unit 108 initializes a number i representing the specification ID of the document of interest (document 0 in this example) to 0 (step A31).

Next, the link complementation unit 108, i is equal to or less than _{N 0} is the number of specifications that belong to the document 0 (step A32). If i <N ₀ (Yes in step A32), the process proceeds to step A33, and if i ≧ N ₀ (No in step A32), the process ends.

  In step A33, the link complementing unit 108 initializes a number j representing the specification ID of the document 1 to 0 (step A33).

Next, the link complementation unit 108, j is determines whether the _N less than ₁ is the number of specifications that belong to the document 1 (step A34). If j _{<N 1} (Yes at Step A34), the process proceeds to step A35, if j ≧ _{N 1} (No in step A34), the process proceeds to step A39.

  In step A35, the link complementation unit 108 refers to the information stored in the link storage unit 103, and determines whether or not a link has already been generated in the specification i of the document 0 (step A35). In other words, the link complementation unit 108 determines whether any specification of the document 1 is associated with the specification i of the document 0. If a link has already been generated in the specification i of the document 0 (Yes in step A35), the process proceeds to step A39. If a link has not yet been generated in the specification i of document 0 (No in step A35), the process proceeds to step A36.

  In step A36, the link complementing unit 108 refers to the relevance value between the specification i of the document 0 and the specification j of the document 1, and determines whether or not the relevance value is equal to or greater than a second threshold value. (Step A36).

  As already described, the storage unit 11 includes a relevance value storage unit (not shown), and the relevance storage unit 106 stores the relevance value between specifications obtained in step A3 in the relevance value storage unit. You can let it go. Alternatively, the relationship storage unit 106 may store the relationship value between the specifications obtained in step A3 in the link storage unit 103. In step A36, the link complementing unit 108 refers to the relevance value of the specifications i and j among the relevance values obtained and stored in step A3, and whether the relevance value is greater than or equal to the second threshold value. What is necessary is just to determine.

  Note that the second threshold value may be set, for example, by a user of the related specification associating system inputting via an input means (not shown) such as a keyboard. For example, the default value of the second threshold value may be stored in advance in the related specification association system. These are examples, and the second threshold setting method is not limited to these methods. The second threshold value may be set to a value such as 0.25, for example, but this value is also an example, and the second threshold value may be a value other than 0.25. However, the second threshold value is set to a value smaller than the first threshold value.

  If the relevance value of the specifications i and j is less than the second threshold value (No in step A36), the process proceeds to step A40. If the relevance value of the specifications i and j is greater than or equal to the second threshold value (Yes in step A36), the link complementation unit 108 determines whether the precondition is satisfied (step A37).

  The precondition is that i = 0, or in the document 0, between the specification i-k before the specification i and the specification i-1 immediately before the specification i, the j-th in the document 1 That is, there is at least one specification associated with the previous specification. That is, “i = 0” and “jth in the document 1 from the specification i−k before the specification i to the specification i−1 immediately before the specification i in the document 0” If at least one of the two of “there is at least one specification associated with the previous specification” is satisfied, the precondition is satisfied. “I = 0” and “in the document 0, between the specification i−k k before the specification i and the specification i−1 immediately before the specification i, If none of “there is at least one specification associated with the specification” is satisfied, the precondition is not satisfied. Note that i = 0 means that the specification i is the first specification in the document 0.

  If the precondition is not satisfied (No in step A37), the process proceeds to step A40. If the precondition is satisfied (Yes in step A37), the link complementation unit 108 determines whether the postcondition is satisfied (step A38).

The post-condition is i = N ₀ −1, or in the document 0, the specification i + 1 after the specification i and the specification i + k after the specification i to the specification i + k after the specification i. That is, there is at least one specification associated with the specification. That is, “i = N ₀ −1” and “from the specification i + 1 next to the specification i in the document 0 to the specification i + k after the specification i, the jth and later in the document 1 If at least one of two of “there is at least one specification associated with the specification” is satisfied, the post-condition is satisfied. “I = N ₀ −1” and “from the specification i + 1 next to the specification i in the document 0 to the specification i + k after the specification i, the jth and subsequent specifications in the document 1 If none of the “corresponding specifications are present” is satisfied, the post-condition is not satisfied. Note that i = N ₀ −1 means that the specification i is the last specification in the document 0.

  If the post-condition is not satisfied (No in step A38), the process proceeds to step A40. If the post-condition is satisfied (Yes in step A38), the link complementation unit 108 supplements the specification j of the document 1 with the specification i of the document 0 (step A39). In step A39, the link complementation unit 108 may store the link between the specification i of the document 0 and the specification j of the document 1 in the link storage unit 103 as illustrated in FIG. However, the link reliability is stored as “Δ”.

  After step A39, the link complementation unit 108 increments the value of j by 1 (step A40). Also, when it is determined No in Steps A36, A37, and A38, the value of j is incremented by 1 (Step A40). After step A40, the processing after step A34 is repeated.

Further, in step A34, if it was j ≧ _{N 1} (No in step A34) and, in step A35, if already linked to specifications i of the document 0 is generated (Yes in step A35), the link complementary portion 108 increments the value of i by 1 (step A39). After step A39, the processes after step A32 are repeated.

  In general, in the development of a system, software, etc., a document for a subsequent process is created based on a document for the previous process. Therefore, the description order of specifications in a document is often the same. In addition, when the specifications are written in the order of work, the description order of the specifications is the same in each document. Therefore, when both the precondition and the postcondition are satisfied as described above, the combination of the specifications i and j is considered as a candidate for link generation. Therefore, it can be said that complementing the link between the specifications i and j (performing step A39) is appropriate. However, the association between specifications performed by the link complementation unit 108 is assumed to have a lower link reliability (reliability related to the appropriateness of association) than the association between specifications performed by the link generation unit 107. Determine.

  Here, the case where the process has been shifted to step A7 (see FIG. 8) has been described as an example. However, in step A8, the document to be focused on is document 1 and the other document is document 0, and the same processing may be performed. .

Hereinafter, a specific example of the operation when the process proceeds to step A7 (see FIG. 8) will be described. At the time of shifting to step A7, it is assumed that the specifications between the documents 0 and 1 are associated as illustrated in FIG. 9, and the link storage unit 103 stores information illustrated in FIG. Here, since the link of document 0 is complemented, N ₀ = 5 and N ₁ = 7. The second threshold is assumed to be 0.25.

The link complementing unit 108 initializes i representing the specification ID of the document 0 to 0 (step A31), and initializes j representing the specification ID of the document 1 to 0 because i <N ₀ (step A33). . And, since it is j _{<N 1,} the process proceeds to step A35. At this time, i = 0 and j = 0.

  When it is confirmed in step A35 whether or not a link is generated in the specification 0 of the document 0, a link is generated in the specification 0 of the document 0 with the specification 0 of the document 1 (FIG. 9, FIG. 9). 7). Accordingly, the process proceeds to step A39, where the link complementing unit 108 increments i and performs the processes of steps A32, A33, A34. As a result, i = 1 and j = 0.

  Next, the link complementation part 108 confirms whether the link is produced | generated by the specification 1 of the document 0 in step A35. Since no link is generated in the specification 1 of the document 0 (see FIGS. 9 and 7), the process proceeds to step A36. Here, the common subsequence between the specification 1 of the document 0 and the specification 0 of the document 1 is only “sales”, and it is determined that there is no relationship between the two specifications, and the relevance value is set to 0. Suppose that Then, since the relevance value between the specification 1 of the document 0 and the specification 0 of the document 1 is less than the second threshold (No in Step A36), the link complementation unit 108 increments j (Step A40). . As a result, i = 1 and j = 1. As already described, the relevance values between the specifications may be stored in a relevance value storage unit (not shown) in the storage unit 11 or stored in the link storage unit 103. May be. In step A36, the relevance value stored in the relevance value storage unit or the link storage unit 103 may be referred to.

  Subsequently, determinations in steps A34 and A35 are made, and the process proceeds to step A36. Here, regarding the specification 1 of the document 0 and the specification 1 of the document 1, since there is no common partial sequence, it is determined that there is no relationship between the two specifications, and the relationship value is set to 0. . Then, since the relevance value between the specification 1 of the document 0 and the specification 1 of the document 1 is less than the second threshold (No in Step A36), the link complementation unit 108 increments j (Step A40). . As a result, i = 1 and j = 2.

  Subsequently, determinations in steps A34 and A35 are made, and the process proceeds to step A36. Here, regarding the specification 1 of the document 0 and the specification 2 of the document 1, it is determined that there are two common subsequences “inventory” and “confirmation”, and the two specifications are related to each other. It is assumed that the relevancy calculation unit 106 has calculated the relevance value as 0.55. Since the relevance value 0.55 is equal to or greater than the second threshold value (0.25), the link complementation unit 108 determines whether or not the precondition is satisfied (step A37).

  Here, a link is generated between the specification 0 immediately before the specification i (i = 1) of the document 0 and the specification 0 of the document 1 (see FIGS. 9 and 7). The specification 0 of the document 1 is a specification before the specification j (j = 2) in the document 1. Accordingly, the link complementation unit 108 determines that the precondition is satisfied (Yes in Step A37), and then determines whether the postcondition is satisfied (Step A38).

  Here, a link is generated between the specification 2 immediately after the specification i (i = 1) of the document 0 and the specification 3 of the document 1 (see FIGS. 9 and 7). The specification 3 of the document 1 is a specification after the specification j (j = 2) in the document 1. Accordingly, the link complementation unit 108 determines that the post-condition is satisfied (Yes in Step A38), and proceeds to Step A39.

  In step A39, the link complementation unit 108 associates the specification i (i = 1) of the document 0 with the specification j (j = 2) of the document 1, and stores the correspondence (link) in the link storage unit 103. Let At this time, “Δ” is stored in the link storage unit 103 as the link reliability of the link. As a result, the information stored in the link storage unit 103 is as shown in FIG. In the information shown in FIG. 12, compared with the information shown in FIG. 7, the correspondence (link) between the specification 1 of the document 0 and the specification 2 of the document 1 is added, and the link reliability “Δ” of the link is also added. ing.

Next, the link complementation part 108 increments j (step A40), and repeats the process after step A34. If i is incremented (step A39), the processing after step A32 is repeated. Then, if i ≧ N ₀ (No in step A32), link complementation for the specification of document 0 (step A7, see FIG. 8) is terminated.

A specific example of the operation when the process proceeds to step A8 (see FIG. 8) after the above operation will be described. In step A8, document 0 and document 1 are replaced and the same processing as in step A7 may be performed. That is, the process shown in FIG. 11 may be performed with the specification ID of document 1 as i and the specification ID of document 0 as j. At this time, N ₀ = 7 and N ₁ = 5.

The link complementing unit 108 initializes i representing the specification ID of the document 1 to 0 (step A31), and since i <N ₀ , j representing the specification ID of the document 0 is initialized to 0 (step A33). . And, since it is j _{<N 1,} the process proceeds to step A35. At this time, i = 0 and j = 0.

  In step A 35, when it is confirmed whether or not a link is generated in the specification 0 of the document 1, a link is generated in the specification 0 of the document 1 and the specification 0 of the document 0. Accordingly, the process proceeds to step A39, where the link complementing unit 108 increments i and performs the processes of steps A32, A33, A34. As a result, i = 1 and j = 0.

  Next, the link complementation part 108 confirms whether the link is produced | generated by the specification 1 of the document 1 in step A35. Since no link is generated in the specification 1 of the document 1, the process proceeds to step A36. Here, regarding the specification 1 of the document 1 and the specification 0 of the document 0, it is determined that there are two common subsequences “order” and “sentence table”, and the two specifications are related to each other. Assume that the relevance calculation unit 106 calculates the relevance value as 0.29. Since the relevance value 0.25 is greater than or equal to the second threshold value (0.25), the link complementation unit 108 determines whether or not the precondition is satisfied (step A37).

  Here, a link is generated between the specification 0 immediately before the specification i (i = 1) of the document 1 and the specification 0 of the document 0. The specification 0 of the document 0 is a specification before the specification j (j = 0) in the document 0. Accordingly, the link complementation unit 108 determines that the precondition is satisfied (Yes in Step A37), and then determines whether the postcondition is satisfied (Step A38).

  Here, a link is generated between the specification 2 immediately after the specification i (i = 1) of the document 1 and the specification 1 of the document 0. The specification 1 of the document 0 is a specification after the specification j (j = 0) in the document 0 (see FIG. 12). Accordingly, the link complementation unit 108 determines that the post-condition is satisfied (Yes in Step A38), and proceeds to Step A39.

  In step A <b> 39, the link complementation unit 108 associates the specification i (i = 1) of the document 1 with the specification j (j = 0) of the document 0 and stores the correspondence (link) in the link storage unit 103. Let At this time, “Δ” is stored in the link storage unit 103 as the link reliability of the link.

Next, the link complementation part 108 increments j (step A40), and repeats the process after step A34. If i is incremented (step A39), the processing after step A32 is repeated. If i ≧ N ₀ (No in step A32), the link complement to the specification of document 1 (step A8 shown in FIG. 8) is terminated.

  As described above, the links between the specifications in document 0 and document 1 when the processing up to step A8 is completed are as shown in FIG. In FIG. 13, a link indicated by a solid line represents a link generated by the link generation unit 107 before step A7. A link indicated by a broken line represents a link supplemented by the link complementation unit 108 in steps A7 and A8.

  Further, as described above, when the processing up to step A8 is completed, the information shown in FIG. The correspondence between specifications shown in FIG. 14 corresponds to the correspondence between specifications schematically shown in FIG.

  In addition, the related specification associating system of the present invention may include an output device that outputs a correspondence relationship between specifications stored in the link storage unit 103. The output device is, for example, a display device or a printing device, but may be another output device. And when outputting the correspondence of the specifications memorize | stored in the link memory | storage part 103 from an output device, the link produced | generated by the link production | generation part 107 (namely, link whose link reliability is "(circle)"). And a link supplemented by the link complementing unit 108 (that is, a link having a link reliability of “Δ”) may be distinguished from each other. For example, as illustrated in FIG. 13, the links generated by the link generation unit 107 may be represented by solid lines, and the links supplemented by the link complementation unit 108 may be represented by broken lines.

  In the above description, specification association between documents has been described. However, the present invention can also be applied to specification association between texts in the same document.

  In the present invention, a link (traceability link) can be generated between texts including a plurality of specifications without manually specifying the relationship between specifications. For example, a link can be generated without manually performing an operation for determining a phrase related to the relationship between specifications. Therefore, it is possible to associate related specifications with each other between texts including a plurality of specifications by minimizing manual work.

  In addition, because the relevance value is not equal to or greater than the first threshold, the link can be supplemented according to the generation status of the links before and after the specification even if the link is not generated by the link generation unit 107. . That is, if a pre-condition and a post-condition are satisfied for a certain specification, the link can be supplemented by assuming that it is appropriate to associate with the specification of another text even if the relevance value is low.

  The phrase conversion unit 104, the relevance determination unit 105, the relevance calculation unit 106, the link generation unit 107, and the link complementation unit 108 are realized by a computer (calculation unit 12) that operates according to a related specification association program. That is, the computer may read the related specification association program and operate as the phrase conversion unit 104, the relevance presence / absence determination unit 105, the relevance calculation unit 106, the link generation unit 107, and the link complementation unit 108 according to the program.

  This related specification association program is a floppy disk (registered trademark), a magnetic disk such as a hard disk, an optical disk such as a CD-ROM or DVD (Digital Versatile Disk), an MO (Magneto-Optical Disk), or a semiconductor memory. It may be stored in such a storage medium and distributed. These storage media are examples, and the medium for storing the related specification association program may be any storage medium that can be read by the computer.

  Further, a part of each process may be executed by middleware such as an operating system running on the computer, database management software, network software, or the like according to a program installed in the computer from the storage medium.

  Further, the computer may download the related specification association program via the LAN or the Internet, store the program or temporarily store the program, and operate according to the program.

  Further, the related specification association program may be stored separately in a plurality of storage media, and the computer may read the related specification association program from the plurality of storage media.

  The computer may be a personal computer or a system in which a plurality of devices are connected via a network.

  Further, the computer is not limited to a personal computer, and may be a device or device that includes an arithmetic processing device included in an information processing device and can realize the functions of the present invention by a program.

  Further, the phrase conversion unit 104, the relevance presence / absence determination unit 105, the relevance calculation unit 106, the link generation unit 107, and the link complementation unit 108 may be realized by different hardware.

  Next, the minimum configuration of the present invention will be described. FIG. 15 is an explanatory diagram showing an example of the minimum configuration of the related specification association system of the present invention. The related specification associating system of the present invention includes relevance presence / absence determining means 71, relevance calculating means 72, and specification associating means 73.

  Relevance presence / absence determination means 71 (for example, relevance presence / absence determination unit 105) is a set of two specifications belonging to different texts from two texts (for example, document 0 and document 1) including a plurality of ordered specifications. And for each selected set of specifications, extract a common character string (for example, a common substring) that is a character string of a predetermined length that appears in common in the two specifications, and extract the common character string Based on the above, it is determined whether or not the two specifications are related.

  When it is determined that the two specifications are not related for each set of two specifications belonging to different texts, the relevance calculation unit 72 (for example, the relevance calculation unit 106) When the relevance value indicating the strength of relevance is set to a predetermined value (for example, 0) and it is determined that the two specifications are related, the reversible compression size of each of the two specifications and 2 A relevance value is calculated based on a lossless compression size of a character string obtained by concatenating character strings of two specifications.

  The specification associating unit 73 (for example, the link generation unit 107), when the relevance value satisfies a predetermined condition (for example, when the relevance value is equal to or greater than the first threshold value), Associate the specifications.

  With such a configuration, it is possible to associate related specifications between texts including a plurality of specifications while minimizing manual work as much as possible.

  A part or all of the above-described embodiment can be described as in the following supplementary notes, but is not limited thereto.

(Supplementary note 1) From two texts including a plurality of ordered specifications, a set of two specifications belonging to different texts is selected, and a predetermined length that appears in common in the two specifications for each selected specification set A common character string that is the same character string, and a relevance presence / absence determining unit that determines the relevance of the two specifications based on the extraction result of the common character string; If it is determined that there is no relevance between two specifications for each set of specifications, a relevance value indicating the strength of relevance between the two specifications is defined as a predetermined value, and the relevance between the two specifications If it is determined that there is a relevance value, the relevance value is calculated based on the lossless compression size of each of the two specifications and the lossless compression size of the character string obtained by concatenating the character strings of the two specifications. Means and the relevance value is a predetermined condition If the relevant specification associating system, characterized in that it comprises a specification associating means for associating the two specifications was the calculation target of the relevance value satisfying.

(Supplementary Note 2) When the number of common character strings extracted from two specifications is greater than or equal to a predetermined number, the relevance presence / absence determining means determines that the two specifications are related, and the common character string The related specification associating system according to supplementary note 1, wherein the two specifications are determined not to be related when the number is less than a predetermined number.

(Supplementary Note 3) When the ratio of the number of extracted common character strings to the sum of the lengths of the character strings of the two specifications is equal to or greater than a predetermined ratio, When the ratio of the number of extracted common character strings to the sum of the lengths of the character strings of the two specifications is less than a predetermined ratio, the two specifications are related. The related specification matching system according to Supplementary Note 1, wherein it is determined that there is no.

(Supplementary note 4) The related specification associating system according to any one of supplementary notes 1 to 3, wherein the relevance presence / absence determining unit extracts a common character string that satisfies a predetermined condition with respect to the common character string.

(Supplementary Note 5) The specification associating means associates the two specifications that are the calculation targets of the relevance value on the condition that the relevance value is equal to or greater than a predetermined association determination threshold. The related specification matching system according to any one of the above.

(Additional remark 6) Based on the matching result regarding the specification of the order before the specification which is not yet matched, and the matching result regarding the specification of the order after the specification which is not yet matched, it is still the correspondence The related specification associating system according to any one of supplementary notes 1 to 5, further comprising specification correspondence complementing means for associating a specification that is not made and a specification belonging to a text different from the specification.

(Supplementary note 7) The specification correspondence complementing unit is a specification that has not yet been matched and is between a first specification belonging to one text and a second specification belonging to a text different from the first specification. In addition, the relevance value between the first specification and the second specification is not less than a complementary threshold, and the specification in the order before the first specification in the text to which the first specification belongs, A specification in the order before the second specification is associated with the specification in the text to which the second specification belongs, and the specification in the order after the first specification in the text to which the first specification belongs. And an additional note that associates the first specification with the second specification when there is a relationship that the second specification and the subsequent specification are associated in the text to which the second specification belongs. 6. The related specification matching system according to 6.

(Additional remark 8) It is provided with the phrase conversion means which converts the phrase contained in each specification according to a predetermined conversion rule, and a relevance presence determination means determines the presence or absence of relevance of two specifications for every set of the specifications after conversion. The related specification matching system according to any one of Appendix 1 to Appendix 7.

(Supplementary note 9) A pair of two specifications belonging to different texts is selected from two texts including a plurality of ordered specifications, and a predetermined length that appears in common in the two specifications for each selected specification set The common character string that is the character string is extracted, and based on the extraction result of the common character string, the presence or absence of relevance between the two specifications is determined, and for each set of two specifications belonging to different texts, If it is determined that there is no relevance between the two specifications, a relevance value indicating the strength of relevance between the two specifications is set to a predetermined value, and it is determined that the two specifications are related. In the case where there are two specifications, the relevance value is calculated based on the reversible compression size of each of the two specifications and the reversible compression size of the character string obtained by concatenating the character strings of the two specifications, If the relationship value is satisfied, Related Specifications mapping method characterized by associating two specifications Tsu.

(Supplementary Note 10) When the number of common character strings extracted from two specifications is equal to or greater than a predetermined number, it is determined that the two specifications are related, and the number of common character strings is a predetermined number. The related specification associating method according to appendix 9, wherein the two specifications are determined to be unrelated when the number is less than

(Supplementary Note 11) When the ratio of the number of extracted common character strings to the sum of the lengths of the character strings of the two specifications is equal to or greater than a predetermined ratio, it is determined that the two specifications are related. Supplementary note 9 that determines that the two specifications are not relevant when the ratio of the number of extracted common character strings to the sum of the lengths of the character strings of the two specifications is less than a predetermined ratio The related specification matching method described.

(Additional remark 12) The related specification matching method in any one of Additional remark 9 to Additional remark 11 which extracts the common character string which satisfy | fills a predetermined condition with respect to a common character string.

(Supplementary note 13) Any one of Supplementary note 9 to Supplementary note 12 for associating the two specifications that are the calculation target of the relevance value on condition that the relevance value is equal to or greater than a predetermined threshold value for association determination The related specification matching method described.

(Additional remark 14) Based on the matching result regarding the specification of the order before the specification which is not yet matched, and the matching result regarding the specification of the order after the specification which has not been matched yet, The related specification associating method according to any one of supplementary note 9 to supplementary note 13, wherein a specification that has not been assigned is associated with a specification that belongs to a text different from the specification.

(Supplementary Note 15) Between the first specification belonging to one text that is not yet associated and the second specification belonging to a text different from the first specification, the first specification And the second specification is greater than or equal to a complementary threshold value, and the specification in the order before the first specification and the second specification belong in the text to which the first specification belongs. The specification in the order before the second specification is associated in the text, the specification in the order after the first specification in the text to which the first specification belongs, and the second specification The related specification correspondence according to appendix 14, wherein the first specification and the second specification are associated with each other when there is a relationship that the specification in the order after the second specification is associated in the text to which it belongs. Attaching method.

(Supplementary note 16) Any one of Supplementary note 9 to Supplementary note 15 for converting a word / phrase included in each specification according to a predetermined conversion rule and determining whether or not the two specifications are related to each other after the conversion is performed. The related specification matching method described.

(Supplementary Note 17) Two sets of specifications belonging to different texts are selected from two texts including a plurality of ordered specifications on a computer, and appear in common in the two specifications for each selected set of specifications. Relevance presence / absence determination processing for extracting a common character string that is a character string of a predetermined length and determining the relevance of the two specifications based on the extraction result of the common character string, belonging to different texts If it is determined that there is no relationship between the two specifications for each set of two specifications, a relevance value indicating the strength of the relationship between the two specifications is set to a predetermined value, and the two specifications are When it is determined that there is a relationship, a relationship value is calculated based on the lossless compression size of each of the two specifications and the lossless compression size of the character string obtained by concatenating the character strings of the two specifications. Sex calculation processing and before If relevance value satisfies a predetermined condition related specifications correspondence program for executing the specification associating process for associating two specifications it became an object of calculation of the relevance value.

(Supplementary Note 18) When the number of common character strings extracted from two specifications is equal to or greater than a predetermined number in the relevance determination process, the computer determines that the two specifications are related, and The related specification association program according to supplementary note 17, wherein when the number of common character strings is less than a predetermined number, the two specifications are determined not to be related.

(Additional remark 19) When the ratio of the number of the extracted common character string with respect to the sum of the length of the character string of two specifications is more than a predetermined ratio by the relevance existence determination processing, it is said 2 If the ratio of the number of extracted common character strings to the sum of the lengths of the character strings of the two specifications is less than a predetermined ratio, it is determined that the two specifications are related. The related specification associating program according to appendix 17, which determines that there is no relevance.

(Supplementary note 20) The related specification associating according to any one of supplementary note 17 to supplementary note 19, which causes the computer to extract a common character string satisfying a predetermined condition for the common character string in the relevance existence determination process. program.

(Supplementary Note 21) In the specification association process, the computer associates the two specifications that are the calculation target of the relevance value on condition that the relevance value is equal to or greater than a predetermined association determination threshold. The related specification association program according to any one of Supplementary Note 17 to Supplementary Note 20.

(Supplementary Note 22) Based on the association result regarding the specification in the order before the specification not yet associated with the computer, and the association result regarding the specification in the order after the specification not yet associated, The related specification associating program according to any one of supplementary note 17 to supplementary note 21, which executes a correspondence specification correspondence complementing process for a specification that has not been associated yet and a specification that belongs to a text different from the specification.

(Supplementary Note 23) A first specification belonging to one text and a second specification belonging to a text different from the first specification, which has not yet been associated with the computer in the specification correspondence complement processing. Between the first specification and the second specification, the relevance value is greater than or equal to a complementary threshold value, and the first specification has an order before the first specification in the text to which the first specification belongs. The specification is associated with the specification in the order before the second specification in the text to which the second specification belongs, and the order after the first specification in the text to which the first specification belongs And the second specification are related to each other in the text to which the second specification belongs. Related specification pair described in appendix 22 Give program.

(Supplementary Note 24) Causes a computer to execute a phrase conversion process for converting a phrase included in each specification according to a predetermined conversion rule. In the relevance determination process, for each set of specifications after the phrase conversion process, two specifications The related specification associating program according to any one of supplementary notes 17 to 23 for determining the presence or absence of relevance.

  While the present invention has been described with reference to the embodiments, the present invention is not limited to the above embodiments. Various changes that can be understood by those skilled in the art can be made to the configuration and details of the present invention within the scope of the present invention. For example, some components may be deleted from all the components shown in the embodiment. Moreover, you may combine another component.

  This application claims the priority on the basis of the JP Patent application 2011-058314 for which it applied on March 16, 2011, and takes in those the indications of all here.

  The present invention is suitably applied to a related specification associating system for associating specifications described in text. The result of associating the specifications with each other can be used for detecting a reflection omission in the subsequent process of the specifications, confirming the consistency, and grasping the range of influence at the time of changing the specifications.

DESCRIPTION OF SYMBOLS 11 Memory | storage part 12 Calculation part 100 Initial specification memory | storage part 101 Conversion word memory | storage part 102 Conversion specification memory | storage part 103 Link memory | storage part 104 Word / phrase conversion part 105 Relevance existence determination part 106 Relevance calculation part 107 Link generation part 108 Link complementation part

Claims (10)

Selecting two sets of text or et two specifications, based on common string that appears in common to the two specifications, the relevance existence determining means determines the presence or absence of association of the two specifications,
If it is determined to be relevant to the two specifications, the two specifications each lossless compressed size, strings of the two specifications based on the lossless compression size of the string linked, the relevance value Relevance calculation means to calculate,
Related specification associating system, characterized in that it comprises a specification correlating means determines whether associating two specifications. The relevance determination unit determines that the two specifications are related when the number of common character strings appearing in common in the two specifications is equal to or greater than a predetermined number, and the number of the common character strings The related specification associating system according to claim 1, wherein when the number is less than a predetermined number, the two specifications are determined not to be related. When the ratio of the number of common character strings that appear in common in the two specifications with respect to the sum of the lengths of the character strings in the two specifications is equal to or greater than a predetermined ratio, determined to be relevant to the specification, to the length sum of the strings of the two specifications, the ratio of the number of the common strings, if it is less than the rate a predetermined, relevant to the two specifications The related specification matching system according to claim 1, wherein it is determined that there is not. The related specification matching system according to any one of claims 1 to 3, wherein the relevance presence / absence determination unit extracts a common character string that satisfies a predetermined condition with respect to the common character string. The specification associating means associates the two specifications that are subject to calculation of the relevance value on condition that the relevance value is equal to or greater than a predetermined association determination threshold value. The related specification matching system according to any one of the above. The relevance determination means selects two sets of specifications from two texts including a plurality of ordered specifications,
The specifications that are not yet associated based on the association results for the specifications in the order before the specifications that are not yet associated and the association results for the specifications that are in the order after the specifications that are not yet associated The related specification matching system according to any one of claims 1 to 5, further comprising: specification matching complementing means that matches a specification belonging to a text different from the specification. The specification correspondence complementing means is a specification that has not yet been associated and is between the first specification belonging to one text and the second specification belonging to a text different from the first specification. A relevance value between one specification and the second specification is equal to or greater than a complementary threshold, and the specification in the order before the first specification in the text to which the first specification belongs; and the second specification The specification in the order before the second specification is associated with the specification in the text to which the specification belongs, the specification in the order after the first specification in the text to which the first specification belongs, and the second 7. The first specification and the second specification are associated with each other when there is a relationship that the specification in the order after the second specification is associated in the text to which the specification belongs. Related specification mapping system. Comprising word conversion means for converting words included in each specification according to a predetermined conversion rule;
The related specification associating system according to any one of claims 1 to 7, wherein the presence / absence determining unit determines whether or not two specifications are related for each set of converted specifications. Selecting two sets of text or et two specifications, based on common string that appears in common to the two specifications, to determine the presence or absence of association of the two specifications,
If it is determined to be relevant to the two specifications, the two specifications each lossless compressed size, strings of the two specifications based on the lossless compression size of the string linked, the relevance value Calculate
It is determined whether or not two specifications are associated with each other . On the computer,
Selecting two sets of text or et two specifications, based on common string that appears in common to the two specifications, relevance existence determining process of determining whether the relatedness of the two specifications,
If it is determined to be relevant to the two specifications, the two specifications each lossless compressed size, strings of the two specifications based on the lossless compression size of the string linked, the relevance value Relevance calculation processing to be calculated, and
A related specification association program for executing a specification association process for determining whether or not two specifications are associated with each other .

Images