JP2900504B2 - Information processing device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION << Industrial Application Field >> The present invention has an expert system which enables a user to easily construct a system suitable for a selection problem and also allows a user to easily perform an operation of searching for information data. The present invention relates to an information processing device.

<< Conventional Technology >> In the information society in recent years, it is considered that the selection problem of selecting only necessary information (selection target) from a variety of influential information will increase in the future. To solve such a selection problem, a database system is used.

This conventional database management system is as follows.

For example, (1) when constructing a normal database,
Uses Data Description Language (DDL).

This defines the structure of the database, and the database is constructed according to this structure.

(2) When selecting data, a data manipulation language (Data Manipulation Language, DML) is used.

The database management system is software (program) that manages a database for realizing independence and sharing of data as a characteristic of the database. Further, an interface between the database and a user is a data manipulation language (Data Manipulation). Language, D
L) etc. are used.

The data description language is used to define the structure of a database, and the structure of the database refers to the data to be registered in the database, the attributes of each data, the relation between the data, and the like.

Data manipulation languages are used to read data or write data.

In addition, the relation between the relational database (RDB) and the SQL described below is described.
From the viewpoint of the method of expressing information, there are three types: a hierarchical model, a network model, and a relational model (relational model). Among these models, the relational database (RDB) is based on the relational model that is currently used most often.

Operations on this RDB are generally performed using SQL (Structured Que
ry Language).

In other words, it is configured to use the SQL to build and use the database.

<< Problems to be Solved by the Invention >> However, in such a conventional method of constructing a database and a method of selecting the same, it is necessary to use a difficult programming language such as a data description language or a data manipulation language. Is a difficult task.
In addition, in the selection method, a problem is selected by setting a key to be selected and a condition to be given to the key, and there is a problem that describing detailed selection conditions requires special technical skills. Was.

The present invention has been made in order to solve the above-mentioned problems, and an object of the present invention is to use an expert system that models a problem solving process of an expert who solves a selection problem, that is, to achieve a desired problem. By assigning meanings that can be understood by even a beginner to each of the selection targets to be selected, the burden on the user when selecting the selection targets is reduced, and fine selection conditions that can be understood by the user can be easily given. An information processing apparatus having an expert system that enables a beginner to easily select and build a database by reducing the burden on the user by using an expert system using the same model in the construction of the database. Is provided.

<< Means for Solving the Problems >> In order to solve the above problems, in the invention according to claim 1,
It has a framework for writing attribute data and other specific data as knowledge information in attributes representing the properties and characteristics of the selection target, and the selection condition is expressed in a sieve type that can be sifted without interdependency. When the attribute name and the specific data are selected, the selected object and the specific data included in the sieve type are searched, and the target selected object and the specific data are selected and extracted. An expert system comprising: a search control unit; and a selection result storage unit for storing a selection result selected and extracted by the control unit.

Further, the invention according to claim 2 has a framework for writing attribute data and other specific data as knowledge information in the attributes representing the properties and characteristics of the selection target, and the selection conditions are mutually dependent. Editing means for editing a sieve type that can be passed through a sieve having no sieve, storage means for storing the sieve type edited by the editing means, and a sieve generation device from a user / other system in addition to the sieve type. The default sieve to which the specific data given through is added is edited using the above sieve type, and the edited default sieve and a sieve group editing means for assigning a name to the sieve type group; and the sieve group editing means. Sieve group storage means for storing the edited default sieve, and selection target editing for editing the attributes of the selection target And an expert system for constructing a system comprising: a selection object storage unit for storing the selection object edited by the selection editing unit.

Furthermore, in the invention according to claim 3, as the attribute name and other specific data assigned to the attribute to be selected according to claims 1 and 2, a framework such as an attribute value and its range, a sieve method, and a comparison method is used. It is characterized by using a sieve type having.

According to a fourth aspect of the present invention, in the first aspect of the present invention, there is provided a framework for writing other specific data of an attribute name representing a property, a characteristic, or the like of a selection target, and the selection conditions are mutually dependent. Editing means for editing a sieve type that can be set so as not to have

According to a fifth aspect of the present invention, each of the sieves according to the first aspect is characterized in that a means for adding a note such as a constraint condition to a selection target and writing the note to a user is provided.

According to a sixth aspect of the present invention, the selection result storage means according to the first aspect is connected to a display means for displaying the selection result stored in the storage means.

<Action> According to the invention of claim 1 having the above configuration, a framework for writing attribute data and other specific data as knowledge information in attributes representing properties, characteristics, and the like of a selection target is provided, and It is configured to store a selection condition expressed in a sieve type that can be sifted without interdependency, to select and extract a target selection target and specific data, and to store the selection result. Therefore, without using a data description language when constructing the database, the name of the selection target is entered into the framework expressed in a sieve format and the desired data is described, so that even a beginner can easily select the necessary selection target and its data And data processing can be easily performed.

According to the invention of claim 2, a sieve type which can be sieved and whose selection conditions do not depend on each other is edited, the edited sieve type is stored, and a user / other system is added in addition to the sieve type. By editing the default sieve to which the specific data given through the sieve generating device is added by using the above-mentioned sieve type, and by assigning a name to the group of the edited default sieve and the sieve type, it is easily configured. A database can be constructed.

Furthermore, in the invention according to claim 3, as the attribute name and other specific data assigned to the attribute to be selected according to claims 1 and 2, a framework such as an attribute value and its range, a sieve method, and a comparison method is used. Since the sieve type is used, the user can write as needed at any time, and can freely make changes and the like, thereby improving usability.

According to the fourth aspect of the present invention,
Because there is a framework for writing other specific data of attribute names representing the properties and characteristics of the selection target, and an editing means for editing a sieve type that can be selected so that the selection conditions do not have a mutual dependency. Thus, the effects described in claims 1 to 3 can be obtained, and the usability is further improved.

In the invention described in claim 5, in each of the sieves described in claim 1, a notice such as a constraint condition is added to a selection target, so that mistakes on the user side are prevented, and excellent operability is achieved. Become.

According to the sixth aspect of the invention, the selection result is displayed in the selection result storage means of the first and second aspects, so that the selection target and the data of the selection target can be surely confirmed.

<< Example >> Hereinafter, an example of the present invention will be described with reference to the drawings.

FIG. 1 shows a first embodiment in which the apparatus of the present invention is applied to a host computer. A mouse 1, which is a sieve generating apparatus, is detachably mounted on a host computer 60 via an expansion board 61. By operating one of the buttons 1a and 1b, a signal of a mode commanded by the mouse 1 is displayed on the display 60a.

In this embodiment, a mouse was used as the sieve generating device. However, as shown in FIG. 2, a touch panel type display is used for the display 60a, and the screen is displayed with a finger or a touch pen.
Needless to say, it is also possible to directly press by 62.

The expert system A used in the host computer 60 has, as shown in the block diagram of FIG.
Schematic configuration of a selection target storage device 2, a sieve type storage device 3, a selection result storage device 4, a sieve storage device 5, a mouse 1, which is a sieve generation device, a sieve control device 6, and an intermediate result storage device 7. Have been.

In the P / ROM of the expert system shown in the circuit block diagram of FIG. 5, data to be stored in the sieve type storage device 3 and the selection target storage device 2 in advance are defined at the time of system construction as a sieve expert program. In the figure, reference numeral 9 denotes an A / D converter.

The selection target storage device 2 includes a name given to an attribute to the selection target and an attribute representing information such as the property and patent of the selection target. It has the function of storing a plurality.

Next, the sieve-type storage device 3, as shown in FIG.
Memorize attribute names, attribute values, sieving methods, comparison methods, etc.
The sieve represents one attribute, and the sieve type is given an attribute name. That is, only the framework for the attribute value / comparison method is prepared.

The sieving method is a method of sieving, and sets conditions such as whether to select what remains after sieving or to select what has been sieved.

Next, as a comparison method, a method of comparing the attribute value of the sieve with the attribute value of the selection target, that is, if these attribute values are “equal” or the attribute value of the selection target is A condition such as "large" or "small" is given.

That is, the sieve type storage device 3 has a function of storing a plurality of sieve types.

The selection result storage device 4 is a device that stores the result of the selection process of the selection target performed by sieving and the selected selection target.
The system is configured to output the selection to a user or another system via an interface circuit 12 to a host computer 13 including a personal computer, a word processor and other systems.

The sieve stored in the sieve storage device 5 stores an attribute name, an attribute value, a sieve method, a comparison method, and the like.

All values are given to the sieve types, and the sieve storage device 5 has a function of storing the sieves and the order of sieving them. On the other hand, it is expressed by giving integer numbers in ascending order from 1. That is, with respect to sieves to which different numbers are given among sieves, a selection target satisfying a condition shared by all sieves is selected from among selection conditions to which those sieves are given (sieve and condition). Further, with respect to the sieves given the same number, a sieve that satisfies one of the sieve conditions among the selection conditions given to those sieves is selected (a sieve or condition).

The mouse 1 serving as the above-mentioned sieve generating device creates a sieve by incorporating an attribute value, a sieve method, and a comparison method given from a user or another system, using the sieve type stored in the sieve type storage device 3 as a framework. Is created, and the created sieve and the order of the sieve are written in the sieve storage device 5 described later.

The sieve control device 6 calls the sieve and the order of sifting to the sieve storage device 5, sieves the selection target read from the selection target storage device 2 in accordance with the sieving order, and performs a selection process. After that, the selection result is written into the selection result storage device 4.

Note that the data stored in the sieve type storage device 3 and the selection target storage device 2 is defined in the expert system.

Next, an algorithm using the apparatus of the present invention will be described with reference to the explanatory diagram of FIG.

(1) As a condition for selecting a selection target from the outside such as a user or another system, a sieve type stored in the sieve type storage device 3 is read out, and an attribute name, an attribute value, a sieve method, and a comparison method are input. Enter the order of sieving them.

(2) Based on the conditions input in (1), a sieving order is generated by the sieving generation device and stored in the sieving storage device 5.

(3) The sieve control unit 6 reads out all sieves and the order of sieving from the sieve storage device 5 and stores them.

(4) Call all the selection targets from the selection target storage device 2 into the sieve control device 6 and write them to the intermediate result storage device 7.

(5) The sieve controller 6 sifts the corresponding sieves in the order of sieving.

Next, the sieving algorithm will be described.

(A) A counter (a sieve order counter) for counting the number of the sieve order (called a sieve number) is initialized to one.

(B) If the value of the sieve order counter is larger than the maximum sieve number, the sieving process ends.

(C) If no selection target exists in the intermediate result storage device 7, the sieving process ends.

(D) Select all sieves having a sieve number equal to the value of the sieve order counter.

(E) The sieve control device 6 reads all selection targets in the intermediate result storage device 7 and deletes all selection targets in the intermediate result storage device 7.

(F) Read the attribute name, attribute value, and comparison method of the sieve,
For the attribute name and attribute value, select all attribute names
Match the attribute values according to the comparison method.

(G) The sieving method of the sieve is read out, the selected objects matched in (b) are sifted according to the sieving method, and the selected selected objects are written in the intermediate result storage device 7.

(H) For all sieves selected in (d), (f)
To (g) are repeated.

(I) When the processing of (h) is completed, the value of the sieve order processing counter is incremented by one.

(J) Return to (b).

(6) If there is no selection target in the intermediate result storage device 7, the selection process ends.

(7) The selection target in the intermediate result storage device 7 is stored in the selection result storage device 4 by the sieve control device 6.

(8) Host computer 13 for users, other systems, etc.
The selection result obtained via the interface circuit 12 is output to the outside.

According to this information processing apparatus as is apparent from the above configuration. Based on the sieve-type model as described above, the process of selecting the target object is performed by sifting the information possessed by the object to be selected through a sieve as a selection condition, thereby providing an expert system for solving the selection problem. Can be applied.

Next, as a second embodiment of the apparatus of the present invention, an example of a tool for constructing an expert system suitable for a choice-type problem will be described.

The expert system construction tool is configured as shown in FIG. That is, the values edited by the sieve type editing device 13 based on the user expert and the user program of the expert are stored in the sieve type storage device 3, and the stored values are stored in the sieve type group editing device.
While storing in the sieve group storage device 15 via 14,
The information stored in the sieve type storage device 3 is stored in the selection target storage device 17 via the selection target editing device 16.

Further, the sieve type edited by the sieve type editing device 13 has a configuration as shown in FIG.
That is, the sieve type is a framework of the sieve, in which the attribute name, the data type, the range of the attribute value, the sieve method, the comparison method, and the like are described.

 Here, each name is defined.

(1) The attribute name indicates the name of the attribute.

(2) The data type indicates the data type of the attribute value.

(3) The attribute value range indicates the range of values that the attribute value can take. For example, when the data type is a numeric type, the maximum and minimum values that the attribute value can take are represented by numerical values. Is done. In the case of a name type, it is represented as a set of names that can be taken as attribute values.

(4) The sieving method refers to a method of sifting to select what has passed through a sieve or to select what has passed through a sieve.

(5) The comparison method is a method of comparing the attribute value of the sieve with the attribute value of the selection target, for example, “equal”, “over”,
"Below".

That is, only the attributes of the attributes of the selection target in the selection problem solving the sieve type are defined. The sieve type defined above is stored in the sieve type storage device 3.

The sieve type storage device 3 at the time of this editing stores the sieve type edited by the sieve type editing device 13.

The sieve group editing device 14 reads out the sieve type stored in the sieve type storage device 3 when setting the differential sieve, and performs editing using the sieve type. The edited character (sieve) The sieve group is written to the sieve group storage device 15. (1)
The sieve differt sieve is a sieve type, in addition to an attribute value, a range of attribute values, a sieve method, and a comparison standard.These are attributed to the attribute values edited by the sieve type editing device 13 as necessary. The range, sieving method and comparison standard have been changed. (2) A sieve group is a group of several default sieves, and a name is given to this group, and this name is called a sieve group name.

The sieve group is given the order of the sieves in the group. Fig. 10 shows the sieving order, and as shown in this figure, the defaults A, B, and C are sieving types A, B, and C, respectively.
Is generated as a framework.

Sieve group X has default sieves A and B.

In addition, sieve group Y has default sieves B and C
And the default sieve B has a default value suitable for each group in the groups X and Y.

The sieve is sifted in the order of default sieves A and B in sieve group X, and all the selection targets are selected under the conditions given by these sieves.

In sieve group X, default sieve B,
C is sieve at the same time, and a selection object satisfying at least one of the sieve conditions is selected from these sieves.

For example, in the selection of resin materials, sieves are used as characteristics / characteristics of the resin, and the sieve group edits the default sieve for the resin application. , Heat resistance temperature is 80 ℃
If the above-mentioned default sieve is prepared and the application is a terminal block, a default sieve with an attribute and heat resistance temperature of 120 ° C or more is prepared.
The selection process can be performed.

Next, the sieve group editing device 14 in the sieve group storage device 15 stores the edited default sieve.

The selection target editing device 16 edits the attributes of the selection target, and it is assumed that the type of the attributes of the selection target has been edited by the sieve type editing device 13.

The selection target editing device 16 edits the name of the selection target and its attributes, and edits the attributes in the sieve type storage device 3.
From the sieve type, and edit the attribute to be selected using the attribute name and attribute value range of the sieve type.
For the attribute value input, enter a value that falls within the attribute value range.

That is, the attribute in FIG. 11 is composed of a pair of an attribute name and an attribute value, and each attribute to be selected is edited using a sieve type.

The selection target editing device 16 is used as a device for storing the selection target edited by the selection target editing.

Thus, in this embodiment, when constructing an expert system, it is possible to construct a system by defining a sieve as a selection condition without defining a rule SQL and defining a selection target.

Next, a system configuration when the present invention apparatus is used will be described.

FIG. 12 shows a third embodiment of the device of the present invention. In this case, the sieve group designation device 18 is required to perform a selection process from the sieve groups stored in the sieve group storage device 15. The user specifies the name of the sieve group that satisfies the following conditions, and writes the default sieve and the order of sifting the specified sieve group into the sieve storage device 3.

In the sieve editing device 13, when it is necessary to change the default sieve's attribute value, sieve method, comparison method, or the like, it is necessary to prepare a sieve other than the default sieve. When performing, the user edits the sieve stored in the sieve storage device 3 or edits a sieve other than the default sieve using the sieve type stored in the sieve storage device 3.

When a sieve is deleted or changed, the order of sieving is calculated from the order of sieving within a sieve group.

That is, the number of the deleted item is calculated by packing its number, and the added item is stored last.

 The editing result is written to the sieve storage device 3.

Next, the intermediate result storage device 7, which is provided in the sieve control device 6 during processing, a so-called register will be described.
Used to store intermediate results during the selection process,
When the selection process is completed, all the selection targets stored here are written in the selection result storage device 4.

The selection result display device 60a displays the selection result to the user.

The processing algorithm of each of the above systems is as follows.

(1) The sieve editing device 13 performs sieve type editing.
The edited sieve type is written to the sieve type storage device 3. FIG. 13 shows a sieve type editing flowchart.

FIG. 13 is a sieve type flowchart, in which step 100
Then, the sieve type is read from the sieve type storage device 3 and, in the case of new creation in the next step 101, the attribute name is input in step 106 and the data type is input in step 107 as indicated by. Then, when the range of the attribute value is input in Step 108, the sieve is input in Step 109, and then Step 1
At 10, a comparison method is input, and the process proceeds to step 104.

When the process is terminated in step 104, the process proceeds to step 105, where the edited sieve type is written into the sieve type storage device 3 and the process ends.

Also, if there is a change in step 102, the process proceeds to step, and if it is changed in step 111, the process proceeds to the next step 112,
Change the attribute name.

If the data type is to be changed in the next step 113, the data type is changed in step 114. When the change is completed, the change of the attribute value range is confirmed in step 115, and if YES, the attribute value range is changed in step 116, and after that, the process proceeds to step 117, where the sieving method is performed. Changes are confirmed, and if any, step 11
The process proceeds to step 8 to change the sieving method, and then proceeds to step 119.

When the comparison method is changed in step 119, the process proceeds to step 120, where the comparison method is changed, and the edited sieve type is written to the sieve type storage device 3 in step 105 in the manner described above, and the process ends.

Further, if it is determined in step 103 that the user wants to delete, the process proceeds to step 121, where a sieve type to be deleted is specified.

Thereafter, in step 122, the specified sieve type is deleted, and the process proceeds to step 105 in the manner described above.
Write the edited sieve type to the sieve type storage device 3 with EN
D.

FIG. 14 is a flowchart of group editing, and FIG. 15 is a flowchart of a selection target. Since each flow is the same as the above-mentioned sieve type flow, detailed description thereof will be omitted.

(2) The sieve group is edited by the editing device 13 in accordance with the flow described above, and the edited default sieve is written in the sieve group storage device 15.

This default sieve editing flowchart is
As shown in the figure, the flow of this flowchart is well understood by those skilled in the art, and a detailed description thereof will be omitted.

(3) The selection target is edited by the selection target editing device 16, and the edited selection target is written in the selection target storage device 2. FIG. 17 shows a flowchart of the selection target editing. The detailed description of this flow is also omitted.

As described above, the sieve group editing device 14 and the selection target editing device 16 perform editing by using the sieve type stored in the sieve type storage device 3. Therefore, in order to perform these edits, the sieve type needs to be edited. Since both the sieve group and the selection target have no dependency, either one may be edited first.

Next, FIG. 20 shows a schematic algorithm of the processing of the system when using the apparatus of the present invention.

(1) The sieve group designation device 18 designates a sieve group.

The specified sieve group is read from the sieve group storage device 15 and written to the sieve storage device 4.

(2) The sieve editing device 13 edits the sieve stored in the sieve storage device 5 and the order of the sieve, and edits the new sieve order by using the sieve type stored in the sieve type storage device 3.

(3) The selection process is performed by the sieve control device 6, and the obtained selection result is written in the selection result storage device 4.

(4) The selection result obtained by the selection result storage device 4 is displayed on the display 60a.

(5) If the result of the selection is different from the purpose, the processing of the above (1) or (2) is repeated (refer to the flowcharts of FIGS. 16 and 17).

FIGS. 21 to 27 show display screens for each mode displayed on the display 60a.

In this embodiment configured in this way, the selection condition, ie, the sieve that becomes the knowledge of the expert performing the selection work is independent of the other sieves, without taking into account the dependency between the multiple sieves, It is possible to easily add, modify, and delete sieves.

In addition, since the control of the inference for performing the selection process is performed by the sieve control device that controls the independent sieve, it is not necessary to describe control knowledge.

 An outline of another embodiment of the present apparatus will be described below.

(1) As shown in FIG. 20, in the system at the time of construction, the configuration of the system at the time of use is combined, and the order in which the sieve control device 6 is sifted is traced. Alternatively, it is possible to construct a debugger that can be used for debugging at the time of construction by stopping after performing a designated sieve and referring to the intermediate result selected in that state from the intermediate result storage device 7.

(2) In the above embodiment, the sieve attribute value is determined in the sieve type definition in order to statically determine the data type. However, by dynamically checking the data type, this description becomes unnecessary.

(3) In the above-described embodiment, the sieve group is specified in the sieve type group specification value. However, the priority order when the default sieve in the sieve group is duplicated among a plurality of groups is defined by the sieve editing device 13. By doing so, it becomes possible to specify a plurality of sieve groups in the sieve group specifying device 18.

(4) Additional information such as more detailed information about the selection target and information that is difficult to sift is prepared in another storage device, and is used when the selection result is displayed on the selection result display device 60a. By displaying the additional information arbitrarily by the user, the user can check and determine whether the selection result is the target selection target by referring to the additional information.

(5) In the above embodiment, the comparison method is “equal”, “over”, “below”, etc. However, the comparison method corresponding to the sieve is a comparison method having ambiguity based on fuzzy theory. Of course, you can do that. This also makes it possible to provide an ambiguous selection condition.

(6) In the above embodiment, the sieving method is a binary method of passing or not passing through the sieve, but in this regard, a sieving method having ambiguity based on fuzzy theory may be used.

Thereby, it is possible to give ambiguity in narrowing down at the time of selection.

(7) The algorithm of each device in the above embodiment may be realized by software or hardware.

(8) In each of the sieves of the above embodiments, a precautionary statement such as a constraint condition may be added to the selection target, and a writing device for performing a selection process in a subsequent sieve and a precautionary note for a user may be added. FIG. 19 and FIG. 20 show this example.

In other words, the precautionary statement is given to the selected object when it passes through a sieve with a condition such as a constraint condition, and this precautionary note is given during the selection process using the sieve. If you use a notice to control whether or not to pass through the sieve and display the result of the selection, display this notice at the same time so that the user can refer to the notice and evaluate the result of the selection. Can be.

In other words, it is possible to independently describe the dependency between the sieves by the precautionary statement, and to control the selection process finely, and to display the selection result to the user instead of simply displaying the selection result. By giving a note to the result, a more accurate selection process can be performed.

In order to add the above-mentioned note control device 20 to the above-described embodiment, if a note is added to the configuration of the sieve as a selection target, it is considered that a selection process using a plurality of sieves is performed on the selection target. Thus, it is configured to have a plurality of notices.

Next, a description will be given of a sieve configured to have one notice.

As the above note method, as shown in FIG. 19, a note issuance condition indicating a condition for issuing the note, an attribute value for changing the contents of a sieve when the note is issued,
It consists of a sieving method, a comparison method, and notes.

The precautionary note issuance condition is added with a condition such as what kind of selection condition is given, what sieve is given in the selection process, or whether it is given.

The attribute value, sieve method, and comparison condition that change the contents of the above sieve are values for changing the sieve when giving a notice, and when changing the respective values of the sieve when giving the notice Used for

The precautionary note control device 20 performs processing according to the precautionary note during the selection process by the sieve in the sieve control device 6, and the processing algorithm of the precautionary note control device 20 will be described below.

That is, (1) when the precautionary note issuance condition is satisfied, the contents of the sieve are rewritten to a value according to the precautionary note method. (2) Using the sieve, a selection process is performed by the sieve control device. (3) A notice is given to the selection target selected in the above (2).

The selection result display device 60a also displays a note on the selection result display.

Thus, according to this embodiment, since the sieve has an independent comparison method and an independent sieve method, it is easy to add and expand a sieve having a new comparison method and a sieve method. it can.

(9) In the second embodiment, the device for storing the selection target is an existing database, and an apparatus for generating a data description language and a data manipulation language of the database is added to the existing database and the expert system according to the second embodiment. The combination can be easily performed.

(10) In the second embodiment, the sieve type editing device 13, the sieve group editing device 14, and the selection target editing device 16 are provided with an editor in the form of a table, so that the user's input can be minimized.

In addition, when an expert system is constructed in this way, the database can be easily created and changed, and a KE (Knowledge Engineer) and a database expert engineer are not required.

<< Effects of the Invention >> As is clear from the above configuration, according to the information processing apparatus according to the present invention, based on the sieve type model as described above, a target to be selected is selected from information having the selected target sieve. By performing the process, a process of selecting a target object is performed, and the present invention can be applied to an expert system for solving a selection problem. That is, according to the present apparatus, the following effects can be obtained.

(1) When constructing an expert system, it is possible to easily construct a system by defining a sieve as a selection condition without defining a rule SQL and defining a selection target.

(2) Sifting conditions, that is, sieves that become the knowledge of the expert performing the election, are independent of other sieves and do not take into account the dependencies among multiple sieves.・ Can be modified / deleted, making it easy to use.

(3) Anyone can easily search and edit by giving meaning to the attribute to be selected.

(4) Since the sieve has an independent comparison method and an independent sieve method, it is possible to easily add and expand a sieve having a new comparison method and a new sieve method. Therefore, when the expert system is constructed, the creation and modification of the database can be easily performed, which has the effect that the KE and the database expert engineer become unnecessary.

[Brief description of the drawings]

FIG. 1 is a perspective view showing a first embodiment of the present invention, and FIG.
FIG. 3 is a perspective view showing a second embodiment of the apparatus of the present invention, FIG. 3 is a block diagram showing the overall circuit block configuration of the apparatus of the present invention, and FIG. Fig. 5, Fig. 5 is a configuration diagram of an expert system based on a sieve model, Fig. 6 is an explanatory diagram showing operation of a sieve control device to be selected, Fig. 7 is an explanatory diagram showing operation of the sieve control device, Fig. 8 Is a configuration diagram of the system at the time of construction, FIG. 9 is an explanatory diagram showing a sieve type structure, FIG. 10 is an explanatory diagram showing a relationship between a default sieve and a sieve, FIG.
FIG. 12 is an explanatory diagram for explaining a structure of a selection target, FIG. 12 is an explanatory diagram showing a configuration of a use-time system, FIG. 13 is a flowchart showing a type editing flow, and FIG. 14 is a flowchart showing a sieve group editing flow. , FIG. 15 is a flowchart showing an editing flow of a selection target, FIG. 16 and FIG. 17 are flowcharts showing a flow of a use-time system, and FIG. FIG. 19 is a block diagram showing a cautionary note type configuration, and FIG.
FIG. 20 is an explanatory diagram showing a system in use with a precautionary note control device added thereto, FIG. 21 is an explanatory diagram showing an editing state of each sieve data displayed on a display of the present application device, and FIG. FIG. 23 is an explanatory diagram showing a debug menu, FIG. 24 is an explanatory diagram showing an embodiment of an attribute edit window for selection, and FIG. 25 is a diagram for selecting and defining each name of the attribute for selection. FIG. 26 is an explanatory diagram showing an embodiment of a file reading operation dialog box, and FIG. 27 is an explanatory diagram showing a state in which evaluation attributes displayed on the display of the device of the present invention are set. It is. A ... Expert system 1 ... Mouse (sieving generation device) 2 ... Selection storage device 3 ... Sieving type storage device 4 ... Selection result storage device 5 ... Sieving storage device 6 ... Sieving control device 13 ... Sieve type editing device 14 ... Sieve group editing device 15 ... Sieve group storage device 16 ... Selection target editing device 17 ... Selection target storage device 60a ... Display

──────────────────────────────────────────────────続 き Continued on the front page (58) Fields surveyed (Int.Cl. ⁶ , DB name) G06F 17/30 G06F 9/44 550 JICST scientific and technical literature file

Claims (6)

(57) [Claims] 1. A sieve which has a framework for writing attribute data and other specific data as knowledge information to attributes representing the properties and characteristics of a selection target, and wherein selection conditions have no mutual dependency. A storage means for storing what is expressed in a sieve type, and when the attribute name and the specific data are selected, the selection target and the specific data captured in the sieve type are searched,
An expert system comprising: a search control unit for selecting and extracting a target selection target and specific data; and a selection result storage unit for storing the selection result selected and extracted by the control unit. An information processing apparatus characterized by the above-mentioned. 2. A sieve having a framework for writing attribute data and other specific data as knowledge information to attributes representing the properties and characteristics of the selection target, and having selection conditions having no mutual dependency. Editing means for editing the sieve type; storage means for storing the sieve type edited by the editing means; and specific data provided via a sieve generating device from a user or another system in addition to the sieve type Edit the default sieve with the added sieve type using the above sieve type, a sieve group editing means for giving a name to the edited default sieve and sieve type group, and a default sieve edited by the sieve group editing means. Sieve group storage means for storing, selection target editing means for editing the attributes of the selection target, and selection and editing thereof The information processing apparatus characterized by comprising a selection storage means for storing the selected object edited by stage, the expert system for system construction consisting of. 3. A sieve type having an attribute value and a framework such as a range, a sieve method, and a comparison method is used as the attribute name and other specific data assigned to the attribute to be selected according to claim 1 or 2. An information processing apparatus, characterized by 4. The invention according to claim 1, further comprising a framework for writing other specific data of an attribute name representing a property / characteristic of a selection target, and having selection conditions mutually dependent. An information processing apparatus comprising editing means for editing a sieve type that can be set not to be edited. 5. The sieve according to claim 1, wherein
An information processing apparatus, comprising: means for adding a note such as a constraint condition to a selection target to make a note to a user. 6. An information processing apparatus according to claim 1, wherein said selection result storage means is connected to a display means for displaying the selection result stored in said storage means.