JPH08123817A - Analysis knowledge management device - Google Patents

Abstract

PURPOSE: To suppress the index quantity to be given to the knowledge information and to flexibly change the index quantity by forming a hierarchical constitution of items of the relative information, producing an index where the component elements are connected together in the hierarchical order and holding this index. CONSTITUTION: An index is produced to be given to the knowledge information when a knowledge base is retrieved, and this index is stored and managed by a management device together with the knowledge information on the knowledge base. An index production means 100 of the management device forms a hierarchical constitution of items of the relative information, produces an index by connecting together the component element in the hierarchical order and holds this index. Then an analysis program deciding means 102 of the means 100 decides an analysis program in response to the analysis sorting information selected by a selection means 101. A connection means 150 connects together the index component elements which are acquired by both means 101 and 102 and an analyzing object selection means 103. Then a connection means 160 connects the identifier of a data card to the index produced by the means 150.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an analysis knowledge management device, and more particularly to an analysis knowledge management device for managing knowledge information used in various analysis operations of an analysis target. For details, when performing various analytical calculations of certain target requirements, in the process of creating data for calculation and when examining the calculation results, specialized knowledge of analysis and various analytical calculation programs (hereinafter, Experiential know-how for each solver is required. The present invention relates to an analysis knowledge management device for managing knowledge, which is information such as appropriate advice and technical terms, and presenting it to analysis processing when supporting such work.

[0002]

2. Description of the Related Art The first conventional method is to register and retrieve knowledge information using an index. FIG.
FIG. 7 is a diagram for explaining the first method of conventional knowledge management. In the figure, when the knowledge information 12 is registered in the knowledge file 2, the index 11 is added and written in the knowledge file 2. When searching the knowledge information registered in the knowledge file 2, the knowledge file 2 is searched using the index 11 as a key.

The second conventional method is to register the information in the knowledge file without adding the index to the information and perform the search with the free keyword. FIG. 18 is a diagram for explaining the second conventional method of knowledge management. In the figure, when the registration information 1 is written as it is in the knowledge file 2 and a search is performed, an arbitrary free keyword 3 representing the registration information 1 is input to search / acquire the knowledge information 4. For example, "virus" is input as a free keyword when searching for "antibody" from the knowledge file 2, or linked free keywords such as "disease, blood, white blood cells" are input for searching.

[0004]

However, in the first conventional method, the index is added to the information and input, but generally, the knowledge about analysis is not quantitative data such as a document or an image. In many cases, there is a problem that it is difficult to express with an index of an appropriate size. Furthermore, in the first conventional method, the data such as the explanation of technical terms and the advice at the time of analysis may change depending on the analysis field and method, and a different index must be given each time. An index is generated and it is difficult to create and manage it.

The second conventional method is an ambiguous search. When one free keyword is input and desired knowledge information cannot be searched, the keyword input direction is changed and input. In addition, there is a possibility that an unintended search result may occur, and it takes time to search the knowledge information because such a method is repeated. The present invention has been made in view of the above points, and when registering the conventional knowledge information, or when performing an analysis process using the registered knowledge information, the amount of the index to be added to the knowledge information. It is an object of the present invention to provide an analysis knowledge management device capable of flexibly changing the index while suppressing the above.

[0006]

FIG. 1 is a block diagram showing the principle of the present invention. The analysis knowledge management device of the present invention hierarchically configures related information items in the analysis knowledge management device that generates an index to be added when searching a knowledge base, stores the index together with the knowledge information of the knowledge base, and manages the index. Then, each component is combined in hierarchical order to generate an index,
It has an index generating means 100 for holding it.

The index generating means 100 described above is also used.
Is an analysis classification information selecting means 101 for selecting analysis classification information indicating a classification of information to be analyzed, and an analysis program determining means for determining an analysis program corresponding to the analysis classification information selected by the analysis classification information selecting means 101. 1
02, analysis object selection means 103 for selecting an object to be analyzed, analysis classification information selection means 101, analysis program determination means 102, and analysis object selection means 10
A first combining unit 150 that combines the constituent elements of the index acquired in 3 to generate the first index;
A second index is generated by combining the first index generated by the first combining means with the identifier of the data card made up of text data or data other than the text data for executing the analysis program.
Of the connecting means 160.

The present invention also provides an index generating means 1
It further has a knowledge base registration unit 200 which registers knowledge information in association with the knowledge information using the index generated by 00 as a registration key. Further, according to the present invention, a search index is created according to the status of the index analysis generated by the index generation unit 100, and the knowledge base is searched by the search index when the index matches the index generated by the index generation unit 100. The search means 300 is provided.

The above-mentioned search means 300 searches the hierarchy one level higher when the knowledge base cannot be searched with the search index, and changes the search index with the above-mentioned constituent elements. It has a changing means 301.

[0010]

According to the present invention, related items are assembled in a hierarchical structure, and an index (hereinafter, referred to as a path) is generated by combining a plurality of the items by tracing the hierarchy. It is to be registered in the knowledge base in association with. In this way, if there is a path in the knowledge base, knowledge information can be obtained by searching the knowledge base by creating a path at the time of searching in the same manner as at the time of creation.

Further, even if the path used at the time of search is not registered in the knowledge base, since the related information is hierarchically held, by changing a part of the path along the upper hierarchy, It is possible to acquire highly relevant knowledge information more flexibly than the knowledge base.

[0012]

EXAMPLES Examples of the present invention will be described in detail below. FIG. 2 is a system configuration diagram of an embodiment of the present invention. The analysis knowledge management device shown in the figure includes an analysis template generation unit 100 for generating an index (template) to be added to knowledge information, an analysis template file 10 for storing the template generated by the analysis template, and knowledge for analysis. A knowledge registration unit 200 for registering information, a knowledge base 20 for registering information in which knowledge information from the knowledge registration unit 200 is associated with a template generated by the analysis template generation unit 100, an analysis template file 10, and a knowledge base 20. The analysis unit 300 refers to and analyzes the target requirement.

FIG. 3 is a diagram for explaining an analysis template generation unit according to an embodiment of the present invention. An input / output unit 10, a selection information file 40 having menu screen information, analysis field display information, analysis target display information, and the like, and an analysis template file for registering the generated analysis template as input / output with the analysis template generation unit 100. 1
0, a data card file 30 that holds a data card to be added to the analysis template, and the like. The data card file 30 will be described later with reference to FIG.

The above-mentioned analysis template generation unit 10 will be described below.
The operation of 0 will be described. [Analysis Template Generation Process] FIG. 4 is a flowchart of the analysis template generation process according to the embodiment of the present invention. (Step 101) First, in the case of newly creating the analysis template file 10, the process proceeds to Step 102,
In other cases, for example, when the analysis template is used when performing an analysis target search or the like, the process proceeds to the process of the flowchart shown in FIG.

Step 102) The analysis template generation unit 100 reads out the analysis target item displayed on the monitor device of the input / output unit 110 from the selection information file 40 and transfers it to the input / output unit 110. Accordingly, the input / output unit 110
Displays an analysis field selection screen as shown in FIG. 6 and prompts the user to make a selection input. In the example of the figure, as the analysis field,
Major classification items in fields such as "non-linear analysis", "linear analysis", and "resin flow", and minor classification items subordinate to the major classification item are displayed. For example, “elastoplastic”, “hyperelasticity”, “creep”, etc. are displayed as small classification items corresponding to the large classification item “non-linear analysis”. The user clicks the display screen with the mouse of the input instructing unit 410 or the like to select the large classification or the small classification.

Step 103) The analysis template generation unit 100 searches the selection information file 40 to determine whether a plurality of analysis program information (solvers) corresponds to the field designated and input by the user. FIG. 6 is an example of an analysis program corresponding to the field information according to the embodiment of the present invention. In the figure, for example, in the selection information file 40, "elastoplasticity" of a small classification item of the analysis field "nonlinear analysis" is included.
It is shown that only the solver that is analyzed for “solver A” corresponds. It is shown that "creep" corresponds to both "solver A" and "solver B". When there are a plurality of solvers for the sub-category of one field, the process proceeds to step 104, and one-to-one
If it corresponds to, the process proceeds to step 105.

Step 104) When a plurality of solvers correspond to a small classification item in one analysis field, one of them is selected by clicking with the mouse. (Step 105) Next, the input / output unit 110 inputs the analysis target information from the keyboard. As shown in FIG.
A hierarchical screen as shown in is displayed, and the user inputs characters in the box on the screen. For example, if you want to analyze the "printer", enter "printer", then "print head", "platen",
"Print head" Serial number of print head "DN4953
Enter 24-001 ”or“ DN495325-003 ”etc.

As a result, layered information including the field a (), the solver s to be used, and the object o () is prepared. In addition, the identifier (ID) of each information shall be set in (). Step 106) Analysis template generation control unit 420
Is a template T1 ((a (). S.) As a template in steps 102, 103, 104 and 105.
o ())). The created template has a hierarchical structure as shown in FIG. In the figure, the thick line indicates the path selected by the user.

(Step 107) The analysis template generation control unit 420 acquires from the data card file 30 a data card group corresponding to the solver of the template T1 ((a (). S.o ())) created in Step 106. I'll do it. As shown in FIG. 9, the data card file 30 is configured to correspond to the analysis field / solver. The data card is the data necessary for the solver to perform various calculations. For example, the constraint condition value and physical property value as the type of data, and the input range and each input value as the generation condition There is information about dependency conditions. The data card read from the data card file 30 has only the ID information of the card, and the data value used by the solver at the time of inputting the data card described later is set.

The data card file 30 shown in FIG. 9 is registered as a set of an analysis target, a solver, and a data card. The data cards corresponding to "Solver A" are load condition card (ID: WT) and restraint condition card (ID:
RT), material characteristic card (ID: MT), calculation control card (ID: CL), non-linear characteristic card (ID: NL), etc. (however, () indicates I of the card.
D).

Step 108) The template to which the data card ID (d) is added in step 107 is stored in the analysis template file 10. The above processing has been described with respect to the case where a new analysis template file 10 is created, but a case where the template T1 is read from the created analysis template file 10 will be described as preprocessing for performing the data card editing processing described later.

FIG. 10 is a flow chart showing the processing when reading the analysis template file according to the embodiment of the present invention. (Step 201) In the input / output unit 110, a screen for selecting an analysis target is displayed on the monitor device, and an existing analysis target is selected with a mouse. (Step 202) The input data card group corresponding to the analysis target selected in step 201 is read from the data card file 30.

Step 203) The template corresponding to the data card read out from the analysis template file 10 is read out. [Knowledge Information Registration Process] Next, a process in which the knowledge registration unit 200 registers knowledge information in the knowledge base 20 will be described.

FIG. 11 is a diagram for explaining a knowledge registration unit according to an embodiment of the present invention. The system shown in the figure inputs a data input section 210 for inputting data to the data card read from the data card file 30 shown in FIGS. , A knowledge registration unit 200 for registering in the knowledge base 20.

FIG. 12 is a flowchart of the knowledge registration processing according to the embodiment of the present invention. (Step 301) First, the card data input unit 210 of the knowledge registration unit 200 executes the data card file 3 in step 107 of FIG. 4 and step 202 of FIG.
Input or change the data used by the solver in the data card read from 0. For example, when inputting a constraint condition card, data such as "heating temperature range: 2000 degrees" is input by an expert or a person who has knowledge of such conditions and various values necessary for executing the solver. Go. The input may be manual input such as a keyboard, or may be input by reading a file.

(Step 302) If knowledge is generated during data input, the process proceeds to step 303, and if not, the process proceeds to step 309. (Step 303) It is determined whether the knowledge input into the data card is text data or other, for example, image data, and if it is text data such as numerical values or conditional statements, the process proceeds to step 304 and other data is used. If there is, the process proceeds to step 305.

Step 304) If the knowledge is text data, the card data input unit 210 inputs it through the keyboard of the input / output unit 110. (Step 305) If it is image data or the like, the card data input unit 210 displays the data from the knowledge data file 50 on the monitor device of the input / output unit 110, and selects the necessary data with the mouse.

(Step 306) The registration unit 200 combines the data card index d with the template T1 ((a (). S.o ())) generated from the analysis template file 10 and analyzed as described above. A template T2 ((a (). S.o ())), d is generated. (Step 307) The registration unit 220 associates the analysis template T2 with the knowledge information input in step 304 or step 305.

(Step 308) The registration unit 230 registers the template T2 and the knowledge information associated in step 307 as a pair in the knowledge base 20. (Step 309) The above processing is performed until the input of all data cards is completed. [Analysis Process] Next, a process of analyzing a certain analysis target using the knowledge stored in the knowledge base 220 will be described.

FIG. 13 is a diagram for explaining the search unit of one embodiment of the present invention. The search unit 300 shown in the figure refers to an analysis instruction unit 310 that inputs an analysis instruction command, an index generation unit 320 that refers to the analysis template file 10 and generates an index that refers to knowledge for analysis, and a knowledge based on the index. Search the base 20,
A knowledge extraction unit 330 that acquires knowledge for analysis, a result output unit 34 that performs analysis processing based on the acquired knowledge and outputs a result
It consists of zero.

FIG. 14 is a flowchart of the knowledge retrieval processing according to the embodiment of the present invention. (Step 401) An instruction command for performing a search is input from the search instruction unit 310. (Step 402) The index generation unit 320 combines the data card index d with the template T1 as an index for searching the knowledge base 20 to obtain the template T2.

Step 403) The template T2 is searched for the analysis template file 10. (Step 404) If the template T2 exists in the analysis template file 10, the process proceeds to step 405, and if not, the process proceeds to step 409. (Step 405) The knowledge extraction unit 330 uses the template T
In step 2, the knowledge base 220 is searched.

(Step 406) As a result of the knowledge extraction unit 330 searching the knowledge base 20, if knowledge information exists, the process proceeds to step 407, and if it does not exist, the process proceeds to step 409. (Step 407) When the knowledge extraction unit 330 can retrieve the knowledge base 220 and acquire the knowledge information, the knowledge is sent to the result output unit 340, and the result output unit 340 outputs the analysis result to the monitor device 350 or the printer. Output to 360.

Step 408) The above processing is repeated until the input of the search target template is completed. (Step 409) In Step 404, the template T2 does not exist in the analysis template file 10,
Alternatively, in step 406, if the knowledge information of interest does not exist in the knowledge base 20, the process of rearranging the template is performed. The shuffling process is shown in FIG.

FIG. 15 is a flowchart of the template rearrangement process at the time of knowledge retrieval according to the embodiment of the present invention. (Step 4090) The “analysis target” is searched from the analysis template file 10. (Step 4091) It is determined whether or not a recombinable “analysis target” exists in the analysis template file 10. If it exists, the process moves to step 4092, and if a recombinable “analysis target” does not exist, Step 4
Move to 095.

Step 4092) Recombinable "analysis target" in the analysis template file 10 is recombined with "analysis target" of the template T2 to form a template T3. (Step 4093) The analysis template file 10 is searched using the rearranged template T3 as an index.

(Step 4094) If the template T3 exists in the analysis template file 10, FIG.
The process moves to step 304 of 4. If it does not exist,
Returning to step 4090, the process of recombining with another template is performed again. (Step 4095) In step 4091, when there is no “analysis target” that can be recombined with the template T2 in the analysis template file 10, another “solver” that can be recombined with the template T2 is searched from the analysis template file 10.

(Step 4096) If the resolvable "solver" exists in the analysis template file 10, the process proceeds to step 4097, and if not, the process proceeds to step 3100. (Step 4097) If the analysis template file 10 has a "solver" that can be recombined with the template T2, it is recombined with the "solver" of the template T2 to form the template T4.

(Step 4098) The analysis template file 10 is searched using the template T4 as an index. (Step 4099) If the template T4 exists in the analysis template file 10, the process proceeds to step 304. If it does not exist, the process proceeds to step 4095.

(Step 4100) In Step 4096, the template T is added to the analysis template file 10.
If there is no “solver” that can be rearranged with 2,
Another "analysis field" is searched from the analysis template file 10. (Step 4101) If an "analysis field" that can be rearranged with the template T2 exists in the analysis template file 10, the process proceeds to step 4102, and if it does not exist, the process is stopped.

(Step 4102) If there is an "analysis field" that can be rearranged with the template, the "analysis field" of the template T2 is rearranged, and the template is designated as template T5. (Step 4103) The analysis template file 10 is searched using the template T5 as an index.

(Step 4104) If the template T5 exists in the analysis template file 10, the process proceeds to step 409, and if not, step 41.
Return to 00. A specific example of each of the above processes will be described below with reference to FIG. [First Specific Example] (a) First, an analysis template is generated.

(1) The analysis template generator 100
From the display screen shown in FIG. 5, "linear analysis" is selected as the analysis field, and "elasticity" is selected as the minor classification (step 103).

(2) The selection information file 40 is searched to determine the solver corresponding to the input analysis field. In the example of FIG. 6, “elasticity” corresponds to “solver A” and “solver B”. In this example, two solvers are applicable, so these two are displayed and the user is prompted to select. Here, the solver selected by the user is "solver A" (step 104).

(3) Next, the analysis target is input from the keyboard. A box as shown in FIG. 7 is displayed on the monitor device, and the target requirement is input. In this example, it is assumed that "print head" and "DN495-002" are input (step 105). (4) The analysis template generation unit 100 has the above (1) to.
An analysis template is generated based on the input of (3). a: Analysis field ... "Linear analysis" s: Solver ... "Solver A" o: Analysis target ... "Print head", "DN495-00"
It becomes 2. Here, it is assumed that an identifier (ID) is given to each item as shown in FIG. Therefore, the generated template T1 is (a (L) .Ao (P, H, DN-2)).

(5) The analysis template generator 100
A data card group corresponding to the template T1 is acquired from the data card file 30. The analysis template generation unit 100 searches the data card file 30 using (a (L) .A) of the template T1 (analysis field and solver) as a key, and the data card for “solver A” “restraint condition card” , "Material characteristic card", "calculation control card" are acquired as a card group (step 107).

(6) The analysis template generator 100
The above template T1 is used as the analysis template file 1
It is stored in 0 (step 108). (B) Data card input processing and data card editing processing Following the analysis template generation processing in (a) above, the knowledge registration unit 2 is added to the card group read in (5) in (a).
The card data input unit 210 of 00 inputs data.

Data is input to the card in the order read in (1) and (5), and the card data input unit 210
Inputs various data used for executing the solver in order to the constraint condition card, the material property card, and the calculation control card, respectively. (C) Process of registering knowledge information in knowledge base (1) Here, it is assumed that the knowledge generated during data input is text data. For example, "Material Properties Card"
When the knowledge of the text data “Material A and Material B are not mixed” occurs at the time of inputting, the data is input from the keyboard of the card data input unit 210.

(3) The registration unit 220 adds a data card index to the template T1 generated in (4) of (a). In this case, the ID of the material property card for which knowledge has occurred is “MT”, so d = MT and the template T2 is “(a (L) .Ao (P, H, DN-2)). MT. It becomes "

(3) The registration unit 220 adds the template T2 “(a (L) .Ao (P) generated in (2) to the knowledge“ material A and material B are not mixed ”generated in (1). , H,
DN-2)). MT "is added and" material A and material B are not mixed "+" (a
(L). A. o (P, H, DN-2)). Register MT "in the knowledge base 20.

(D) Search processing (1) The template T1 "(a (L) .Ao (P, H, DN-2)) is input from the search instruction section 310 of the search section 300 (step 401). (2) The index generator 320 adds the index d = of the data card to the template T1 input in (1).
MT is combined to produce the "(a (L) .Ao (P, H, DN-2)). MT" template T2.

(3) Next, the index generator 320
The analysis template file 10 is searched with the template T2. At this time, since the template corresponding to the above template T2 exists in the analysis template 10, the template T2 is sent to the knowledge extracting unit 330. (4) The knowledge extraction unit 330 has the index generation unit 330.
The knowledge base 20 is searched with the template T2 generated by, and "the material A and the material B are not mixed" is obtained as a search result, and the result output unit 340 is sent.

(5) The result output unit 340 outputs the search result to the monitor device 350. [Second Specific Example] In this specific example, another example of the search processing of (d) of the first specific example will be described. (1) The template T1 "(a (L) .Ao (P, PT, DN-2)) is input from the search instruction unit 310 of the search unit 300 (step 401), that is, the analysis target (P).
T) represents the "platen".

(2) The index generator 320
The index d = MT of the data card is combined with the template T1 input in (1), and “(a (L) .Ao (P, PT, DN-2)). M is obtained.
T ″ template T2 is generated. (3) Next, the index generation unit 320 searches the analysis template file 10 with the template T2. At this time, if the template corresponding to the template T2 does not exist in the analysis template 10. , Analysis target is "print head", and "print head" ID "H"
Searches the analysis template file 10 to determine whether there is an analysis target that can be rearranged.

(4) In this case, it is assumed that there is an analysis target that can be recombined, and template T2 "(a (L) .Ao (P, PT, DN-2)). M.
T ”is recombined with the analysis target“ H ”to obtain a template T3“ (a (L) .Ao (P, H, DN-2)). MT "is generated.

(5) Template T generated in (4)
In step 3, the analysis template file 10 is searched. Here, the template T3 is the analysis template file 10
The template generation unit 320 sends the template T3 to the knowledge extraction unit 330. (6) The knowledge extraction unit 330 has the index generation unit 330.
The knowledge base 20 is searched by the template T3 generated by, and "the material A and the material B are not mixed" is acquired as a search result, and is sent to the result output unit 340. [Third Concrete Example] In this concrete example, still another example of the search processing of (d) of the second concrete example will be described.

(1) The template T1 "(a (L) .Bo (MD, SV)) is input from the search instruction unit 310 of the search unit 300 (step 401).
"D" is "magnetic disk", "SV" is "cymbal",
“B” refers to “solver B”. (2) The index generation unit 320 uses the template T1 input in (1) and the data card index d =
MT is combined to generate a “(a (L) .Bo (MD, SV)). MT” template T2.

(3) Next, the index generator 320
The analysis template file 10 is searched with the template T2. At this time, when the template corresponding to the above template T2 does not exist in the analysis template 10, the analysis target is set to “floppy disk”, the analysis template file 10 is searched by the ID “FDH” of “floppy disk”, and the analysis template file 10 can be rearranged. It is determined whether or not there is another analysis target.

(4) In this case, since there is no analysis target that can be recombined, it is necessary to search for another solver.
Here, from the template T2 of (2), “(a (L) .Bo (MD, SV)). MT”, the solver “B” is replaced with “A”, and the template 4 “(a (L) .Ao (MD, SV)). MT ".

(5) Template T generated in (4)
In step 4, the analysis template file 10 is searched. Here, the template T4 is the analysis template file 10
, The index generation unit 320 sends the template T4 to the knowledge extraction unit 330. (6) The knowledge extraction unit 330 has the index generation unit 330.
The knowledge base 20 is searched by the template T4 generated by, and "the material A and the material B are not mixed" is acquired as a search result, and is sent to the result output unit 340.

In this example, by changing the solver, it is possible to display the knowledge generated when the same problem is searched by another solver. However, only templates with the same data card are available. [Fourth Concrete Example] In this concrete example, still another example of the search process of the third concrete example will be described.

(1) The template T1 "(a (NL) .Bo (MD, SV)) is input from the search instruction unit 310 of the search unit 300 (step 401).
“L” indicates “non-linear”. (2) The index generation unit 320 adds the index d = of the data card to the template T1 input in (1).
The MTs are combined to generate the template T2 "(a (NL) .Bo (MD, SV)). MT".

(3) Next, the index generator 320
The analysis template file 10 is searched with the template T2. At this time, when the template corresponding to the above template T2 does not exist in the analysis template 10, the analysis target is set to “floppy disk”, the analysis template file 10 is searched by the ID “FDH” of “floppy disk”, and the analysis template file 10 can be rearranged. It is determined whether or not there is another analysis target.

(4) In this case, since there is no analysis target that can be rearranged, it is necessary to search for another solver.
It is determined whether the analysis template file 10 has a solver that can be rearranged by the solver “A”. (5) In this case, since there is no resolvable solver, the analysis template file 10 is searched for “nonlinear” = “NL” as another analysis field, and it is determined whether or not a resolvable analysis field exists.

(6) Here, since there is a recombination analysis field “linear” = “L”, here, the template T2 of (2), “(a (NL) .Bo (MD, SV)” is used. ) .MT ”, analysis field“ Nonlinear ”=“ NL ”is changed to“ Linear ”=
The template 5 "(a (L) .Bo (MD, SV)). MT" is generated by recombining with "L".

(7) The template T generated in (6)
In step 5, the analysis template file 10 is searched. Here, the template T5 is the analysis template file 10
The template generation unit 320 sends the template T5 to the knowledge extraction unit 330. (6) The knowledge extraction unit 330 has the index generation unit 330.
The knowledge base 20 is searched by the template T4 generated by, and "the material A and the material B are not mixed" is acquired as a search result, and is sent to the result output unit 340.

The reassignment of the analysis field of the fourth specific example described above is to display the knowledge previously generated while inputting the data card of the template. The analysis field has a hierarchical structure like the analysis target, and it is possible to present the knowledge registered in a situation close to the target knowledge as related knowledge. Similar to the knowledge registration process, the search process of the above specific example searches for knowledge corresponding to the automatically created analysis template, so the user only has to issue a search instruction by pressing the search button or the like. Good. If the knowledge corresponding to the template already exists, the knowledge is displayed, but if the related knowledge does not exist, the elements of the template of the tree structure (hierarchical structure) are rearranged in order from bottom to top. By doing so, it is possible to perform a search from the one having a high similarity.

[0068]

As described above, according to the present invention, rather than adding and managing a fixed index to knowledge, it is possible to register and search by combining information of each element indicating the analysis situation as an index. By doing so, the time required for the index search can be shortened and the search can be performed flexibly.

[Brief description of drawings]

FIG. 1 is a principle configuration diagram of the present invention.

FIG. 2 is a system configuration diagram of an analysis knowledge management device according to an embodiment of the present invention.

FIG. 3 is a diagram illustrating an analysis template generation unit according to an embodiment of the present invention.

FIG. 4 is a flowchart of an analysis template generation process according to an embodiment of the present invention.

FIG. 5 is a diagram showing a display example of an analysis field according to an embodiment of the present invention.

FIG. 6 is a diagram showing an example of an analysis program corresponding to field information according to an embodiment of the present invention.

FIG. 7 is a diagram showing an example of analysis target information input display according to an embodiment of the present invention.

FIG. 8 is a diagram showing an example of combining analysis templates according to an embodiment of the present invention.

FIG. 9 is a diagram showing an example of a data card file according to an embodiment of the present invention.

FIG. 10 is a flowchart for reading an analysis template file according to an embodiment of the present invention.

FIG. 11 is a diagram illustrating a knowledge registration unit according to an embodiment of the present invention.

FIG. 12 is a flowchart of knowledge registration processing according to an embodiment of the present invention.

FIG. 13 is a diagram illustrating an analysis unit according to an embodiment of the present invention.

FIG. 14 is a flowchart of a knowledge search process according to an embodiment of the present invention.

FIG. 15 is a flowchart of template rearrangement processing at the time of knowledge search according to an embodiment of the present invention.

FIG. 16 is a diagram showing an example of combining analysis templates according to an embodiment of the present invention.

FIG. 17 is a diagram for explaining a first method of conventional knowledge management.

FIG. 18 is a diagram for explaining a second conventional method of knowledge management.

[Explanation of symbols]

 10 analysis template file 20 knowledge base 30 data card file 40 selection information file 50 knowledge data file 100 analysis template generation unit 110 input / output unit 200 knowledge registration unit 210 card data input unit 220 registration unit 300 analysis unit 310 search instruction unit 320 index generation Section 330 knowledge extraction section 340 result output section 350 monitor device 360 printer

Claims (5)

[Claims] 1. An analysis knowledge management apparatus for generating an index to be added when searching a knowledge base, storing the knowledge base together with the knowledge information of the knowledge base, and managing the same, and hierarchically configuring related information items, and configuring each configuration. An analysis knowledge management device comprising: an index generation unit that generates and holds an index in which elements are combined in hierarchical order. 2. The index generating means, analysis classification information selecting means for selecting analysis classification information indicating a classification of information to be analyzed, and analysis corresponding to the analysis classification information selected by the analysis classification information selecting means. Analysis program determining means for determining a program, analysis object selecting means for selecting an object to be analyzed, the analysis classification information selecting means, the analysis program determining means, and the index acquired by the analysis object selecting means First combining means for combining constituent elements to generate a first index, and text data for causing the first index generated by the first combining means to execute the analysis program or the text data Second combination for combining the identifiers of data cards consisting of data other than to generate a second index Analysis knowledge management apparatus according to claim 1 consisting of steps. 3. The analysis knowledge management device according to claim 1, further comprising a knowledge base registration unit that associates the knowledge information with the index generated by the index generation unit as a registration key and registers the knowledge information in the knowledge base. 4. The knowledge base according to claim 1 or 2, wherein a search index is created according to the analysis condition, and the knowledge base is searched with the search index when the index is matched with the index created by the index creating means. Analysis knowledge management device. 5. The search means traces the hierarchy one level higher and changes the search index by the higher-order component when the knowledge base cannot be searched by the search index. The analysis knowledge management device according to claim 4, further comprising an index changing unit for searching.