JP2002082984A - Device and method for retrieving data base - Google Patents

Abstract

PROBLEM TO BE SOLVED: To appropriately support the retrieval processing of a data base without previously setting a specified information. SOLUTION: The data base retrieval device is provided with a data base 1 holding data being the object of retrieval, a similar example data base 5 making a similar example into the data base and holding it, a retrieval condition input part 2 inputting a retrieval condition, a retrieval condition developing part 6 retrieving the similar example data base 5 based on the retrieval condition and outputting a developed retrieval condition obtained by developing the retrieval condition, a data base retrieval part 7 retrieving the data base 1 based on the developed retrieval condition, a retrieval result output part 8 outputting a retrieval result, a retrieval operation holding part 3 holding the retrieval operation of a user by including the retrieval condition, a similar example analysis part 11 analyzing the similar example based on selection information which is just before and the retrieval operation held by the retrieval operation holding part 3 and a similar example data base update part 12 updating the content of the similar example data base 5 based on an analysis result by the similar example analysis part 11.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a database search apparatus and a database search method for searching a database, and more particularly to other similar search conditions related to a user's search request in order to support a user searching a wide range of related items. The present invention relates to a database search device having a search support function capable of presenting search conditions for narrowing search targets and a method thereof.

[0002]

2. Description of the Related Art Conventionally, there is a database search device for searching for desired data from a database. For example, FIG. 12 is a block diagram showing a configuration of a conventional database search device (see Japanese Patent Application Laid-Open No. H6-176072). This database search device is a search device that uses an index that uses word similarity. In FIG. 12, this database search device stores a database 27 that stores data expressed by pairs of attributes and attribute values.
And a first search unit 25 for roughly narrowing down the data amount from the data in the database 27 using the first index before performing the similarity search, and a second search unit 25 A second search unit that performs a similarity search based on the index.

When a search condition consisting of an attribute and an attribute value is input via the input device 21 and the input / output control unit 23, the inference processing unit 24 sends the first search unit 25 the same The data having the attribute is retrieved from the database 27. The first search unit 25 includes the first index generation unit 2
8 is instructed to generate a first index. The first index generation unit 28 determines the range of the similarity between the searched attribute values, calculates the similarity between the attribute values of the respective attributes together with the importance of the attributes, and performs the first search on the calculation result. Output to the unit 25. The first search unit 25 holds, as a first index, the degree of similarity between the attribute values of the search results. After that, the inference processing unit 24 causes the second search unit 26 to perform a similarity search. The second search unit 26 compares the similarity between the input search condition and each attribute value, and outputs the similarity to the inference processing unit 24 as a similar search result when the similarity is within the range of the similarity. The inference processing unit 24 outputs the similarity search result to the output device 22 via the input / output control unit 23.

[0004] On the other hand, some dialogue apparatuses that achieve a task by performing a dialogue with a user can provide optimal operation guidance suitable for the user's operation proficiency. For example, FIG. 13 is a block diagram showing a configuration of a conventional interactive device to which an operation guidance output method of a search operation is applied (see Japanese Patent Application Laid-Open No. Hei 8-44520). In FIG. 13, after receiving a notification of an operable state from the dialog management unit 44, the dialog device receives input information from the input control unit 34 in response to a user's input operation using the voice input device 31 or the touch panel 32. Based on the operation time measurement unit 36 that measures the time until the input (operation time), and the operation time and the operation time threshold registered in the threshold registration unit 38 in advance, the proficiency level of the user who is using For example, "professional"
It has a proficiency determination unit 40 that determines proficiency levels classified as “beginner” and “general person” and stores the determination result in a proficiency storage unit 41.

[0005] The operation guidance output process by the dialog management unit 44 is based on the proficiency stored in the proficiency storage unit 41 and extracts the operation guidance corresponding to the proficiency from the operation guidance database 43 and displays it on the display 2.
9 or the speaker 30.

[0006] The proficiency determining section 40 not only determines the proficiency of the user based on the operation time, but also determines the proficiency of the user based on the input conditions. In the exemplary procedure registration unit 39, the original input procedure of the search condition for searching the currently targeted database is registered in advance. Here, the proficiency determination unit 40 determines the difference between the input search condition and the model procedure using the registered model procedure as a correct answer, and determines the user's skill based on the magnitude of the difference.

[0007]

By the way, in the above-mentioned conventional database search apparatus, the input search condition is determined based on the attribute importance and similarity of a pair of the attribute and the attribute value of the search condition registered in advance. Based on the half value of, an appropriate similarity range is set, the result is output, or the index is updated, but the similarity to the attribute of the search condition for setting the similarity range is registered in advance. If the information is not registered, there is a problem that the similarity range cannot be determined, that is, a similar case cannot be searched.

Further, in the above-described conventional interactive device, the skill level is determined based on the difference between the input search condition and the model procedure registered in advance, and the skill level is determined in advance according to the skill level. Although the guidance contents are output by reference, there is a problem that various data to be used for the guidance must be created in advance regardless of the search tendency, such as inputting search conditions for the search target database. was there.

[0009] The present invention has been made in view of the above,
An object of the present invention is to provide a database search device and a method thereof that can appropriately support a database search process without setting specific information in advance.

[0010]

In order to achieve the above object, a database search apparatus according to the present invention converts a database holding search target data and a similar case of the search target data stored in the database into a database. A similar case database to be held, input means for inputting search conditions of the database, and a search for the similar case database based on search conditions input to the input means, and an expanded search condition obtained by expanding the search condition. Search condition expanding means for outputting, search means for searching the database based on the expanded search condition expanded by the search condition expanding means, output means for outputting a search result by the search means, and input means The input search condition and the selection information selected by the user for the search result output by the output unit are included. A search operation holding unit that holds a user's search operation related to the database search; a similar case analysis unit that analyzes a similar case based on the selection information and the search operation held in the search operation holding unit; Updating means for updating the contents of the similar case database based on the analysis result by the similar case analyzing means.

According to the present invention, the search condition expanding means includes:
The similar case database is searched based on the search condition input to the input unit, and an expanded search condition obtained by expanding the search condition is output. The search unit also searches the expanded search condition expanded by the search condition expanding unit. And search the database for
An output unit outputs a search result. On the other hand, a search operation holding unit holds a user's search operation related to a search of the database including a search condition input to the input unit and selection information selected by a user with respect to a search result output by the output unit. A similar case analysis unit that analyzes a similar case based on the selection information and the search operation held in the search operation holding unit, and an update unit uses the similar case analysis unit based on an analysis result obtained by the similar case analysis unit. By updating the content of the similar case database, the user can appropriately search for data of conditions similar to the search conditions based on the search history so far, for example, such as the specification of parent parts and small parts. In addition, even if a combination of search conditions cannot be set in advance, it is possible to search for similar cases.

According to another aspect of the present invention, there is provided a database search device which stores a search object data, input means for inputting search conditions of the database, and a next similarity search corresponding to the search conditions input from the input means. A similarity search procedure database for converting conditions into a database and holding the similarity search procedure as a similarity search procedure, and searching the similarity search procedure database based on the search conditions input to the input means, and searching for a similarity search condition corresponding to the search conditions , A search unit that searches the database based on the next similar search condition output by the guidance unit, an output unit that outputs a search result by the search unit, and an input that is input to the input unit. The data including search conditions and selection information selected by the user for the search results output by the output unit. Search operation holding means for holding a user's search operation related to database search, similarity search procedure analysis means for analyzing a similarity search procedure based on the selection information and the search operation held in the search operation holding means, Updating means for updating the contents of the similarity search procedure database based on the analysis result by the similarity search procedure analysis means.

According to this invention, the guidance means searches the similarity search procedure database based on the search condition input to the input means, and outputs the next similarity search condition corresponding to the search condition. Search the database based on the next similar search condition output by the guidance means,
An output unit outputs a search result. On the other hand, a search operation holding unit holds a user's search operation related to a search of the database including a search condition input to the input unit and selection information selected by a user with respect to a search result output by the output unit. A similarity search procedure analyzing means for analyzing a similarity search procedure based on the selection information and the search operation held in the search operation holding means, and an updating means for analyzing an analysis result by the similarity search procedure analysis means. At this time, the contents of the similarity search procedure database are updated.
If the current search result does not match the request, the guidance means determines what to do next search operation, what to input as search conditions, etc. in the search history and preferences of many users so far. It is presented as guidance along with it.

[0014] A database search apparatus according to the next invention is a database for holding search target data, input means for inputting search conditions of the database, and database of similar cases of search target data stored in the database. A similar case database to be held, a database of the next similar search conditions corresponding to the search condition input from the input means, and a similar search procedure database to be stored as a similar search procedure; and a search condition input to the input means. A search condition expansion unit for searching the similar case database based on the search condition and outputting an expanded search condition obtained by expanding the search condition; and searching the similarity search procedure database based on the search condition input to the input unit. Guidance means for outputting a next similar search condition corresponding to the search condition; Search means for searching the database based on the expansion search condition output by the case expansion means or the next similar search condition output by the guidance means; output means for outputting a search result by the search means; and input means Search operation holding means for holding a user's search operation related to the search of the database including search conditions input to the user and selection information selected by the user with respect to the search result output by the output means; and Analysis means for analyzing similar cases and similar search procedures based on the search operation held in search operation holding means, and similar case update for updating the contents of the similar case database based on the analysis result by the analysis means Means for updating a similarity search procedure database based on an analysis result by the analysis means Characterized by comprising a stage.

According to the present invention, the search condition expanding means includes:
The similar case database is searched based on the search condition input to the input means, and an expanded search condition obtained by expanding the search condition is output. The guidance means performs similarity search based on the search condition input to the input means. Search the search procedure database,
The next similar search condition corresponding to the search condition is output, and the search unit searches the database based on the expanded search condition output by the search condition expansion unit or the next similar search condition output by the guidance unit. And output means for outputting a search result by the search means. On the other hand, a search operation holding unit holds a user's search operation related to a search of the database including a search condition input to the input unit and selection information selected by a user with respect to a search result output by the output unit. Analyzing means for analyzing a similar case and a similar search procedure based on the selection information and the search operation held in the search operation holding means, and a similar case updating means based on an analysis result by the analysis means. Then, the content of the similar case database is updated, and the similarity search procedure changing means updates the content of the similarity search procedure database based on the analysis result by the analysis means.

In the database search device according to the next invention, in the above invention, a search condition for data stored in the database is provided to the input means, and an editing process is performed based on a search result output to the output means. And an analyzing unit that obtains an editing result by the editing unit and adds the editing result to analyze the similar case or the similar search procedure.

According to this invention, the editing means gives the input means a search condition for the data stored in the database, performs an editing process based on the search result output to the output means, and The means acquires an editing result by the editing means, and analyzes the similar case or the similar search procedure by adding the editing result.

A database search method according to the next invention includes an input step of inputting search conditions for a database holding search target data, a first storage step of storing search conditions input in the input step, and the search step. A search condition expanding step of searching a similar case database holding similar cases of search target data stored in the database based on a condition, and outputting an expanded search condition obtained by expanding the search condition; A search step of searching the database based on the expanded search condition expanded by the step, an output step of outputting a search result of the search step, and a selection selected by a user with respect to the search result output by the output step A second holding step of holding information;
An analyzing step of analyzing a similar case based on the search condition held by the first holding step and the selection information held by the second holding step; and the similar case database based on an analysis result by the analyzing step. And an updating step of updating the content of

According to the present invention, in the input step, a search condition for the database holding the data to be searched is input, and in the first holding step, the search condition input in the input step is held, and the search condition expanding step is performed. Based on the search condition, a similar case database that holds similar cases of the search target data stored in the database is searched, and an expanded search condition obtained by expanding the search condition is output. The database is searched based on the expanded search condition expanded in the search condition expanding step, the search result in the search step is output in the output step, and the search result output in the output step is output in the second holding step. In response to the selection information selected by the user, the analysis step stores the selection information selected by the user. The similar case is analyzed based on the condition and the selection information held by the second holding step, and the content of the similar case database is updated by the updating step based on the analysis result of the analyzing step. .

A database search method according to the next invention includes an input step of inputting search conditions for a database holding search target data, a first storage step of storing search conditions input in the input step, and a similarity search. A guidance step of searching for and outputting the next similar search condition corresponding to the search condition from the similarity search procedure database holding the procedure, and the database based on the next similar search condition searched by the guidance step. A search step of searching for a search result; an output step of outputting a search result of the search step; a second holding step of holding selection information selected by a user with respect to the search result output by the output step; A similarity search procedure based on the search condition held in the step and the selection information held in the second holding step. An analysis step of analyzing, characterized in that it comprises an update step of updating the contents of the analogous case database based on the analysis result by the analysis step.

According to the present invention, in the input step, search conditions for the database holding the data to be searched are input, in the first holding step, the search conditions input in the input step are held, and in the guidance step, From the similarity search procedure database holding the similarity search procedure, a next similarity search condition corresponding to the search condition is searched and output, and a search step is performed based on the next similarity search condition searched by the guidance step. Searching the database, outputting the search result of the search step by an output step, holding the selection information selected by the user with respect to the search result output by the output step by a second holding step, Thus, the search condition held in the first holding step and the search condition held in the second holding step The-option information by analyzing the similarity search procedure on the basis of the update process, and to update the contents of the analogous case database based on the analysis result by the analysis step.

A database search method according to the next invention includes an input step of inputting search conditions for a database holding search target data, a first storage step of storing search conditions input in the input step, and the search step. A search condition expanding step of searching a similar case database holding similar cases of the search target data stored in the database based on the conditions, and outputting an expanded search condition obtained by expanding the search condition; A guidance step of searching for and outputting the next similar search condition corresponding to the search condition from the held similar search procedure database, and an expansion search condition output by the search condition expansion step or the guidance according to a user's selection Search for searching the database based on the next similar search condition searched by the process An output step of outputting a search result of the search step; a second holding step of holding selection information selected by a user with respect to the search result output by the output step; An analysis step of analyzing similar cases and similar search procedures based on the retrieved search conditions and the selection information held by the second holding step, and the contents of the similar case database and the Updating the contents of the similarity search procedure database.

According to the present invention, in the input step, a search condition for the database holding the search target data is input, and in the first holding step, the search condition input in the input step is stored, and the search condition expanding step is performed. Based on the search condition, a similar case database holding similar cases of the search target data stored in the database is searched, and an expanded search condition obtained by expanding the search condition is output. Searching and outputting the next similar search condition corresponding to the search condition from the similarity search procedure database in which the search procedure is held, and performing the search by the user's selection in the search step, the expansion search output by the search condition expansion step The database is based on conditions or the following similar search conditions searched in the guidance process. Search for the search result, outputting the search result of the search step by the output step, holding the selection information selected by the user for the search result output by the output step by the second holding step, by the analysis step, A similar case and a similar search procedure are analyzed based on the search condition held in the first holding step and the selection information held in the second holding step, and an update step is performed based on an analysis result of the analysis step. Then, the contents of the similar case database and the contents of the similar search procedure database are updated.

[0024] In a database search method according to the next invention, in the above invention, the input step is performed by an editing means for performing an edit process based on a search result in the output step. Is input, and the analysis step performs analysis by adding an editing result edited by the editing means based on the search result in the output step.

According to this invention, the input step inputs a search condition relating to data stored in the database from editing means for performing an editing process based on a search result in the output step, and the analysis step includes: The editing means performs analysis by adding an editing result edited based on a search result in the output step.

[0026]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

Embodiment 1 FIG. 1 is a block diagram showing a configuration of the database search device according to the first embodiment of the present invention. In FIG. 1, the database search device includes a search condition input unit 2 for inputting search conditions for searching for information stored in a database 1, and a search operation storage as search operation storage data for an operation related to a search performed by a user. A search operation holding processing unit 4 for performing processing to be held by the unit 3, a search condition expanding unit 6 that refers to the similar case database 5 and outputs expanded search conditions obtained by expanding the input search conditions, and a search condition expanding unit 6. In addition, a database search unit 7 for searching the database 1, a search result output unit 8 for outputting a search result by the database search unit 7, and a user evaluation result (selection result) for the search result are searched as search result evaluation data. The search result evaluation unit 10 and the search result evaluation data storage unit 9 hold the result evaluation data storage unit 9 and the search operation storage unit 3, respectively. A search result display data and retrieval operations search operation data held in the holding section 3 for analyzing the original in similar cases the similar case analyzer 11 are,
A similar case database updating unit 12 for updating the storage content of the similar case database 5 based on the analysis result.

Here, a search process by the database search device shown in FIG. 1 will be described. First, a search condition is input to the search condition input unit 2 by a user. The search condition is held in the search operation holding unit 3 by the search operation holding processing unit 4 and output to the search condition development unit 6. Thereafter, the search condition developing unit 6 searches the similar case database 5 based on the search conditions, and expands the search conditions into a plurality of search conditions including similar cases, that is, expanded search conditions. After that, the database search unit 7
The database 1 is searched based on the expanded search condition including the search condition input from the search condition input unit 2. The search result by the database search unit 7 is output to the search result output unit 8.

Thereafter, the user selects one or more of the search results output from the search result output unit 8 that best match the search conditions. The selection result is input to the search result evaluation unit 10 and is stored as search result evaluation data in the search result evaluation data storage unit 9 and is also stored as search operation storage data in the search operation storage unit 3.

Thereafter, the similar case analysis unit 11 analyzes similar cases based on search operation storage data and search result evaluation data obtained by a user operation. The similar case database update unit 12 updates the contents of the similar case database 5 based on the analysis result of the similar case analysis unit 11.

Here, a specific similarity search process using the similar case database 5 will be described. In this case, it is assumed that the parts list necessary for the repair work of the elevator is stored together with the specification information in the similar case database 5 as a database, and is necessary for the repair work for the new product of the specification to be released in the future. A case will be described in which a component list for renovation work similar to past specifications is searched to create a simple component list. To be stored in the similar case database 5 as a database means that the similar case database updating unit 12 has updated the contents of the similar case database 5 by a plurality of search results.

First, as shown in FIG. 2, the user sets the item “construction name” to “PT7” and sets the item “control after modification” to “PT7”.
Is entered as search condition data D1. On the other hand, in the similar case database 5, as shown in data D2 of FIG. 2, there are items such as "construction name", "specification item name", "specification name 1", "specification name 2", and "frequency". The corresponding data is registered as a database. In the data D2, in the “repair control method” of the “specification items”, “ACE2” is replaced with “ACE1” and “ACEE”
0 ". Therefore, the search condition expanding unit 6 searches the data D2 in the similar case database 5 based on the data D1 of the search condition and expands the search condition.
The content of the "control system after renovation" is "ACEE2", "AC
EE1 "and" ACEE0 ".

When the number of expanded search conditions increases, the “frequency” of the data D2 in the similar case database 5 is increased.
Is set based on the threshold value, and the number of expansion search conditions to be expanded is limited. For example, if the threshold value is set to “50” for the data D2 shown in FIG. 2, “ACEE2” and “ACEE2” whose “frequency” is “100”
"1" is expanded as an expansion search condition, and "ACEEE0" is not expanded. In this case, as in the case of the data D3 shown in FIG.
Only “2” and “ACEE1” are output as expansion search conditions. Here, the database search unit 7 searches the database 1 based on the expanded search condition, and outputs a search result to the search result output unit 8.

Next, a specific similarity analysis process performed by the similar case analysis unit 11 and a specific change process performed by the similar case database update unit 12 will be described.
First, the user repeats the search operation as shown in FIG. 2 several times, and evaluates the selection result using the search result evaluation unit 10 each time. That is, the user:
With respect to a certain search result, one or more data of the search result which seems to match the input search condition is selected. The evaluation result by the selection is output to the search result evaluation data holding unit 9 and also to the search operation holding unit 3.

For example, as shown in FIG. 3, when the search condition data D11 is entered with "Construction name" as "PT7" and "Repair operation method" as "2C2BC",
The search result output unit 8 outputs some search result data D12 as search results. The user evaluates and selects the two data D12 among the data D12 of some search results as data relatively close to the data requested by the user. This selected data D12
Is held in the search result evaluation data holding unit 9 as search result evaluation data, and is held as search operation holding data in the search operation holding unit 3.

The similar case analysis unit 11 analyzes similar cases based on each search result evaluation data and each search operation holding data. The analysis of the similar case is performed on all the search result evaluation data. For example, when the data D12 which is the two data shown in FIG. 3 is selected, the user:
Although the values of the “specification items” are different, they are determined to be generally similar because they were selected as being in line with the user's request. Therefore, the similar case analysis unit 11 compares each of the “specification items” of the data D12, which is the two data stored in the database 1, and if the two “specification items” are the same, sets them as extraction targets. First, extract different "specification items".

In the case of FIG. 3, the values of the "control system after repair" and "door closing system after repair" among the specification items are different. Therefore,
The similar case analysis unit 11 causes the similar case database 5 to retain the "specification item" in the similar case database 5 so that even if the "specification item" is different from the search result, the similar case analysis unit 11 can include the "specification item" instead of excluding it from the search result. To The similar case database 5 is held by the similar case database updating unit 12 based on the analysis result by the similar case analyzing unit 11.

By repeating the similar case analysis by the similar case analysis unit 11 and the updating of the similar case database by the similar case database updating unit 12,
The internal capacity of the similar case database 5 increases. In this case, the similar case database update unit 12 updates the similar case database by organizing the similar case database 5 into a format that is easy to use, instead of simply adding the case to the similar case database 5. . Specifically, as shown in data D13 in FIG. 3, the item “frequency” is managed, and when the items to be stored in the similar case database 5 are the same, the content of the search case itself is added. Instead, the total is calculated based on the "frequency" item.

According to the first embodiment, the more the past search results are accumulated, the more the contents of the similar case database 5 become richer, and this is reflected in the search results of the subsequent searches. Since the similar case database 5 is used, even if there is no data of a specification that completely matches the search condition, a similar specification can be automatically added as a search condition to obtain data of a similar specification. And the convenience is improved. Also, even if there is data of a specification that completely matches the search condition, other similar data can be obtained, and data of the matching specification can be compared with data of a similar specification, Based on the result of the comparison and examination, appropriate data desired by the user can be retrieved and output.

In addition, the appropriate data can be output as a search result, the size (capacity) of the similar case database 5 can be reduced, and the consumption of computer resources can be suppressed.

Embodiment 2 Next, a second embodiment of the present invention will be described. In the first embodiment described above,
Although the appropriate data can be automatically searched for using the similar case database 5, the second embodiment uses the similar search procedure database 13 instead of the similar case database 5 to obtain appropriate data. Is automatically searched.

FIG. 4 is a block diagram showing a configuration of the database search device according to the second embodiment of the present invention. In FIG. 4, the database search apparatus includes a search condition input unit 2 for inputting search conditions for searching for information stored in a database 1, and a search operation storage as search operation storage data in which an operation relating to a search performed by a user is performed. A search operation holding processing unit 4 for performing processing to be held by the unit 3 and a search condition guidance unit 14 for referring to the similar search procedure database 13 and presenting appropriate search conditions based on the similar operation procedure.
A database search unit 7 for searching the database 1 based on a search condition selected by the user and presented by the search condition guidance unit 14, a search result output unit 8 for outputting a search result by the database search unit 7, The search result evaluation unit 10 and the search result evaluation data holding unit 9 hold the user's evaluation result (selection result) for the search result as search result evaluation data in the search result evaluation data holding unit 9 and the search operation holding unit 3, respectively. A search procedure analysis unit 15 for analyzing a similar search procedure based on the retrieved search result display data and the search operation data held in the search operation holding unit 3, and a similar search procedure database 13 is stored based on the analysis result. Similarity search procedure database updating unit 16 for updating contents
And

Here, search processing by the database search device shown in FIG. 4 will be described. First, a search condition is input to the search condition input unit 2 by a user. The search condition is held in the search operation holding unit 3 by the search operation holding processing unit 4 and output to the search condition guidance unit 14. Thereafter, the search condition guidance unit 14
Searches the similar search procedure database 13 based on the input search conditions, and
When the search is further advanced, that is, when narrowing down is performed, a recommended search condition is output as a guidance message. Thereafter, the database search unit 7 searches the database 1 based on the search condition selected by the search condition guidance unit 14. Database search unit 7
Is output to the search result output unit 8.

Thereafter, the user selects one or more of the search results output to the search result output unit 8 that best match the search conditions. The selection result is input to the search result evaluation unit 10 and is stored as search result evaluation data in the search result evaluation data storage unit 9 and is also stored as search operation storage data in the search operation storage unit 3.

After that, the search procedure analyzing unit 15 analyzes the similar search procedure based on the search operation storage data and the search result evaluation data obtained by the user's operation. The similarity search procedure database updating unit 16 includes a search procedure analysis unit 15
The content of the similarity search procedure database 13 is updated based on the result of the analysis.

Here, a specific similarity search procedure search process using the similarity search procedure database 13 will be described. FIG. 5 is a diagram illustrating an example of a format of search operation holding data held in the search operation holding unit 3. The search operation holding processing unit 4 outputs the input search condition and the like for each search operation. In FIG. 5, as items to be recorded,
The specification item names such as "construction name" and "repair control method" are arranged, and data input as search conditions are recorded as one line for each search operation. That is, FIG.
In the first search operation, "construction name" is searched for as "PT7", and the second search operation is performed for "construction name" as "PT7".
Then, the "repair control method" is searched as "ACE2", and the third search operation is "PT7" for "construction name", "ACEE2" for "repair control method", and "repair operation method".
Indicates the search operation holding data when the search is performed as “2C2BC”.

FIG. 6 is a diagram showing an example of the similarity search procedure database 13 created by the search procedure analyzer 15 and the similarity search procedure database updater 16 from the search operation holding data shown in FIG. FIG. 6 illustrates the first to third search operation procedures from left to right on the drawing. For example, if the “construction name” is searched for as “PT7” for the first time, in the past search operation procedure, for the second time, “construction name” and “after-repair control method” or “construction name” and “after-repair operation” It can be seen that either of the search conditions of “method” was input. In addition, by incorporating information of “number of cases” in each search operation procedure, it is possible to indicate which search condition is most frequently input in the past search operation procedures. For example, in FIG. 6, when the first “construction name” is searched for as “PT7”, the number of times of the second “construction name” and the “control after modification” search operation is “20”, This means that the number of times of the second search operation of “construction name” and “operation method after repair” is “4”.

FIG. 7 is a diagram for explaining the guidance processing by the search condition guidance unit 14 using the similar search procedure database 13. Here, it is assumed that a similar search procedure database having the same contents as the similar search procedure database shown in FIG. 6 is stored. Here, as shown in FIG. 7, when the search condition data D21 in which the “construction name” is “PT7” and the “repair control method” is “ACEE2” is input, the search condition guidance unit 14 From the similar search procedure database 13, the location corresponding to the currently input search condition, that is, “construction name” is “PT7”,
Search for a place where the procedure in which the “repair control method” was input as a condition remains. If this location is found, the procedure located on the right side of this location, that is, in the past search operation procedure, if the current search condition is not met, what search condition will be used as the next search operation procedure To see if you were typing

In the example of FIG. 7, the search conditions of "construction name" and "repair control method" are linked next to "reconstruction name", "repair control method", and "repair operation". Method ". If the next similar search procedure is found, the search condition guidance unit 14 may, for example, enter "Construction name, post-repair control method, post-repair operation method" as the next refinement condition (D22).
For example, the message is changed to a message that is easy for the searcher to understand, guidance output is performed, and the user inputs new search conditions according to the presentation.

According to the second embodiment, the current search result is insufficiently narrowed down. However, for a user who cannot think of the next search condition to be input due to the narrowing down, the past searcher is By using the similar search procedure database to guide what kind of search conditions were input in the same situation, it is possible to reduce trial and error related to narrowing down and to improve the efficiency of search operations. Can be.

The search result evaluation data used in the similar case arrangement shown in the first embodiment may be added to the end of the search operation holding data shown in FIG. In this case, it is possible to determine information as to which search operation has selected data matching the request of the searcher. The search result evaluation data can be incorporated into the similarity search procedure database 13 via the search procedure analyzer 15 and the similarity search procedure database updater 16.

Embodiment 3 FIG. Next, a third embodiment of the present invention will be described. In the third embodiment, the similar case database 5 and the similar search procedure database 13 in the first embodiment are used together.

FIG. 8 is a block diagram showing the configuration of the database search device according to the third embodiment of the present invention. 8, this database search device has a configuration in which a search condition guidance unit 14 of the database search device shown in FIG. 4 is further provided after the search condition development unit 6 of the database search device shown in FIG. With the additional configuration of the search condition guidance unit 14, the corresponding similar search procedure database 13 and similar search procedure database update unit 16 are also added, and the search case analysis unit 11
A search evaluation result analysis unit 17 that integrates a search procedure analysis unit 15 with the search evaluation result analysis unit 15 is provided.

Here, a search process by the database search device shown in FIG. 8 will be described. First, a search condition is input to the search condition input unit 2 by a user. The search condition is held in the search operation holding unit 3 by the search operation holding processing unit 4 and output to the search condition development unit 6. Thereafter, the search condition developing unit 6 searches the similar case database 5 based on the search conditions, and expands the search conditions into a plurality of search conditions including similar cases, that is, expanded search conditions. The database search unit 7 searches the database 1 based on the expanded search condition including the search condition input from the search condition input unit 2. The search result by the database search unit 7 is output to the search result output unit 8.

On the other hand, as in the second embodiment, the search condition guidance unit 14 searches the similar search procedure database 13 based on the input search conditions, and further searches according to the currently input search conditions. When proceeding, that is, when narrowing down, a recommended search condition is output as a guidance message. The database search unit 7 searches the database 1 based on the search conditions, and the search result is output to the search result output unit 8. It should be noted that the search of the database 1 based on the similar search condition using the similar case database 5 is performed or the database 1 based on the similar search condition using the similar search procedure database 13 is performed.
Can be arbitrarily selected by the user. Here, the user can select according to the current number of search results.

Thereafter, the user selects, from the search results output to the search result output unit 8, one or more items that best match the search conditions. The selection result is input to the search result evaluation unit 10 and is stored as search result evaluation data in the search result evaluation data storage unit 9 and is also stored as search operation storage data in the search operation storage unit 3.

Thereafter, the search evaluation result analysis unit 17 analyzes similar cases and similar search procedures based on search operation storage data and search result evaluation data obtained by a user operation. The similar case database update unit 12 updates the contents of the similar case database 5 based on the analysis result of the search evaluation result analysis unit 17, and the similar search procedure database update unit 16 updates the analysis result of the search evaluation result analysis unit 17 The contents of the similarity search procedure database 13 are updated based on the original.

FIG. 9 is a diagram showing an example of search operation holding data used by the search evaluation result analysis unit 17. The search evaluation result analysis unit 17 implements the processing functions of the similar case analysis unit 11 and the search procedure analysis unit 15. In order to realize these processing functions by one search evaluation result analysis unit 17, The amount of information that can be processed increases, and it is possible to change to database information that more strongly expresses the past influence of the search operation. The difference between the search operation holding data shown in FIG. 9 and the search operation holding data shown in the second embodiment is that the number of developments to similar conditions (data D31 to D33) is recorded for each specification item name. It is that you are.

FIG. 10 is a diagram showing an example of the similarity search procedure database 13. In the similarity search procedure database 13 according to the second embodiment, the index is used to select the next operation item based on the number of cases, but in the similarity search procedure database 13 according to the third embodiment, Introducing points that take into account the number of deployments, and using the number of points as guidance indicators. For example, for guidance items that are likely to be developed in similar cases, set a negative point in the guidance so that they are not recommended as much as possible, or store the normal number of cases and the number of similar developed cases side by side. In addition, at the time of guidance, the best one from each viewpoint is displayed, so that the searcher can select the information.

According to the third embodiment, the first embodiment using the similar case database 5 and the second embodiment using the similar search procedure database 13 are realized by one apparatus. In this case, since the information of the search of the similar case and the search of the similar search procedure can be used in association with each other, more appropriate similar case database 5 and similar search procedure database 13 can be further formed, and Desired and appropriate data search can be reliably and easily realized. In addition, since the search for similar search conditions using the similar case database 5 and the search for similar search conditions using the similar search procedure database 13 can be selectively performed, an appropriate search according to the user's request can be performed. Processing can be performed. That is, when the similar case database 5 is used, information that matches the search conditions can be searched without omission. When the similar search procedure database 13 is used, a large number of pieces of information can be used as search conditions. It is possible to perform a search for appropriately narrowing down matched information.

Embodiment 4 Next, a fourth embodiment of the present invention will be described. In the fourth embodiment, a data editing unit 18 for editing data is further provided in the configuration of the third embodiment, and the data editing result is stored in the similar case database 5.
And the similarity search procedure database 13.

FIG. 11 is a diagram showing a configuration of a database search device according to the fourth embodiment of the present invention. In FIG. 11, a data editing unit 18 performs data editing processing with reference to a search result of the database 1. Therefore, the data editing unit 18 can input the search condition for the data being edited to the search condition input unit 2. The data editing result is held in the newly created data holding unit 19 as newly created data. The search evaluation result analysis unit 20 corresponds to the search evaluation result analysis unit 17 and further obtains newly created data, and analyzes the search evaluation result including the newly created data. Other configurations are the same as those of the third embodiment, and the same components are denoted by the same reference numerals.

The data editing unit 18 performs data editing processing with reference to the search results of the database 1. here,
The search condition input unit 2 may acquire a data area having the same name as the specification item of the database 1 from the data editing unit 18 and set this as a search condition. The data editing unit 18 appropriately edits and creates data, and the newly created data is held in the newly created data holding unit 19. The search evaluation result analysis unit 20 performs an analysis process in consideration of newly created data in addition to the search operation holding data and the search result evaluation data. The analysis result is reflected on the similar case database 5 via the similar case database update unit 12 and the similar search procedure database 13 via the similar search procedure database update unit 16.

According to the fourth embodiment, a database search can be executed during data editing, and the time and effort required for inputting search conditions can be omitted. Since the analysis process is performed using the newly created data, it is possible to save the user the time required for the selection evaluation.

[0065]

As described above, according to the present invention, the search condition expanding means searches the similar case database based on the search condition input to the input means, and expands and develops the search condition. The condition is output, the search means searches the database based on the expanded search condition expanded by the search condition expanding means, and the output means outputs the search result. On the other hand, a search operation holding unit holds a user's search operation related to a search of the database including a search condition input to the input unit and selection information selected by a user with respect to a search result output by the output unit. A similar case analysis unit that analyzes a similar case based on the selection information and the search operation held in the search operation holding unit, and an update unit uses the similar case analysis unit based on an analysis result obtained by the similar case analysis unit. By updating the content of the similar case database, the user can appropriately search for data having conditions similar to the search conditions based on the search history so far. For example,
Even in the case of a combination of search conditions that cannot be set in advance, such as the specification of a parent part and a small part, it is possible to search for similar cases. This has the effect that the search can be performed appropriately and without omission.

According to the next invention, the guidance means is:
A similarity search procedure database is searched based on the search condition input to the input means, and a next similarity search condition corresponding to the search condition is output, and the search means outputs the next similarity search condition output by the guidance means. , And the output means outputs a search result. On the other hand, a search operation holding unit holds a user's search operation related to a search of the database including a search condition input to the input unit and selection information selected by a user with respect to a search result output by the output unit. A similarity search procedure analyzing means for analyzing a similarity search procedure based on the selection information and the search operation held in the search operation holding means, and an updating means for analyzing an analysis result by the similarity search procedure analysis means. At this time, the contents of the similarity search procedure database are updated. If the current search result does not match the request, the guidance means will ask the next search operation,
Since what to input as search conditions is presented as guidance in line with the search history and preferences of many users so far, it is appropriate to narrow down the search for desired information that matches the search conditions. The effect that it can be performed is produced.

According to the next invention, the search condition expanding means searches the similar case database based on the search condition input to the input means, and outputs the expanded search condition obtained by expanding the search condition. Search the similarity search procedure database based on the search condition input to the input means, and output the next similarity search condition corresponding to the search condition,
Search means searches the database based on the expanded search condition output by the search condition expanding means or the next similar search condition output by the guidance means, and the output means outputs the search result by the search means. On the other hand, a search operation holding unit holds a user's search operation related to a search of the database including a search condition input to the input unit and selection information selected by a user with respect to a search result output by the output unit. Analyzing means for analyzing a similar case and a similar search procedure based on the selection information and the search operation held in the search operation holding means, and a similar case updating means based on an analysis result by the analysis means. Since the content of the similar case database is updated, and the similarity search procedure changing means updates the content of the similarity search procedure database based on the analysis result by the analysis means, information matching the search condition is obtained. And the process of narrowing down and searching for information that matches the search conditions from a large number of information can be selectively performed. Selection of, the user can arbitrarily select in accordance with the number of the search results to date, further, there is an effect that it is possible to find the appropriate information that matches the search criteria.

According to the next invention, the editing means provides the input means with a search condition relating to the data stored in the database, performs editing processing based on the search result output to the output means, The analyzing means obtains the editing result by the editing means, and analyzes the similar case or the similar search procedure by adding the editing result. Therefore, during the data editing process, the information matching the search condition is performed. A search can be performed, and the time and effort required for inputting search conditions can be reduced. In addition, since the edited result is treated equally as selection information, the time and effort required for user selection evaluation can be reduced. This has the effect that it can be performed.

According to the next invention, by the input step,
A search condition for a database holding search target data is input, a search condition input in the input step is held in a first holding step, and a search condition expansion step stores the search condition in the database based on the search condition. A similar case database holding similar cases of the search target data to be stored is searched, an expanded search condition obtained by expanding the search condition is output, and the expanded search condition expanded by the search condition expanded step is also searched by the search step. And the output step outputs the search result of the search step, and the second holding step holds the selection information selected by the user with respect to the search result output by the output step. A search condition held in the first holding step and a search condition held in the second holding step. The similar case is analyzed based on the information, and the update process updates the content of the similar case database based on the analysis result of the analysis process. In addition, there is an effect that the search can be performed without omission.

According to the next invention, by the input step,
A search condition for a database holding search target data is input, a first holding step holds the search condition input by the input step, and a similarity search procedure database holds a similarity search procedure by a guidance step. Searching and outputting the next similar search condition corresponding to the search condition, searching the database based on the next similar search condition searched in the guidance step in a search step, and performing the search in the output step Outputting the search result by the step, holding the selection information selected by the user with respect to the search result output by the output step by the second holding step, and searching by the analysis step the search condition held by the first holding step And analyzing the similarity search procedure based on the selection information held in the second holding step, Therefore, since the content of the similar case database is updated based on the analysis result of the analysis step, it is possible to appropriately perform a narrowing-down search of desired information that matches a search condition. .

According to the next invention, by the input step,
A search condition for a database holding search target data is input, a search condition input in the input step is held in a first holding step, and a search condition expansion step stores the search condition in the database based on the search condition. A similar case database that holds similar cases of the search target data to be stored is searched, and an expanded search condition obtained by expanding the search condition is output. From a similar search procedure database in which a similar search procedure is held by a guidance process, Searching and outputting the next similar search condition corresponding to the search condition, and in the search step, by the user's selection, the expanded search condition output in the search condition expansion step or the next similar search searched in the guidance step Search the database based on the conditions, and by the output process,
The search result of the search step is output, the selection information selected by the user for the search result output by the output step is held by the second holding step, and the selection information is held by the first holding step by the analysis step. Analyzing similar cases and similar search procedures based on search conditions and the selection information held by the second holding step,
Since the content of the similar case database and the content of the similar search procedure database are updated based on the analysis result of the analysis process, a process of completely searching for information that matches the search condition, And a process of narrowing and searching for information that matches the search condition can be selectively performed, and furthermore, this selection can be arbitrarily selected by the user according to the number of search results up to now, Further, it is possible to search for appropriate information that matches the search condition.

According to the next invention, in the input step, a search condition relating to data stored in the database and input from the editing means for performing an edit process based on a search result in the output step is input, and the analysis step is performed in the input step. Since the editing unit performs analysis by adding the edited result obtained by performing the editing process based on the search result in the output step, it is possible to search for information that matches the search condition during the data editing process. It is possible to reduce the time and effort required for inputting search conditions and to treat the edited result equally as selection information, thereby reducing the time and effort required for user selection evaluation. Play.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration of a database search device according to a first embodiment of the present invention.

FIG. 2 is a diagram showing a process of expanding search conditions by the database search device shown in FIG. 1;

3 is a diagram showing a similar case analysis process by the database search device shown in FIG. 1;

FIG. 4 is a block diagram illustrating a configuration of a database search device according to a second embodiment of the present invention;

FIG. 5 is a diagram illustrating an example of search operation holding data by the database search device illustrated in FIG. 4;

FIG. 6 is a diagram illustrating an example of a similarity search procedure database of the database search device illustrated in FIG. 4;

FIG. 7 is a diagram illustrating search condition guidance processing by the database search device illustrated in FIG. 4;

FIG. 8 is a block diagram illustrating a configuration of a database search device according to a third embodiment of the present invention;

FIG. 9 is a diagram illustrating a search evaluation result analysis process performed by the database search device illustrated in FIG. 8;

FIG. 10 is a diagram illustrating an example of a similarity search procedure database of the database search device illustrated in FIG. 8;

FIG. 11 is a block diagram illustrating a configuration of a database search device according to a fourth embodiment of the present invention.

FIG. 12 is a block diagram showing a configuration of a conventional database search device.

FIG. 13 is a block diagram showing a configuration of a conventional interactive device.

[Explanation of symbols]

1 database, 2 search condition input section, 3 search operation holding section, 4 search operation holding processing section, 5 similar case database, 6 search condition expansion section, 7 database search section, 8 search result output section, 9 search result evaluation data storage Section, 10 search result evaluation section, 11 similar case analysis section, 1
2 similar case database update unit, 13 similar search procedure database, 14 search condition guidance unit, 15 search procedure analysis unit, 16 similar search procedure database update unit,
17, 20 Search evaluation result analysis unit, 18 Data editing unit, 19 Newly created data holding unit.

Claims (8)

[Claims] 1. A database that stores search target data; a similar case database that stores similar cases of search target data stored in the database in a database; and input means for inputting search conditions of the database; A search condition expansion unit that searches the similar case database based on the search condition input to the input unit and outputs an expanded search condition obtained by expanding the search condition; and an expansion search condition expanded by the search condition expansion unit. Search means for searching the database on the basis of: an output means for outputting a search result by the search means; a user selecting a search condition input to the input means and a search result output by the output means Search operation holding means for holding a user's search operation regarding the search of the database including the selected information, A similar case analysis unit that analyzes a similar case based on the selection information and the search operation held in the search operation holding unit; and a content of the similar case database based on an analysis result by the similar case analysis unit. A database search device, comprising: updating means for updating. 2. A database for storing search target data; input means for inputting search conditions of the database; and a similarity search procedure corresponding to the search conditions input from the input means. A similarity search procedure database held as: a search means for searching the similarity search procedure database based on a search condition input to the input means, and outputting a next similarity search condition corresponding to the search condition; A search unit for searching the database based on the next similar search condition output by the guidance unit; an output unit for outputting a search result by the search unit; a search condition input to the input unit; and the output unit. The user's search operation related to the search of the database including the selection information selected by the user for the output search result A search operation holding means for holding a work, a similar search procedure analyzing means for analyzing a similar search procedure based on the selection information and the search operation held in the search operation holding means, and a similar search procedure analyzing means. Updating means for updating the contents of the similarity search procedure database based on the analysis result. 3. A database holding search target data; input means for inputting search conditions of the database; a similar case database storing similar cases of search target data stored in the database in a database; A similarity search procedure database that stores the next similarity search condition corresponding to the search condition input from the input means and stores the similarity search procedure as a similarity search procedure, and the similar case database based on the search condition input to the input means. Search condition expanding means for searching and outputting expanded search conditions obtained by expanding the search conditions; searching the similar search procedure database based on the search conditions input to the input means; Guidance means for outputting a similar search condition, and expansion search conditions output by the search condition expansion means. A search means for searching the database based on the next similar search condition output by the guidance means; an output means for outputting a search result by the search means; a search condition input to the input means; Search operation holding means for holding a user's search operation related to the search of the database including the selection information selected by the user with respect to the search result output by the output means; and Analysis means for analyzing similar cases and similar search procedures based on the search operation; similar case update means for updating the contents of the similar case database based on the analysis results by the analysis means; and analysis by the analysis means Similarity search procedure updating means for updating the contents of the similarity search procedure database based on the result. Database search device. 4. An editing unit for providing a search condition for data stored in the database to the input unit, and performing an edit process based on a search result output to the output unit, The database search device according to claim 1, wherein an edit result obtained by the edit unit is acquired, and the similar case or the similar search procedure is analyzed by adding the edit result. 5. An input step of inputting a search condition for a database holding search target data; a first holding step of holding a search condition input in the input step; and the database based on the search condition. A search condition expanding step of searching a similar case database holding similar cases of search target data stored in the search condition and outputting an expanded search condition obtained by expanding the search condition; and an expanded search condition expanded by the search condition expanding step. A search step of searching the database based on a search result; an output step of outputting a search result of the search step; and a second holding step of holding selection information selected by a user with respect to the search result output by the output step. And similarity based on the search condition held in the first holding step and the selection information held in the second holding step. A database search method, comprising: an analysis step of analyzing an example; and an update step of updating the contents of the similar case database based on an analysis result of the analysis step. 6. An input step of inputting search conditions for a database holding search target data; a first holding step of holding search conditions input in the input step; and a similarity search procedure holding a similarity search procedure. A guidance step of searching and outputting a next similar search condition corresponding to the search condition from a database; a search step of searching the database based on the next similar search condition searched by the guidance step; An output step of outputting a search result by the search step; a second holding step of holding selection information selected by a user with respect to the search result output by the output step; a search condition held by the first holding step; An analysis step of analyzing a similarity search procedure based on the selection information held by the second holding step; Database search method characterized by comprising the updating step of updating the contents of the analogous case database based on the analysis result. 7. An input step of inputting a search condition for a database holding search target data; a first storage step of holding a search condition input in the input step; and the database based on the search condition. A search condition expansion step of searching a similar case database holding similar cases of search target data stored in the search target and outputting expanded search conditions obtained by expanding the search conditions; and a similar search procedure database holding similar search procedures. A guidance step of searching for and outputting a next similar search condition corresponding to the search condition; and a development search condition output by the search condition development step or a next similarity searched by the guidance step, depending on a user's selection. A search step of searching the database based on search conditions; An output step of performing, a second holding step of holding selection information selected by a user with respect to a search result output by the output step, and a search condition held by the first holding step and holding by the second holding step. An analysis step of analyzing similar cases and similar search procedures based on the selected information, and an update step of updating the contents of the similar case database and the similar search procedure database based on the analysis result of the analysis step And a database search method comprising: 8. The input step includes inputting a search condition relating to data stored in the database from an editing unit that performs an editing process based on a search result obtained in the output step. 8. The database search method according to claim 5, wherein an analysis is performed by adding an edited result based on the search result in the output step.