JP2019164628A - Instance retrieval system - Google Patents

Abstract

To provide an instance retrieval system with which an instance that can be a good reference can easily be retrieved.SOLUTION: An instance retrieval system according to an embodiment of the present invention has a processing unit. The processing unit, when being input with a character string corresponding to a first structure, selects a second structure based on the character string, determines evaluation of the second structure, and acquires an instance associated with the second structure. The processing unit displays a first screen including identification information of the second structure, the evaluation, and the instance.SELECTED DRAWING: Figure 1

Description

  Embodiments described herein relate generally to a case search system.

  In the case search system, it is desirable that reference cases can be found more easily.

JP 2010-244465 A

  The problem to be solved by the present invention is to provide a case search system that can more easily find a reference case.

  The case search system according to the embodiment includes a processing unit. When the character string corresponding to the first structure is input, the processing unit selects the second structure based on the character string, determines an evaluation regarding the second structure, and the second structure. Get a case associated with an object. The processing unit displays a first screen including identification information of the second structure, the evaluation, and the case.

It is a schematic diagram which illustrates the structure of the case search system which concerns on embodiment. It is a flowchart which illustrates operation | movement of the case search system which concerns on embodiment. It is a schematic diagram which illustrates the display screen by the case search system which concerns on embodiment. It is a schematic diagram which illustrates the display screen by the case search system which concerns on embodiment. It is a schematic diagram which illustrates the display screen by the case search system which concerns on embodiment. It is a schematic diagram which illustrates the display screen by the case search system which concerns on embodiment. It is a schematic diagram which illustrates the display screen by the case search system which concerns on embodiment. It is a schematic diagram which illustrates the structure of another example search system which concerns on embodiment. It is a schematic diagram which illustrates the display screen by another example search system which concerns on embodiment. It is a flowchart which illustrates operation | movement of another example search system which concerns on embodiment. It is a schematic diagram which illustrates the hardware which implement | achieves the case search system which concerns on embodiment.

Embodiments of the present invention will be described below with reference to the drawings.
In the present specification and each drawing, the same elements as those already described are denoted by the same reference numerals, and detailed description thereof will be omitted as appropriate.

FIG. 1 is a schematic view illustrating the configuration of a case search system according to the embodiment.
The case search system 1 according to the embodiment includes a processing unit 10. In the example illustrated in FIG. 1, the case search system 1 further includes an input unit 20, a display unit 30, a first storage unit 41, a second storage unit 42, and a third storage unit 43.

  The user operates the input unit 20 to input a character string corresponding to a newly configured structure to the processing unit 10. When the processing unit 10 receives an input of a character string, the processing unit 10 displays the identification information of another structure similar to the structure corresponding to the character string, its evaluation, and an example of another structure on the display unit 30. Let

Here, the structure refers to an object composed of a plurality of parts. The structure is, for example, a power plant such as a thermal power plant, a hydro power plant, a nuclear power plant, or a component thereof. Such structures are typically designed and assembled over time. For example, if there is an improper design in the structure, the structure cannot be assembled properly and significant losses may occur. Therefore, when designing a new structure, it is desirable to grasp the problem that has occurred in the structure that has already been constructed and its countermeasure, so that the same problem is not repeated. Examples of problems include dimensional design errors, material selection errors, and unintentional breakage of parts.
However, a structure such as a power plant has a long period from receiving an order to delivery, and the lifetime of one structure is generally long. The number of times a single designer is involved in designing a new structure is limited. Conventionally, the causes and countermeasures of past problems are often accumulated as experienced knowledge of experienced designers. When designing a new structure, a young designer was instructed by an experienced designer and learned from past problem cases related to the new structure. In the transfer of experience knowledge from person to person, there is a risk that knowledge to be inherited with the person is lost. For this reason, in recent years, minutes, reference cases, success cases, non-conformance cases, etc. are recorded, and a database is constructed by collecting them. In the minutes, for example, the contents and results of a design review meeting held in the past are recorded. In the nonconformance case, for example, the content of the problem that has occurred in the past structure, its cause, and countermeasures are recorded. The reference case records the problems found in the design and their solutions. Success stories describe effective measures to improve the function of the structure and shorten the construction period. By constructing a database that summarizes these cases, more designers can refer to and learn more cases.
When retrieving a case from a database, a designer usually inputs a keyword related to a structure to be designed. If the structure is a power plant, instead of keywords, the type of power plant such as thermal power, hydropower, nuclear power, maximum output, generator speed, power generation method (for example, steam, gas, combined, etc. Type) may be specified.
However, when only these search conditions are specified, a large number of related cases are output as search results. In this case, it takes a lot of time to confirm the output record. In addition, young designers who are particularly inexperienced may not be able to determine which case of search results to focus on.

With respect to this problem, according to the case search system 1 according to the embodiment, as described above, the identification information of another structure, its evaluation, and the case of another structure are displayed based on the input character string. Is done. The input character string corresponds to a newly designed structure.
Therefore, according to this case search system 1, it becomes easy to search for a case that may be used as a reference when configuring a new structure. For this reason, the time required for the case investigation can be shortened. In addition, for example, by confirming a case associated with a structure with high similarity, the user can effectively learn a case that is more likely to be a reference.

FIG. 2 is a flowchart illustrating the operation of the case search system according to the embodiment.
3 to 7 are schematic views illustrating display screens by the case search system according to the embodiment.

  3-7 has illustrated the case where a structure is a power plant. For example, the processing unit 10 displays the screen 100a illustrated in FIG. 3 on the user terminal. In this example, the screen 100a includes three input areas 101 to 103. The user inputs a character string for specifying a structure formed from any of the input areas 101 to 103. When the user inputs a character string and clicks on the icon 104, the input character string is transmitted to the processing unit 10.

  When receiving the input of the character string, the processing unit 10 refers to the first storage unit 41. The first storage unit 41 stores a document related to a newly configured structure. The processing unit 10 acquires a first document related to the structure specified by the character string from those documents (step S1 in FIG. 2). As an example, for a newly constructed structure, a design document, a minutes of examination, a report, and the like are acquired as the first document. Hereinafter, the structure specified by the character string input by the user is referred to as a new structure (first structure). Moreover, the structure comprised in the past is called an existing structure (2nd structure).

  The processing unit 10 refers to the second storage unit 42. The 2nd memory | storage part 42 memorize | stores the document regarding the structure comprised in the past. The processing unit 10 acquires one or more second documents similar to the first document from the second storage unit 42 (step S2). One storage unit may function as the first storage unit 41 and the second storage unit 42.

  The processing unit 10 selects an existing structure associated with each second document as a structure similar to the new structure (step S3). Moreover, the process part 10 determines the evaluation which shows the similarity between a new structure and the selected structure (step S4). This evaluation is determined based on, for example, the similarity between the first document and the second document.

  The processing unit 10 refers to the third storage unit 43. The 3rd memory | storage part 43 has memorize | stored the example regarding the existing structure, for example. The processing unit 10 acquires an associated case for each of the selected existing structures (step S5). The processing unit 10 displays the first screen (step S6). The first screen includes the selected existing structure, an evaluation indicating the similarity between the new structure and the existing structure, and a case associated with the existing structure.

  For example, the processing unit 10 displays the screen 100b (first screen) illustrated in FIG. The screen 100b includes information 111 to 113 of selected existing structures, evaluations 121 to 123 of these existing structures, and cases 131 to 133 associated with these existing structures.

  In the example illustrated in FIG. 4, the name of the existing structure and the type of the existing structure are displayed as information 111 to 113 for identifying the existing structure. As the information 111 to 113, an ID of an existing structure may be displayed.

  As evaluation 121-123, the bar which shows the similarity (similarity degree) with each of a new structure and each existing structure is displayed. A longer bar indicates that the existing structure is more similar (more similar) to the new structure. Instead of the bar, a score indicating similarity may be displayed. Alternatively, the similarity ranking may be displayed.

  As examples 131-133, the name of a case and a part (snippet) of the contents of a case are displayed, for example. For example, when a case name or snippet is selected, details of the case are displayed on the screen.

In the example shown in FIG. 4, scroll bars 131 a to 133 a are provided in the case display area so that the user can confirm more cases within a predetermined range. The user can display different cases in the display area of each case by operating the scroll bars 131a to 133a.
Alternatively, in the initial state, only a part of the cases may be displayed by folding the case display area. For example, when the user clicks an area in which some cases are displayed, the user can expand the folded area and display more cases on the same screen.

  If the display area of the case is too large on the output screen, it is difficult to confirm all existing structures when there are many similar existing structures. By providing the scroll bar or the folding of the display area, it is easy to check all similar existing structures while checking the outline of the examples related to each existing structure.

  The processing unit 10 may further display a word (keyword) 105 on the output screen. The displayed word is at least a part of a plurality of words extracted from the first document and matched between the first document and the second document. By displaying the word, the user can guess the reason why the displayed existing structure is selected. Thereby, for example, when the selected existing structure is different from the user's assumption, the user can take measures such as changing the input character string.

  As illustrated in FIG. 5, the processing unit 10 may display a word (keyword) for each existing structure on the output screen. In the example of FIG. 5, the word 105 a is displayed in association with the first existing structure specified by the information 111. The words 105b and 106c are displayed in association with the second existing structure related to the information 112 and the third existing structure related to the information 113, respectively. For example, at least one of the words 105a is different from at least one of the words 105b and at least one of the words 105c.

  The word 105a is at least part of a plurality of words extracted from the first document and matched between the first document and the second document associated with the first existing structure. Similarly, the word 105b is extracted from the first document and is at least a part of a plurality of words matched between the first document and another second document associated with the second existing structure. The word 105c is extracted from the first document and is at least a part of a plurality of words matched between the first document and another second document associated with the third existing structure.

  As illustrated in FIG. 6, the processing unit 10 may display information 111 to 114 regarding a plurality of structures, evaluations 121 to 124 of each structure, and links 141 to 144 on the screen 100 b. When the user clicks one of the links 141 to 144, the processing unit 10 displays a list of examples of existing structures corresponding to the link, for example, a screen 100c illustrated in FIG. By displaying the link together with the identification information and evaluation of the existing structure instead of the case, more identification information of the existing structure can be displayed on the display unit 30.

Details of the case search system 1 according to the embodiment will be described below.
In the case search system 1 according to the embodiment, the processing unit 10 includes a reception unit 11, a first acquisition unit 12, a selection unit 13, a second acquisition unit 14, and an output unit 15, as illustrated in FIG.

  The user inputs a character string using the input unit 20. The input character string may be anything that can identify the structure. The character string includes, for example, at least one of a structure name, a structure ID, and a structure part name. The accepting unit 11 accepts an input character string.

  The first acquisition unit 12 refers to the first storage unit 41 and acquires one or more documents related to the new structure corresponding to the input character string. Alternatively, the user may specify a document to be input to the processing unit 10 using the input unit 20. In this case, the step of acquiring a document from the first storage unit 41 is omitted.

  The selection unit 13 refers to the second storage unit 42 and extracts a second document whose written text is similar to the first document. For the similarity determination between the first document and another document, a general method for determining the similarity between texts can be used. In this similarity determination, synonyms of words included in the first document may be further used. Specifically, when a word extracted from the first document is registered in the synonym dictionary, the word registered as a synonym of the word is also used for similarity determination.

In this similarity determination, the semantic similarity of sentences included in the first document may be further used. Specifically, first, a similarity score is obtained by a general similarity determination method between texts. This similarity score is corrected using the similarity score based on the semantic similarity of the sentence. In the correction, for example, each obtained similarity score is added with a weight. The semantic similarity of sentences is obtained by, for example, a technique for obtaining a distributed expression of sentences. In the distributed representation of sentences, semantically similar sentences are arranged at close distances in the vector space. For this reason, the similarity can be obtained using, for example, the distance between two vectors.
When a plurality of first documents are acquired by the first acquisition unit 12, the selection unit 13 combines them and handles them as one document, for example.

  The selection unit 13 selects the existing structure corresponding to the second document as an existing structure similar to the new structure specified by the user. In addition, the selection unit 13 determines an evaluation indicating the similarity between the new structure and the selected existing structure.

  For example, the selection unit 13 sets the similarity between the first document and the second document as an evaluation indicating the similarity between the new structure and the selected existing structure. When a plurality of second documents correspond to one existing structure, for example, the selection unit 13 combines the second documents and handles them as one document. For example, the selection unit 13 calculates the similarity between the one document and the first document, and uses the similarity as the evaluation. Alternatively, the selection unit 13 may calculate respective similarities between the plurality of second documents and the first document, and an average value of these similarities may be used as the evaluation.

  The second acquisition unit 14 acquires a case associated with the selected existing structure from the third storage unit 43. The output unit 15 outputs the identification information (name, type, ID, etc.) of the selected existing structure, the evaluation, and the example to the display unit 30 in association with each existing structure.

  According to this case search system 1, as described above, it is easy to search for a case that may be used as a reference when configuring a new structure.

FIG. 8 is a schematic view illustrating the configuration of another example search system according to the embodiment.
FIG. 9 is a schematic view illustrating a display screen by another example search system according to the embodiment.

  In the case search system 2 illustrated in FIG. 8, the processing unit 10 further includes a calculation unit 16. The calculation unit 16 calculates the priority of each selected existing structure. The calculation of the priority is based on, for example, the importance of the case associated with the existing structure.

  For example, a nonconforming case is associated with an existing structure. Non-conformance cases may include characters indicating their importance. For example, a nonconformity case with low importance relates to a minor nonconformity that is easy to correct. Non-conformance cases with high importance relate to serious nonconformities such as a delay in the construction period due to a backlash of design and assembly. The character indicating the importance may be called, for example, a rank, a division, or an evaluation. For example, the character is any one of A, B, C, and D. A indicates that the nonconforming case is the most important, and D indicates that the nonconforming case is the least. The character may be a number. For example, the larger the number, the more important the nonconformance case is.

  The calculation unit 16 calculates priority using characters indicating importance. For example, when the character is any one of A, B, C, and D, A is 6, B is 4, C is 2, and D is 1, and the importance is scored. The calculation part 16 scores the importance of all the cases associated with a certain existing structure, and sets the average value as a priority. The calculation unit 16 calculates the priority for each selected existing structure. The output unit 15 outputs, for example, information on the selected existing structure, evaluation, at least part of the case, and priority to the display unit 30 in association with each existing structure.

  The output unit 15 may display a numerical value indicating the priority. Alternatively, the output unit 15 may make one display mode (first display mode) of a plurality of existing structures different from another display mode (second display mode) of the plurality of existing structures. The difference between the first display mode and the second display mode is based on a difference in priority between one of the plurality of existing structures and another one of the plurality of existing structures.

  For example, as illustrated in FIG. 9, the output unit 15 makes the display modes of the areas 151 to 153 on the screen 100 b different from each other. The area 151 includes an area where the information 111 and the evaluation 121 are displayed. Similarly, the area 152 includes an area where the information 112 and the evaluation 122 are displayed. The area 153 includes an area where the information 113 and the evaluation 123 are displayed. In this example, dots are added to the areas 151 and 152, so that the display mode of these areas is different from the display mode of the area 153. In addition, the display mode of the region 151 is different from the display mode of the region 152 because the dot density of the region 151 and the dot density of the region 152 are different. For example, the higher the dot density, the higher the priority of the associated case. That is, the priority of the case associated with the existing structure specified by the information 111 is lower than the priority of the case associated with the existing structure specified by the information 112, and the existing structure specified by the information 113 It is higher than the priority of the case associated with the object.

  Or when the feedback from the user with respect to the case is memorize | stored with a case, a priority may be calculated using this feedback. As an example, an icon labeled “helpful” and an icon labeled “not useful” are displayed on the case browsing screen. The user can send feedback on the case by clicking on any icon after reading the case. Cases that are fed back to many users as “helpful” may be more useful than other cases, even if they are less important. By calculating the priority using feedback from the user, it becomes easier for the user to search for such cases.

FIG. 10 is a flowchart illustrating the operation of another example search system according to the embodiment.
In the flowchart shown in FIG. 10, steps S1 to S5 are executed in the same manner as steps S1 to S5 in the flowchart shown in FIG. The processing unit 10 calculates the priority of each selected existing structure (step S7). The processing unit 10 causes the first screen to display the selected existing structure, the evaluation indicating the similarity between the new structure and the existing structure, the case associated with the existing structure, and the priority of the existing structure. (Step S8).

FIG. 11 is a schematic diagram illustrating hardware that implements the case search system according to the embodiment.
The case search system 1 (or 2) according to the above-described embodiment includes a CPU (Central Processing Unit) 111, an input device 112, an output device 113, a ROM (Read Only Memory) 114, and a RAM (Random Access Memory). 115, a storage device 116, a communication device 117, and a bus 118. Each unit is connected by a bus 118.

  The CPU 111 executes various processes in cooperation with various programs stored in advance in the ROM 114 or the storage device 116, and comprehensively controls the operation of each unit constituting the case search system 1. In the processing, the CPU 111 uses a predetermined area of the RAM 115 as a work area. The CPU 111 implements the input device 112, the output device 113, the communication device 117, and the like in cooperation with a program stored in advance in the ROM 114 or the storage device 116.

  The input device 112 includes, for example, at least one of a keyboard, a mouse, and a touch panel. The input device 112 receives information input from the user as an instruction signal, and outputs the instruction signal to the CPU 111. The output device 113 is a monitor, for example. Based on the signal output from the CPU 111, the output device 113 outputs various information so as to be visible.

  The ROM 114 stores a program used for controlling the case search system 1 and various setting information in a non-rewritable manner. The RAM 115 is a volatile storage medium such as SDRAM (Synchronous Dynamic Random Access Memory). The RAM 115 functions as a work area for the CPU 111. Specifically, it functions as a buffer or the like that temporarily stores various variables and parameters used by the case search system 1.

  The storage device 116 is a rewritable recording device such as a semiconductor storage medium such as a flash memory or a magnetically or optically recordable storage medium. The storage device 116 stores programs and various setting information used for controlling the case search system 1. The storage device 116 functions as the second storage unit 42. The storage device 116 may further function as the first storage unit 41. The communication device 117 is used to communicate with an external device to transmit / receive information.

  According to the embodiment described above, it is possible to provide a case search system that can more easily find a case that can be a reference for a new structure.

  The embodiments of the present invention have been described above with reference to specific examples. However, embodiments of the present invention are not limited to these specific examples. For example, regarding the specific configuration of each element such as a processing unit, an input unit, an output unit, and a storage unit, those skilled in the art appropriately implement the present invention by appropriately selecting from a known range, and obtain similar effects As long as it is possible, it is included in the scope of the present invention.

  Moreover, what combined any two or more elements of each specific example in the technically possible range is also included in the scope of the present invention as long as the gist of the present invention is included.

  In addition, any case search system that can be implemented by a person skilled in the art based on the case search system described above as an embodiment of the present invention can be implemented as long as it includes the gist of the present invention. Belonging to.

  In addition, in the category of the idea of the present invention, those skilled in the art can conceive of various changes and modifications, and it is understood that these changes and modifications also belong to the scope of the present invention. .

  Although several embodiments of the present invention have been described, these embodiments are presented by way of example and are not intended to limit the scope of the invention. These novel embodiments can be implemented in various other forms, and various omissions, replacements, and changes can be made without departing from the scope of the invention. These embodiments and modifications thereof are included in the scope and gist of the invention, and are included in the invention described in the claims and the equivalents thereof.

  1, 2 case search system, 10 processing unit, 11 reception unit, 12 first acquisition unit, 13 selection unit, 14 second acquisition unit, 15 output unit, 16 calculation unit, 20 input unit, 30 display unit, 41 first Storage unit, 42 second storage unit, 43 third storage unit, 100a, 100b, 100c screen, 101-103 input area, 104 icon, 111-114 information, 121-124 evaluation, 131a-133a scroll bar, 131-133 Case, 141-144 link, 151-153 area, S1-S8 step

Claims (12)

When a character string corresponding to the first structure is input, a second structure is selected based on the character string, an evaluation regarding the second structure is determined, and associated with the second structure Take a case,
Displaying the first screen including the identification information of the second structure, the evaluation, and the case;
Case search system with a processing unit.   The processing unit includes a plurality of different second structures, a plurality of evaluations corresponding to the plurality of second structures, and a plurality of cases corresponding to the plurality of second structures, respectively. The case search system according to claim 1, wherein: is displayed on the first screen. When a character string corresponding to the first structure is input, a second structure is selected based on the character string, an evaluation regarding the second structure is determined, and associated with the second structure Take a case,
Displaying a first screen including identification information of the second structure, the evaluation, and a link for displaying the case;
Case search system with a processing unit.   The processing unit includes a plurality of different second structures, a plurality of evaluations corresponding to the plurality of second structures, and a plurality of links corresponding to the plurality of second structures, respectively. The case search system according to claim 3, wherein: is displayed on the first screen.   5. The case search according to claim 2, wherein in the first screen, one first display mode of the plurality of second structures is different from another second display mode of the plurality of second structures. system.   The case search system according to claim 5, wherein the processing unit makes the first display mode different from the second display mode according to the importance of the case associated with the one of the plurality of second structures. .   The processing unit makes the first display mode different from the second display mode according to feedback from the user with respect to the case associated with the one of the plurality of second structures. The case search system described.   The case search system according to claim 1, wherein the processing unit further displays a plurality of words related to the first structure and the second structure on the first screen.   The case search system according to any one of claims 1 to 8, wherein the evaluation indicates a similarity between the first structure and the second structure. When a character string corresponding to the first structure is input, a first document associated with the first structure is acquired from a first storage unit that stores a plurality of documents.
Using the respective similarities between the first document and the plurality of documents, obtaining a second document from a second storage unit storing another plurality of documents;
Selecting a second structure associated with the second document;
Obtaining a first case associated with the second structure from a third storage unit for storing a plurality of cases;
Displaying the first screen including the identification information of the second structure and the first case;
Case search system with a processing unit.   The processing unit extracts and acquires the second document by further using inter-document similarity between the first document and the plurality of documents when acquiring the second document. Item search system of claim | item 10.   When the processing unit acquires the second document, the processing unit further uses respective semantic similarities between the sentences included in the first document and the plurality of sentences included in the plurality of documents. The case search system according to claim 10 or 11, wherein the second document is extracted and acquired.

Images