TWI735892B - Search device, and search method - Google Patents

Abstract

The search device of the present invention searches for a specification data similar to a search condition from a specification data group that is individually composed of one or more specification item values indicating the details of an item representing the specification of a product. The search device (1) includes a similarity learning unit (14) which learns all combinations of the specification item values for the extent to which the two specification item values of the same item are similar to each other, based on a specification data group held by a product design database (13); an input unit (12) which accepts the input of the specification item value used as the search condition; and a similar specification search department (16) which searches for a specification data similar to the input specification data composed of the specification item values accepted and input by the input unit (12) based on the learning result learned by the similarity learning unit (14).

Description

Retrieval device and retrieval method

The present invention relates to a retrieval device, retrieval method, and machine learning device for retrieving data related to specifications similar to the specifications displayed by the input data from data related to specifications of past products.

At the design site of products with diversified specifications requested by customers, part of the data from design cases of past products with similar specifications is repeatedly used when designing new products, so as to improve the efficiency of design work and the inspection of design adequacy ( Simplification of check). However, in order to efficiently retrieve the data of design cases of past products with similar specifications, it is necessary to have in-depth knowledge of the product, and a large part of it depends on the experience of the person in charge of the design.

In view of this, in order to efficiently search data of design cases of past products without relying on the experience of the designer, an invention described in Patent Document 1 below has been proposed. The invention described in Patent Document 1 has a design case base (base) that stores data of past design cases, a knowledge base that stores knowledge of design specifications that should be applied from the required specifications, and uses the knowledge base to determine the design that should be applied The inference engine of the specification (engine), consider Retrieval of similar design cases based on the design restrictions and interference relationships among the specification items that constitute the required specifications.

(Prior technical literature) (Patent Document)

Patent Document 1: Japanese Patent Application Publication No. 8-221437

However, in the invention described in Patent Document 1, it is necessary to manually describe the knowledge of the design specifications that should be applied from the required specifications to the knowledge base as a rule, which has a problem of very laborious time.

The present invention is developed in view of the above circumstances, and aims to obtain a retrieval device that can reduce the workload of users.

In order to solve the above-mentioned problems and achieve the objective, the present invention is a retrieval device that retrieves specification data similar to the retrieval condition from a specification data group composed of more than one specification item value showing the details of the item representing the specification of the product. The retrieval device is equipped with: a similarity learning section, which is based on the specification data group held in the product design database (database), for the degree to which two specification item values of the same item are similar to each other, learning for all combinations of specification item values; And the input part accepts the input of the value of the specification item used as the search condition. In addition, the search device is equipped with a similar specification search unit, which is based on similarity As a result of the learning by the learning unit, the specification data similar to the input specification data constituted by the specification item values received by the input unit is retrieved.

The retrieval device of the present invention achieves the effect of reducing the workload of the user.

1. 1a to 1c, 1e: retrieval device

11: Control Department

12: Input section

13: Product design database

14, 14d: Similarity Learning Department

15: Similarity data

16, 16c, 16e, 16f: similar specification search department

17: Search result display section

18: Search result details display section

21: Weight data change department

31: Similarity Adjustment Department

32: Adjusted similarity data

41: Use parts extraction part

42: Similarity calculation department

43: Use parts selection department

44: Use part selection definition

51: Special specification input section

61, 61c: Specification similarity calculation department

62: Data Column

63: Input specification data

64: Search results of similar specifications

65: Similarity weight data

81: Special specification similarity calculation department

82: Expanded specification similarity calculation department

83: Input special specification data

84: Expand similarity weight data

85: Weight Adjustment Department

86: Project-specific keywords

101: processing circuit

102: memory device

103: input device

104: display device

131: Specification list

132: Parts Composition Table

201: Processor

202: memory

601: Model ID

602: Series name

603: Rated output

604: Rated rotation speed

605: Brake

606: Special specifications

701: parts hierarchy

702: Part ID

703: Part name

704: number

1301 to 1304, 3701: input field

1501, 3301, 3901, 4501: Specification items

1502, 3902: similarity weight

3302: List of associated part numbers

4502: keyword list

PA01, PB02, PC02, PA001 to PA004, PA010, PA011, PA020, PA021, PB010, PB011, PB020, PB021, PC020, PC021: parts

S11 to S14, S21 to S25, S31 to S33, S41, S51, S61, S71, S72, S1: Step

X, Y: model

Fig. 1 is a diagram showing an example of the configuration of the search device of the first embodiment.

Fig. 2 is a diagram showing an example of the hardware configuration of the search device of the first embodiment.

Fig. 3 is a diagram showing another example of the hardware configuration of the search device of the first embodiment.

FIG. 4 is a diagram showing an example of the detailed structure of the similarity learning unit included in the retrieval device of the first embodiment.

Fig. 5 is a diagram showing an example of the structure of a product design database included in the search device of the first embodiment.

Fig. 6 is a diagram showing an example of a list of specifications held in the product design database of the first embodiment.

Fig. 7 is a diagram showing an example of the parts composition table held in the product design database of the first embodiment.

Fig. 8 is a flow chart showing an example of the operation of the similarity learning unit included in the retrieval device of the first embodiment.

Figure 9 shows the relationship between the item values of the similarity learning unit included in the retrieval device of the first embodiment. A conceptual diagram of the similarity calculation method.

Fig. 10 is a diagram showing an example of similarity data generated by the similarity learning unit included in the retrieval device of the first embodiment.

Fig. 11 is a diagram showing another example of similarity data generated by the similarity learning unit included in the retrieval device of the first embodiment.

Fig. 12 is a flowchart showing an example of the search operation performed by the search device of the first embodiment.

Fig. 13 is a diagram showing an example of a specification input screen displayed by the search device of the first embodiment.

Fig. 14 is a diagram showing an example of the detailed structure of the similar specification search unit included in the search device of the first embodiment.

Fig. 15 is a diagram showing an example of similarity weight data used by the similar specification search unit of the search device of the first embodiment.

Fig. 16 is a flowchart showing an example of the operation of the similar specification search unit included in the search device of the first embodiment.

Fig. 17 is a diagram showing an example of a calculation result calculated by the specification similarity calculation unit of the retrieval device of the first embodiment.

Fig. 18 is a diagram showing an example of search results of similar specifications displayed on the search result display unit of the search device of the first embodiment.

Fig. 19 is a diagram showing a first modification of the search device of the first embodiment.

FIG. 20 is a flowchart showing an example of the search operation performed by the search device according to the first modification of the first embodiment.

Fig. 21 is a diagram showing an example of search results of similar specifications displayed on the search result detail display section of the search device according to the first modification of the first embodiment.

Fig. 22 is a diagram showing a second modification of the search device of the first embodiment.

FIG. 23 is a flowchart showing an example of the search operation performed by the search device according to the second modification of the first embodiment.

Fig. 24 is a diagram showing an example of a similarity weight data change screen displayed by the search device of the second modification of the first embodiment.

FIG. 25 is a diagram showing an example of a similar specification search result displayed by the search result detail display unit of the search device according to the second modification of the first embodiment.

Fig. 26 is a diagram showing an example of the configuration of the search device of the second embodiment.

Fig. 27 is a flowchart showing an example of the specification similarity learning process of the retrieval device of the second embodiment.

Fig. 28 is a diagram showing an example of a similarity adjustment screen between item values displayed by the search device of the second embodiment.

Fig. 29 is a diagram showing an example of the adjusted similarity data output by the similarity adjusting unit of the retrieval device of the second embodiment.

Fig. 30 is a diagram showing a configuration example of a similar specification search unit included in the search device of the second embodiment.

Fig. 31 is a diagram showing a configuration example of a similarity learning unit included in the retrieval device of the third embodiment.

Fig. 32 is a flowchart showing an example of the operation of the similarity learning unit included in the retrieval device of the third embodiment.

Fig. 33 is a diagram showing an example of the definition of the selection of used parts used by the similarity learning unit of the retrieval device of the third embodiment.

Fig. 34 is a diagram showing the concept of the specification similarity learning process executed by the similarity learning unit of the retrieval apparatus of the third embodiment.

Fig. 35 is a diagram showing an example of the configuration of the search device of the fourth embodiment.

Fig. 36 is a flowchart showing an example of the search operation performed by the search device of the fourth embodiment.

Fig. 37 is a diagram showing an example of a specification input screen displayed by the search device of the fourth embodiment.

Fig. 38 is a diagram showing a configuration example of a similar specification search unit included in the search device of the fourth embodiment.

Fig. 39 is a diagram showing an example of expanded similarity weight data used by the similar specification search unit of the search device of the fourth embodiment.

Fig. 40 is a flowchart showing an example of the operation of the similar specification search unit included in the search device of the fourth embodiment.

Fig. 41 is a diagram showing the concept of the calculation of the degree of similarity of the special specifications performed by the special-standard similarity calculation unit of the similar-standard search unit included in the search device of the fourth embodiment.

Fig. 42 is a diagram showing an example of search results of similar specifications displayed on the search result detail display section of the search device of the fourth embodiment.

Fig. 43 is a diagram showing a configuration example of a similar specification search unit included in the search device of the fifth embodiment.

Fig. 44 shows the operation of the similar specification search unit included in the search device of the fifth embodiment The flow chart of an example.

Fig. 45 is a diagram showing an example of item-specific keywords (keywords) used by the similar specification search unit of the search device of the fifth embodiment.

Fig. 46 is a diagram showing an example of search results retrieved by the search device of the fifth embodiment.

Hereinafter, the retrieval device, retrieval method, and machine learning device of the embodiment of the present invention will be described in detail based on the drawings. In addition, the present invention is not limited to the following embodiments.

Implementation mode 1.

Fig. 1 is a diagram showing an example of the configuration of the search device according to the first embodiment of the present invention. The search device 1 of the first embodiment includes a control unit 11, an input unit 12, a product design database 13, a similarity learning unit 14, a similar specification search unit 16, and a search result display unit 17.

The control unit 11 controls the overall processing executed by the search device 1, and in this embodiment, controls the input unit 12, the similarity learning unit 14, the similarity standard search unit 16, and the search result display unit 17.

The input unit 12 is provided for the user to input information on the specifications of the product as a condition for retrieving necessary data from the data held in the product design database 13. The input unit 12 acquires information on the specifications of the product input by the user. The information on the specifications of the product acquired by the input unit 12 is assumed to display the information on the specifications of the newly designed product.

The product design database 13 holds data related to the specifications of products designed in the past. The details of the data held in the product design database 13 will be described later.

The similarity learning unit 14 analyzes the data held in the product design database 13 and learns the similarity between the item values of the product specifications, and generates the similarity data 15 used in the search process of the similar specification search unit 16. The so-called product specification item value refers to the value of each item showing the product specification. The similarity learning unit 14 constitutes a machine learning device.

The similar specification retrieval unit 16 is based on the similarity data 15 that is the learning result learned by the similarity learning unit 14, and retrieves the information displayed by the display and input unit 12 from the specification data group held in the product design database 13 Specification data of similar specifications.

The search result display unit 17 displays the search results searched by the similar specification search unit 16.

In addition, in the present embodiment, the description is provided on the premise that the product design database 13 is present inside the search device 1, but the product design database 13 may also be configured to exist outside the search device 1. For example, it may be configured such that the search device 1 and the product design database 13 are connected via a communication network (network), and the search device 1 obtains necessary data from the product design database 13 via the communication network.

Fig. 2 is a diagram showing an example of the hardware configuration of the search device of the first embodiment. The retrieval device 1 is realized by the processing circuit 101, the memory device 102, the input device 103, and the display device 104 shown in FIG. 2.

The processing circuit 101 is a circuit that performs various processing in the search device 1, such as a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, and an ASIC (Application Specific Integrated Circuit; specific Application of integrated circuit), FPGA (Field-Programmable Gate Array; Field-Programmable Gate Array), or a circuit formed by combining the above circuits. Memory device 102 is hard disk drive, flash memory (flash memory) and so on. The input device 103 is a keyboard, a mouse, a touch panel, etc., as long as it is a machine capable of inputting information required for the operation of the retrieval device 1 there is no limitation. The display device 104 is a display, a liquid crystal monitor, and the like. The display device 104 is not limited as long as it allows the user to refer to the information input using the input device 103, or the search result retrieved by the search unit 16 with similar specifications, and may be a projector or the like.

The control unit 11, the similarity learning unit 14, and the similar specification retrieval unit 16 of the search device 1 are realized by a processing circuit 101. The product design database 13 is realized by the memory device 102. In addition, the storage device 102 also holds similarity data 15, information input by the user via the input unit 12, and search results retrieved by the similar specification search unit 16, and the like. The input unit 12 is realized by the input device 103, and the search result display unit 17 is realized by the display device 104.

In addition, all or a part of the control unit 11, the similarity learning unit 14, and the similar specification search unit 16 of the search device 1 can also be realized by a general-purpose processor and memory. At this time, the hardware configuration of the search device 1 is as shown in FIG. 3. Fig. 3 is a diagram showing another example of the hardware configuration of the search device of the first embodiment.

The processor 201 shown in Figure 3 is a CPU (Central Processing Unit, also known as a central processing unit, processing device, arithmetic device, microprocessor, microcomputer, DSP (Digital Signal Processor; digital signal processor) Processor)), system (system) LSI (Large Scale Integration; large integrated circuit), etc. Memory 202 is RAM (Random Access Memory), ROM (Read Only Memory), EPROM (Erasable Programmable Read Only Memory; Erasable Programmable Read Only Memory), EEPROM (Registrar Mark) (Electrically Erasable Programmable Read Only Memory; Electronically Erasable Programmable Read Only Memory), etc. The memory device 102, the input device 103, and the display device 104 shown in FIG. 3 are the same as the memory device 102, the input device 103, and the display device 104 shown in FIG.

The hardware system shown in Figure 3 can also be a tablet PC (Personal Computer), desktop PC or server-client server and client PC. When the control unit 11, the similarity learning unit 14, and the similar specification retrieval unit 16 are realized by the processor 201 and the memory 202 shown in Fig. 3, the processor 201 will be used as the control unit 11 and the similarity learning The functions of the part 14 and the similar specification search part 16 are stored in the memory 202 and the like in advance. In addition, the processor 201 reads and executes the program stored in the memory 202, so that the control unit 11, the similarity learning unit 14, and the similar specification retrieval unit 16 are realized.

In addition, the retrieval devices of the second and subsequent embodiments described later are also realized by hardware having the configuration shown in FIG. 2 or FIG. 3.

Hereinafter, the operation of the search device 1 of Embodiment 1 will be described. First, the specification similarity learning process performed by the similarity learning unit 14 will be described, and then, the search operation performed by the search device 1 will be described. The following is a description of an example of the search operation when designing a motor. However, the present invention can be applied to a product having data equivalent to the data held in the product design database 13 described later.

(The specification similarity learning process performed by the similarity learning unit 14)

FIG. 4 is a diagram showing an example of the detailed configuration of the similarity learning unit 14 included in the search device 1 of the first embodiment. The similarity learning section 14 that constitutes the machine learning device is equipped with parts for use The extraction unit 41 and the similarity calculation unit 42. The used parts extraction unit 41 extracts the used parts list for each specification item value from the product design database 13. The similarity calculation unit 42 calculates the similarity between the item values based on the used parts list for each specification item value extracted by the used parts extraction unit 41.

FIG. 5 is a diagram showing a configuration example of the product design database 13 provided in the search device 1 of the first embodiment. The product design database 13 contains: a specification list 131 that lists the specifications of products designed in the past; and a part composition table 132 that records the part composition of each product. The specification list 131 is equivalent to the specification data. An example of the specification list 131 is shown in FIG. 6, and an example of the parts configuration table 132 is shown in FIG. 7.

Fig. 6 is a diagram showing an example of a list of specifications held by the product design database 13 of the first embodiment. The specification list 131 contains at least: the model ID (Identification; identification number) 601 used to identify each product; the series name 602 belonging to the name of the product group that aggregates specific products with similar specifications; the products of each series Rated output 603; rated rotation speed 604 of products of each series; brake 605 showing whether the products of each series have an electromagnetic brake (brake); and special specifications 606 showing the special specifications of products of each series. In the special specification 606, natural sentences, that is, text data, are stored. The series with the special specification 606 as "-" means that no text data showing the special specification is stored. In fact, although the motor has many specifications such as the maximum rotation speed, the presence or absence of an oil seal, and the shape of the shaft end, the following is for the convenience of explanation, and only the 6 types of items shown in Figure 6 are used except for special items. 5 types of items other than the standard 606. In addition, there is no restriction on the type of items used for the search as long as there is one or more items. In the following, products belonging to the same series, that is, the entire product group to which the same model ID is assigned, may be referred to as a model.

Fig. 7 is a diagram showing an example of the parts composition table held by the product design database 13 of the first embodiment. The part composition table 132 contains at least: a model ID 601 for identifying each product, a part hierarchy 701, a part ID 702 for identifying each part, a part name 703, and a number 704 showing the number of parts used in a product. The parts configuration table 132 is information on the constituent parts of each product corresponding to the model ID 601. The model ID 601 included in the parts configuration table 132 is the same as the model ID 601 included in the specification list 131. It is assumed that the model ID 601 is used as a code common to the specification list 131 and the parts configuration table 132. The part hierarchy 701 shows which hierarchy the part belongs to when the part is managed in a hierarchical manner. For example, when a product is composed of plural types of first parts, and a certain first part is composed of plural second parts, the part hierarchy 701 of the first part becomes "1", and the part hierarchy 701 of the second part becomes " 2". In addition, when the second part is composed of the third part, the part hierarchy 701 of the third part becomes "3". In fact, the motor also contains other items such as materials and dimensions, but for the sake of simplicity of description, only the five types of items shown in Fig. 7 are used below.

Fig. 8 is a flowchart showing an example of the operation of the similarity learning unit 14 provided in the search device 1 of the first embodiment, and is a program showing the specification similarity learning process. For example, when the search device 1 accepts an operation instructing the start of specification similarity learning from the user, the similarity learning unit 14 starts the specification similarity learning process shown in FIG. 8.

When the similarity learning unit 14 starts the specification similarity learning process, it first executes the specification list acquisition process (step S11). Specifically, the used parts extraction unit 41 obtains the specification list 131 of the content shown in FIG. 6 from the product design database 13.

Next, the similarity learning unit 14 repeats the processing of step S12 and step S13 for each specification item included in the specification list 131. Specifically, the similarity learning department 14 sets needles For each specification item of series name 602, rated output 603, and rated rotation speed 604, the processing of step S12 and step S13 is repeated. At this time, the similarity learning unit 14 repeats the processing of step S12 for each specification item value belonging to the specification item. For example, in a case where the specification item values a1, a2, and a3 belong to the specification item A, the similarity learning unit 14 performs the processing of step S12 and step S13 with the specification item A as the target, and responds to the specification item values a1, a2. Each of and a3 repeats the processing of step S12. The similarity learning unit 14 extracts the parts corresponding to the specification item values from the parts composition table 132 in step S12. The similarity learning unit 14 calculates the similarity between item values in step S13 using the method described later. Step S12 is a process executed by the used part extraction unit 41, and step S13 is a process executed by the similarity calculation unit 42.

For example, the series name 602 in the specification list 131 shown in Fig. 6 includes A series, B series, C series, D series,... Therefore, the similarity learning unit 14 performs the processing of step S12 and step S13 for the series name 602, and repeats the processing of step S12 for A series, B series, C series, D series, ... The extracted parts are then used to calculate the similarity between the item values.

Here, in the case of the processing of step S12 for the A series, the A series is a product with an organic species ID of "1" and a product with a model ID of "2". Therefore, the similarity learning unit 14 acquires all the parts used in the product with the model ID "1" and all the parts used in the product with the model ID "2" from the parts configuration table 132 shown in FIG. 7. For example, parts with part IDs of PA001, PA002, PA003, PA004, ... belong to the parts used in products whose model ID is "1".

In addition, the rated output 603 in the specification list 131 shown in FIG. 6 includes 3.5 kW, 2.0 kW, 1.5 kW, 400 W,... Therefore, the similarity learning department 14 department In the processing of step S12 and step S13 for the rated output 603, for 3.5kW, 2.0kW, 1.5kW, 400W,..., the processing of step S12 is executed to extract the corresponding parts, and then the extracted parts are used The similarity between the calculated item values of the parts.

Here, in the case of the processing of step S12 for 2.0kW, the product with a rated output of 2.0kW is a product with a product type ID of "2", a product with a model ID of "3", and a product with a model ID of " 4" products. Therefore, the similarity learning unit 14 obtains all the parts used in the product with the model ID "2", all the parts and the model ID used in the product with the model ID "3" from the parts configuration table 132 shown in Fig. 7 All parts used in the product of "4".

In step S13, the similarity calculation unit 42 calculates the similarity between the displayed item values based on the data of the used parts obtained by executing step S12 on the value of each specification item included in the specification item of the processing target by the used part extraction unit 41 That is, the value of how similar the values of the specification items are to each other is calculated. Hereinafter, the value calculated by the similarity calculation unit 42 is referred to as the degree of similarity. The similarity calculation unit 42 is a value belonging to SimSpecValue (Va, Vb) between item values that calculates the similarity between the specification item value Va and the specification item value Vb in accordance with the following formula (1).

Here, Parts(V) is a collection of parts used by all models that have V as the value of the specification item. In addition, Count(P) is the number of types of parts in the parts set P. FIG. 9 is a diagram showing the concept of the calculation method of the similarity between the item values of the similarity learning unit 14 included in the search device 1 of the first embodiment. For example, suppose the parts used in the A series are PA001, There are 90 types of PA002, ..., PA090, and 80 types of parts used in the B series models are PB001, PB002, ..., PB080. When 70 types of parts are common, it becomes SimSpecValue ("A series", "B series")=70/(90+80-70)=0.7.

Although it is an example of calculating the similarity between item values for the A series and the B series, when there are other series besides the A series and the B series, the calculation is performed for the combination of all series in step S13. Similarity between item values. For example, when there is a C series in addition to the A series and the B series, the similarity calculation unit 42 calculates the similarity between the item values of the A series and the B series, the similarity between the item values of the A series and the C series, and the B series The similarity between the item values of the C series.

In step S13, the similarity calculation unit 42 generates and outputs the similarity data 15 containing all the calculated similarities between the item values after calculating the similarity between the item values for all combinations of the specification item values.

As described above, the similarity learning unit 14 analyzes the specification list 131 and the part composition table 132 belonging to the specification data group held in the product design database 13, and learns all combinations of specification item values for each of the specification items that display the specifications of the product. To what extent the values of two specification items of the same specification item are similar to each other. That is, the similarity learning unit 14 learns how similar two specification item values of the same specification item are to each other for all combinations of specification item values for the specification data group held by the product design database 13.

FIG. 10 is a diagram showing an example of the similarity data 15 generated by the similarity learning unit 14 included in the search device 1 of the first embodiment. The similarity data 15 series contains at least: specification items, specification item value #1, specification item value #2, and similarity. In addition, in the following example, the similarity between the value of the specification items is based on a one-to-one relationship, that is, it is used for a single specification item However, as shown in FIG. 11, the similarity calculation unit 42 may also calculate the same similarity between a combination of a plurality of specification items to generate similarity data 15 as shown in FIG. 11. Fig. 11 shows an example of the similarity data 15 for the combination of the series name and the rated output, and the similarity for the combination of the rated output and the rated rotation speed.

(Search operation by search device 1)

Next, the search operation performed by the search device 1 will be described. Fig. 12 is a flowchart showing an example of the search operation performed by the search device 1 of the first embodiment. For example, when the retrieval device 1 receives an operation instructing the start of retrieval processing from the user, the retrieval device 1 starts the operation according to the flowchart shown in FIG. 12.

When the search device 1 starts the search process, the input unit 12 first accepts specification input, that is, the specification item value input by the designer whose reason belongs to the user is used as the search condition (step S21). In this step S21, the search device 1 displays, for example, the specification input screen shown in FIG. 13 on the display portion omitted from the description in FIG. 1, and waits for the user to input the specification item value. In addition, the display unit is realized by the display device 104 shown in FIG. 2 and FIG. 3. The specification input screen shown in Fig. 13 includes an input column 1301 for the series name, an input column 1302 for the rated output, an input column 1303 for the rated rotation speed, and an input column 1304 for the presence or absence of brakes. In the example of the specification input screen shown in FIG. 13, it is configured to input the specification item value by selecting a pull-down menu, but it may be configured to input by any method. It can also automatically set the value of the specification item based on the order data by cooperating with other systems such as business systems.

When the acceptance of the input of the specification item value by the input unit 12 is completed, the similar specification search unit 16 then performs a similar specification search (step S22). That is, similar specifications The search unit 16 searches the product design database 13 based on the similarity data 15 and the specification item value input in step S21, and finds a model with a specification similar to the specification displayed by the specification item value input in step S21. Hereinafter, the details of the search processing performed by the similar specification search unit 16 will be described.

FIG. 14 is a diagram showing an example of the detailed configuration of the similar specification search unit 16 included in the search device 1 of the first embodiment. The similar specification retrieval unit 16 includes a specification similarity calculation unit 61 and a data aligning unit 62.

The specification similarity calculation unit 61 is based on the similarity data 15 generated by the similarity learning unit 14 and the similarity weight data 65, for each model stored in the work-in-process design database 13, that is, the specifications of the past models, and calculate and compare The similarity of the specifications shown in the input specification data 63 output by the input unit 12. The input specification data 63 is data containing the value of each specification item received by the input unit 12 in step S21. The details of the similarity weight data 65 are explained separately.

The data aligning part 62 rearranges the calculation result of the specification similarity calculation part 61, that is, the similarity of the specifications of each past model in the order of similarity, and outputs the similarity of the specifications of each past model after the rearrangement as Similar specifications search result 64.

Fig. 15 is a diagram showing an example of similarity weight data used by the similar specification search unit 16 of the search device 1 of the first embodiment. The similarity weight data 65 contains the specification item 1501 and the similarity weight 1502. The similarity weight data 65 of the example shown in Fig. 15 contains the series name, rated output, rated rotation speed, and each of the brakes included in the specification list 131 of the product design database 13 that belong to the specifications. The similarity weight of the weight coefficient. These similarity weights are used when calculating the similarity of the specifications described later. About each regulation The weighting system of grid items is manually defined. For example, it is configured that the search device 1 has a function of creating the similarity weight data 65, and the designer uses this function to create the similarity weight data 65. The designer may also use an external device, for example, a personal computer (personal computer) to create the similarity weight data 65. It may also be configured that when the operation of the retrieval device 1 is started, the retrieval result is evaluated by a test operation using the temporarily set similarity weight, and the value of each similarity weight is adjusted according to the evaluation result.

Fig. 16 is a flowchart showing an example of the operation of the similar specification search unit 16 provided in the search device 1 of the first embodiment, and shows a program of similar specification search processing. For example, when the search device 1 receives an operation instructing the start of search for similar specifications from the user, the similar specification search unit 16 starts the similar specification search processing shown in FIG. 16.

When the similar specification search unit 16 starts the similar specification search process, it first executes the specification list acquisition process (step S31). Specifically, the specification similarity calculation unit 61 obtains the specification list 131 from the product design database 13.

Next, the similar specification search unit 16 repeats the process of step S32 for each past model included in the specification list 131. For example, when models with model IDs 1 to 5 are included in the specification list 131, the similar specification search unit 16 executes the processing of step S32 for each of the 5 types of past models with model IDs 1 to 5, and calculates the items described later. The similarity of different specifications.

In step S32, the specification similarity calculation unit 61 of the similar specification retrieval unit 16 calculates the similarity between the specifications of the past model stored in the product design database 13 and the input specification data 63 based on the similarity data 15 and the similarity weight data 65 . Specifically, the specification similarity calculation unit 61 calculates the similarity SimSpec (Sinp, Smx) between the input specification Sinp and the specification Smx of the model mx in accordance with the following formula (2).

Here, Sim(Sinp(s), Smx(s)) is a value determined based on whether the sth specification item Sinp(s) and Smx(s) among the 4 types of specification items included in the specification list 131 are consistent. . Sim(Sinp(s), Smx(s)) is "1" when Sinp(s) and Smx(s) are the same, and when they are inconsistent, it becomes the similarity degree contained in the similarity data 15 shown in Figure 10 Value. In addition, Wdef(s) is the value of the similarity weight defined in the similarity weight data 65 corresponding to the s-th specification item.

FIG. 17 is a diagram showing an example of a calculation result calculated by the specification similarity calculation unit 61 of the similar specification search unit 16 included in the search device 1 of the first embodiment. The example shown in Fig. 17 shows that when the specifications of the content shown in Fig. 13 are input in the aforementioned step S21, the specification similarity calculation unit 61 calculates the model ID with the specification list 131 shown in Fig. 6 as "1 "," "2", "3" of the similarity of each model of the SimSpec results.

In the specifications shown in Figure 13, the series name is "A series", the rated output is "2.0kW", the rated rotation speed is "2000r/min", and the brake is "None". Therefore, for example, according to the content shown in Figure 6, the series name, rated rotation speed, and brake specifications of the past model family with the model ID "1" are consistent, and the similarity of these specification items becomes "1". On the other hand, the specifications of the rated output do not match, and the similarity degree of the rated output becomes the similarity degree "0.05" between the rated output 3.5kW and 2.0Kw included in the similarity data 15 shown in Fig. 10.

In the example of the calculation result shown in Figure 17, the similarity degree SimSpec (Sinp, Sm1) of the past model with the model ID "1" becomes 0.83, and the similarity degree SimSpec (Sinp, Sm1) of the past model with the model ID "2" Sm2) becomes 0.75 and the past model with model ID "3" The species similarity SimSpec (Sinp, Sm3) becomes 0.89. Although the series name of the past model with the model ID "3" is not different from the series name input in the aforementioned step S21, the effect of the aforementioned similarity data 15 becomes the highest similarity among the three models.

Returning to the description of Fig. 16, after the specification similarity calculation unit 61 performs the processing of step S32 on all past models, the data alignment unit 62 uses the similarity calculation result SimSpec(Sinp) calculated by the specification similarity calculation unit 61 , Smx) rearrange (step S33). That is, the data arranging unit 62 rearranges the similarity calculation result SimSpec (Sinp, Smx) in a manner of arranging the higher similarity in order.

Returning to the description of Fig. 12, when the similar specification search process of step S22 by the similar specification search unit 16 ends, the search result display unit 17 displays the search results of similar specifications, and presents the search results to the designer (step S23) . Fig. 18 is a diagram showing an example of search results of similar specifications displayed by the search result display unit 17 of the search device 1 of the first embodiment. Fig. 18 shows an example in which the search results of similar specifications are displayed together with the specifications input in step S21. In the example shown in Figure 18, the model ID of the past model and the similarity score (score) belonging to the aforementioned similarity SimSpec (Sinp, Smx) are displayed in order of the similarity score. In addition, when the number of models in the past is large, the search result display unit 17 may also extract and display those with a similarity score greater than a preset value, or it may be configured to sequentially extract the preset number from the one with the highest similarity score. Of the past models to display. The designer appropriately selects a model similar to the required specifications based on the search results displayed on the search result display unit 17, and designs a new model that meets the required specifications.

In addition, the retrieval device 1 can also be configured to indicate why the model was retrieved when the retrieval result retrieved by the retrieval unit 16 with similar specifications is displayed, and it will be used as the basis for the retrieval result. The search score is displayed in detail. Fig. 19 is a diagram showing a first modification of the search device of the first embodiment. The retrieval device 1a shown in FIG. 19 is obtained by replacing the retrieval result display unit 17 of the retrieval device 1 shown in FIG. 1 with the retrieval result detail display unit 18. The configuration of the search device 1a other than the search result detail display unit 18 is the same as that of the search device 1.

FIG. 20 is a flowchart showing an example of a search operation performed by the search device 1a of the first modification of the first embodiment. The flowchart shown in FIG. 20 is obtained by replacing step S23 in the flowchart shown in FIG. 12 with step S24. Step S23 is a process executed by the search result detail display unit 18. That is, in step S23, the search result detail display unit 18 performs display based on the search result output from the similar specification search unit 16, and provides the search result of the similar specification to the designer. The search result detail display unit 18 displays, for example, search results of similar specifications to the configuration shown in FIG. 21. The example shown in FIG. 21 is configured to display the details of the similarity score equivalent to the search score in addition to the content of the similar specification search result shown in FIG. 18. The search result detail display unit 18 is equivalent to a search result display unit that displays the search result retrieved by the similar specification search unit 16 and the similarity corresponding to the search result together.

In addition, it is also possible to easily change the search conditions for obtaining the search results desired by the designer by configuring the search device 1 to have search weight changing means for changing the weight of the similarity of the specifications. Fig. 22 is a diagram showing a second modification of the search device of the first embodiment. The retrieval device 1b shown in FIG. 22 is obtained by adding a weight data changing unit 21 belonging to the retrieval weight changing means to the retrieval device 1a shown in FIG. 19. The configuration other than the weight data changing unit 21 of the search device 1b is the same as that of the search device 1a.

FIG. 23 is a flowchart showing an example of the search operation performed by the search device 1b according to the second modification of the first embodiment. The flowchart shown in Figure 23 is shown in Figure 20 Step S25 is added after step S24 in the flowchart of FIG.

In step S25, the weight data changing unit 21 accepts the operation of changing the similarity weight from the user. At this time, the weight data changing unit 21 displays, for example, the similarity weight data change screen shown in FIG. 24 on the display unit (not shown) in FIG. 22, and waits for the user to perform an operation for correcting the similarity weight data. Figure 24 shows an example when the similarity weight is changed to emphasize the consistency of series names. Specifically, Fig. 24 shows an example when the similarity weight of the series name is changed from 2.0 to 5.0, and the similarity weight of the rated rotation speed is changed from 1.5 to 1.0.

When the change operation of the similarity weight data is accepted in step S25, the search device 1b executes step S22 using the changed similarity weight data to perform a new search, and displays the result of the new search in step S23. Figure 25 is a diagram showing an example of the result of a re-search after changing the similarity weight data. By changing the similarity weight in the above-mentioned step S25, the past model with the same series name, such as the intention of the designer, becomes the upper candidate.

As described above, the retrieval device of this embodiment is equipped with: a similarity learning unit, which learns the similarity between the item values of the specifications of the past model of the product designed in the past, based on the related data of the product designed in the past; and retrieval of similar specifications The part is based on the similarity data that belongs to the result of learning by the similarity learning part, and retrieves the specification data showing the specifications similar to the specifications specified by the designer. In this way, the designer does not need to create complicated rules such as the knowledge base described in Patent Document 1 in advance, and it is possible to reduce the workload of the designer in searching for the reused data. In addition, it is possible to realize an efficient search that does not rely on the experience of the designer.

In addition, when the search device of this embodiment adopts the configuration shown in Fig. 19 and is configured to display the details of similarity scores in addition to the search results of similar specifications, the basis of the search results can be reported to the designer.

In addition, the search device of this embodiment adopts the configuration shown in Fig. 22 to display the details of similarity scores in addition to the search results of similar specifications, and to accept changes in the weighting of the similarity of each specification item , You can change the search conditions in a way that the designer expects the search results.

Implementation form 2.

Fig. 26 is a diagram showing an example of the configuration of the search device of the second embodiment. The search device 1c of the second embodiment replaces the similar specification search unit 16 of the search device 1 (see FIG. 1) of the first embodiment with a similar specification search unit 16c and adds a similarity adjustment unit 31. In the search device 1c, the similar specification search unit 16c uses the adjusted similarity data generated by the similarity adjustment unit 31 to perform search processing. The configuration other than the similarity adjustment unit 31 and the similar specification search unit 16c of the search device 1c is the same as that of the search device 1, so the same reference numerals as those in FIG. 1 are assigned and their descriptions are omitted.

The basic operation when the search device 1c searches for past models with specifications similar to those specified by the designer is the same as that of the search device 1 (refer to FIG. 12). The retrieval device 1c is different from the retrieval device 1 in the content of a part of the specification similarity learning process.

Fig. 27 is a flowchart showing an example of the specification similarity learning process of the retrieval device 1c of the second embodiment. The flowchart shown in Fig. 27 is obtained by adding step S41 to the flowchart shown in Fig. 8. The steps other than step S41 are the same as the specification similarity learning processing described in the first embodiment, so the same step numbers as the corresponding processing in FIG. 8 are assigned and the description thereof will be omitted.

In step S41, the similarity between the item values is adjusted. This adjustment is an adjustment made to the similarity data 15 that is the result of learning by the similarity learning unit 14. The calculation is performed using the similarity adjustment unit 31. When step S41 is executed, the similarity adjustment unit 31 causes, for example, the similarity adjustment screen between item values shown in FIG. 28 to be displayed on a display unit that is omitted from the description in FIG. 26. The similarity adjustment unit 31 waits for the designer to perform an operation in a state where the similarity adjustment screen between item values is displayed on the display unit. When the similarity adjustment is required, the designer enters the adjusted similarity for the similarity to be adjusted, and when the input is completed, press the save button to end the operation. When the adjustment of the similarity is not required, the designer does not input the adjusted similarity and presses the save button or the cancel button to end the operation. In addition, Fig. 28 shows an example when the user adjusts the similarity between the A series and the B series and the similarity between the A series and the C series.

The similarity adjustment unit 31 outputs the similarity between all item values including the adjusted similarity as the adjusted similarity data 32 when the operation of adjusting the similarity by the designer is completed. Figure 29 is a diagram showing an example of the similarity data 32 after adjustment. The adjusted similarity data 32 is a structure in which the adjusted similarity is added to the similarity data 15 (refer to FIG. 10) described in the first embodiment.

Fig. 30 is a diagram showing a configuration example of a similar specification search unit 16c provided in the search device 1c of the second embodiment. The similar specification search unit 16c is obtained by changing the specification similarity calculation unit 61 of the similar specification search unit 16 (refer to FIG. 14) described in the first embodiment to the specification similarity calculation unit 61c. The specification similarity calculation unit 61c calculates the specifications of the past model stored in the product design database 13 and the input output from the input unit 12 based on the adjusted similarity data 32 and the similarity weight data 65 generated by the similarity adjustment unit 31 The similarity of specification data 63. At this time, the specification similarity calculation unit 61c uses the adjusted similarity included in the adjusted similarity data 32 to perform calculations. The similarity calculation method performed by the specification similarity calculation unit 61c is the same as The similarity calculation method performed by the specification similarity calculation unit 61 is the same, so the description is omitted.

As described above, the retrieval device 1c of the present embodiment includes the similarity adjustment unit 31 that adjusts the similarity data 15 belonging to the result of learning by the similarity learning unit 14. This allows the designer to specify the similarity to be strengthened and the similarity to be suppressed, and the value of the similarity can be adjusted appropriately. In addition, according to the retrieval device 1c, even if the designer does not adjust the similarity data, he can individually confirm whether the value of the similarity data 15 has been properly learned.

Implementation mode 3.

Fig. 31 is a diagram showing a configuration example of the similarity learning unit 14d included in the retrieval device of the third embodiment. The structure other than the similarity learning unit 14d of the retrieval device of the third embodiment is the same as that of the retrieval device 1 of the first embodiment. Therefore, the description of the components other than the similarity learning unit 14d is omitted.

The similarity learning unit 14d is formed by using the component selection unit 43 in addition to the similarity learning unit 14 described in the first embodiment.

Fig. 32 is a flowchart showing an example of the operation of the similarity learning unit 14d provided in the retrieval device of the third embodiment. The flowchart shown in Fig. 32 is obtained by adding step S51 to the flowchart shown in Fig. 8. The processing other than step S51 is the same as each processing included in the flowchart shown in FIG. 8, so the same step numbers as the corresponding processing in FIG. 8 are assigned, and the description thereof will be omitted.

Step S51 is a part selection process for item values. In this step S51, the used part selection unit 43 selects the extraction used when the similarity calculation unit 42 calculates the similarity between the item values from the extraction results extracted by the used part extraction unit 41 based on the used part selection definition 44. result. Figure 33 is a diagram showing an example of using the part selection definition 44. As shown in Figure 33 As shown, the used part selection definition 44 contains at least a specification item 3301 and a related part number list 3302 corresponding to the specification item 3301. The use part selection definition 44 is created in advance by a designer, for example. The creation of the use part selection definition 44 may be performed using a device outside the search device, such as a personal computer, or it may be configured that the search device has a function of creating the use part selection definition 44, and the designer uses this function to make it.

In step S51, the part selection unit 43 is used to select the parts corresponding to the specification item values extracted in step S12, and there is a part number described in the related part number list 3302 associated with the corresponding specification item in the use part selection definition 44 Parts. The used part selection unit 43 outputs the selection result to the similarity calculation unit 42.

In step S13, the similarity learning unit 14d does not calculate the similarity between the item values using all the item values extracted by the part extraction unit 41, but calculates the item value based on the item values selected by the part selection unit 43. The similarity between.

Fig. 34 is a diagram showing the concept of the specification similarity learning process executed by the similarity learning unit 14d of the retrieval apparatus of the third embodiment. In the example shown in Figure 34, when the specifications of the computer type X and the model Y are similar, the similarity will vary depending on which of the specification item "rated output" and the specification item "rated rotation speed" is concerned. . That is, if we focus on the "rated output", since the part PA010 and the part PA011 are common, it can be said that the similarity is high. On the other hand, if you focus on the "rated rotation speed", because the model X is composed of parts PB020 and PB021, and the model Y is composed of parts PC020 and PC021, there is no commonality between the two, and it can be said that the similarity is low. The similarity learning unit 14d uses the part selection definition 44 to select the characteristic part groups for each specification item and then calculates the similarity, so that the calculation accuracy of the similarity data 15 can be improved.

As described above, in the retrieval device of the present embodiment, the similarity learning unit 14d includes the used part selection unit 43, and the used part selection unit 43 selects and issues characteristic parts for each specification item before calculating the value of the item. The similarity. In this way, the calculation accuracy of the similarity data can be improved.

Implementation mode 4.

Fig. 35 is a diagram showing an example of the configuration of the search device of the fourth embodiment. The search device 1e of the fourth embodiment replaces the similar specification search unit 16 of the search device 1a shown in FIG. 19 with a similar specification search unit 16e, and adds a special specification input unit 51. The configuration other than the special specification input unit 51 and the similar specification search unit 16e of the search device 1e is the same as that of the search device 1a, so the same reference numerals as those in FIG. 19 are assigned, and the description thereof will be omitted.

Fig. 36 is a flowchart showing an example of the search operation performed by the search device 1e of the fourth embodiment. The flowchart shown in FIG. 36 is obtained by adding step S61 to the flowchart shown in FIG. 20. The processing other than step S61 is the same as each processing included in the flowchart shown in FIG. 12, so the same step number as the corresponding processing in FIG. 20 is assigned and the description thereof is omitted.

In step S61, the special specification input unit 51 receives the input of the special specification from the designer. At this time, the search device 1e displays the specification input screen shown in FIG. 37 on the display unit (not shown) in FIG. 35. The special specification input unit 51 waits for the designer to operate in a state where the specification input screen is displayed on the display unit. The specification input screen shown in FIG. 37 is a configuration in which a special specification input field 3701 is added to the specification input screen shown in FIG. 13. The specification input screen in Fig. 37 is displayed at the beginning of step S21.

The designer inputs the series name, rated output, etc. to the retrieval device 1e, and also inputs special specifications. The special specification input unit 51 is a special specification input operation performed by the designer. When the work is over, data showing the content input to the input field 3701 of the special specification is generated and output to the similar specification search unit 16e.

Fig. 38 is a diagram showing a configuration example of the similar specification search unit 16e provided in the search device 1e of the fourth embodiment. The similar specification search unit 16e replaces the specification similarity calculation unit 61 of the similar specification search unit 16 (refer to FIG. 14) described in Embodiment 1 with a special specification similarity calculation unit 81 and an expanded specification similarity calculation unit 82. become.

The special specification similarity calculation unit 81 calculates the similarity between the input special specification data 83 belonging to the data output from the special specification input unit 51 and the special specification of the past model stored in the product design database 13. The expanded specification similarity calculation unit 82 calculates the specifications of the past model stored in the work-in-process design database 13 based on the similarity of the special specification calculated by the similarity calculation unit 81 of the special specification and the expanded similarity weight data 84, and the input unit 12 The similarity of the output input specification data 63.

Fig. 39 is a diagram showing an example of expanded similarity weight data 84. The expanded similarity weight data 84 contains the specification item 3901 and the similarity weight 3902. The expanded similarity weight data for the example shown in Figure 39 contains the series name, rated output, rated rotational speed, brake, and special specifications included in the specification list 131 of the product design database 13 belonging to the specification items. The similarity weight of the weight coefficient. The weighting system for each specification item is manually defined. It may also be configured to evaluate the retrieval result by a test operation using the temporarily set similarity weight when the operation of the retrieval device 1e is started, and then adjust the value of each similarity weight according to the evaluation result.

FIG. 40 is a flowchart showing an example of the operation of the similar specification search unit 16e provided in the search device of the fourth embodiment. The flowchart shown in Fig. 40 is the flow chart shown in Fig. 16 Step S32 of the flowchart is replaced with step S71 and step S72. The processing other than step S71 and step S72 is the same as each processing included in the flowchart shown in FIG. 16, so the same step number as the corresponding processing in FIG. 16 is assigned and the description is omitted.

Step S71 is a special specification similarity calculation process. In step S71, the special specification similarity calculation unit 81 calculates the similarity between the special specification of the past model stored in the product design database 13 and the input special specification data 83. Specifically, the special specification similarity calculation unit 81 calculates the similarity SpecialSimSpec (SSinp, SSmx) between the input special specification SSinp and the special specification SSmx of the model mx according to the following formula (3).

Here, SSinp(t1) and SSmx(t2) are the t1 and t2 texts when each special specification is separated by a comma (,). SSim(SSinp(t1), SSmx(t2)) is Is the edit distance between these texts. In addition, MAX(1-SSim(SSinp(t1), SSmx(t2))/L) shows the value of the left formula relative to the value of SSinp(t1) at the maximum and t2. In addition, L is the total number of characters in the texts SSinp(t) and SSmx(t), and n is the number of characters that will be entered when the special specification is separated by a comma (Japanese; equivalent to a comma in Chinese). number. In addition, the calculation of the edit distance between texts is performed using a well-known technique, for example, the technique disclosed in the patent document "International Patent Publication No. 2015/040793."

Figure 41 is a diagram showing the concept of a special specification similarity calculation performed by the special specification similarity calculation unit 81 of the similar specification search unit 16e included in the search device of the fourth embodiment. picture. As shown in Figure 41, the special specification similarity calculation unit 81 divides the text displayed in the input special specification data 83 and the text corresponding to the special specification of the past model mx with a dot, and calculates similarity for the divided text Spend. In the example shown in FIG. 41, the special-standard similarity calculation unit 81 divides each text into two parts, and calculates the similarity of the divided text. The special-standard similarity calculation unit 81 outputs the calculated average value of the similarity as the special-standard similarity.

Step S72 is the calculation processing of the similarity of the specifications of the expanded items. In step S72, the expanded specification similarity calculation unit 82 calculates and stores the WIP design database 13 based on the similarity degree of the special specification calculated by the special specification similarity calculation unit 81 in step S71 and the expanded similarity weight data 84 The similarity between the specifications of the previous model and the input specification data 63. Specifically, the extended specification similarity calculation unit 82 calculates the similarity ESimSpec (Sinp, SSinp, Smx, SSmx) between the input specification Sinp and the input special specification SSinp and the model mx specification Smx and the special specification SSmx according to the following formula (4) ).

Here, Sim(Sinp(s), Smx(s)) is the same as the description of the first embodiment, according to the sth specification item Sinp(s) and Smx(s) among the 4 types of specification items included in the specification list 131 ) Is a value that depends on whether it is consistent. That is, Sim(Sinp(s), Smx(s)) becomes "1" when Sinp(s) and Smx(s) are consistent, and becomes included in the similarity data 15 shown in Figure 10 when they are inconsistent. The value of similarity. In addition, EWdef(s) is the value of the similarity weight defined in the expanded similarity weight data 84 corresponding to the s-th specification item.

In step S33 in FIG. 40, the data aligning unit 62 rearranges the similarity calculation result ESimSpec (Sinp, SSinp, Smx, SSmx) calculated by the expanded specification similarity calculation unit 82. That is, the data arranging part 62 rearranges the similarity calculation results ESimSpec (Sinp, SSinp, Smx, SSmx) in order from the higher similarity.

The search result detail display unit 18 displays, for example, search results of similar specifications with the configuration shown in FIG. 42 on the display unit, and presents the search results to the designer (FIG. 36, step S24).

As described above, the similar specification search unit 16e of the search device of this embodiment is equipped with the special specification similarity calculation unit 81 that analyzes the special specification described in the natural sentence belonging to the text and calculates the similarity of the special specification, and the special specification is also considered. Search for past models with specifications similar to the specifications specified by the designer under the similarity degree. In this way, it is possible to search for past models with high similarity in consideration of special specifications, and the search accuracy can be improved.

Implementation mode 5.

FIG. 43 is a diagram showing a configuration example of a similar specification search unit 16f included in the search device of the fifth embodiment. The configuration other than the similar specification retrieval unit 16f of the retrieval device of the fifth embodiment is the same as that of the retrieval device 1e of the fourth embodiment. Therefore, the description of the components other than the similar specification search unit 16f is omitted.

The similar specification search unit 16f is obtained by adding a weight adjustment unit 85 to the similar specification search unit 16e of the fourth embodiment. The weight adjustment unit 85 adjusts and expands the similarity weight data 84 according to the input special specification data 83 and the item category keyword 86.

Figure 44 shows a search of similar specifications provided by the search device of the fifth embodiment A flowchart of an example of the operation of the part 16f. The flowchart shown in FIG. 44 is obtained by adding step S81 to the flowchart shown in FIG. 40. The processing other than step S81 is the same as each processing included in the flowchart shown in FIG. 40, and therefore the same step numbers as the corresponding processing in FIG. 40 are assigned and descriptions thereof will be omitted.

Step S81 is the similarity weight adjustment process. In step S81, the weight adjustment unit 85 confirms whether the input special specification data 83 contains the keyword registered in the item category keyword 86, and adjusts the value of the expanded similarity weight data 84 based on the confirmation result. The expanded similarity calculation unit 82 is input with the expanded similarity weight data 84 adjusted by the weight adjustment unit 85. The weight adjustment unit 85 adjusts the expanded similarity weight data 84 through a pre-defined procedure. For example, when the input special specification data 83 contains a keyword registered in the item category keyword 86, the weight adjustment unit 85 expands the similarity weight contained in the similarity weight data 84 and the input special specification data 83. The similarity weight of the specification item corresponding to the keyword is changed to a large value. An example of the process of changing the similarity weight to a large value by the weight adjustment unit 85 is a process of multiplying the similarity weight by a coefficient having a value greater than 1.

FIG. 45 is a diagram showing an example of item-specific keywords 86 used by the similar specification search unit 16f of the search device of the fifth embodiment. The item keyword 86 includes a specification item 4501 and a keyword list (list) 4502.

Fig. 46 is a diagram showing an example of search results retrieved by the search device of the fifth embodiment. The example shown in Figure 46 shows that the series name contained in the information input using the input unit 12 and the special specification input unit 51 is "B series", the rated output is "2.0kW", and the rated rotation speed is "1000r/min" , The brake is "Yes", the special specification is "brake torque special" and the keyword list 4502 is the search result shown in Figure 45.

In the above example, the special specification is "brake torque special", and the keyword list of the specification item "brake" of the item-specific keyword 86 shown in Fig. 45 has "brake torque special". Therefore, the search result shown in Fig. 46 is obtained by adjusting the similarity weight 3902 corresponding to the brake of the specification item 3901 of the expanded similarity weight data 84 shown in Fig. 39 to twice to calculate the similarity, and then rearranging the result. become. The reason for adjusting the similarity weight corresponding to the brake to 2 is to simplify the description. Can also be adjusted to other values.

The item-specific keywords 86 are created in advance by a designer, for example. The creation of the item-specific keywords 86 can be performed using a device outside the retrieval device, such as a personal computer, or it may be configured that the retrieval device has the function of creating the item-specific keywords 86 and the designer uses the function to create it.

As described above, the similar specification search unit 16f of the search device of the present embodiment includes the weight adjustment unit 85 that adjusts the similarity weight of the specification item based on the item-specific keyword and the text contained in the special specification. In this way, the similarity calculation is performed after the similarity weight is adjusted according to the description content of the special specification, so that past models with high similarity can be searched.

The configuration shown in the above embodiment only shows an example of the content of the present invention, and can also be combined with other known technologies, and a part of the configuration can be omitted or changed without departing from the spirit of the present invention.

1: Retrieval device

11: Control Department

12: Input section

13: Product design database

14: Similarity Learning Department

15: Similarity data

16: Similar specifications retrieval department

17: Search result display section

Claims (10)

A retrieval device that retrieves specification data similar to the retrieval condition from a specification data group that contains more than one specification item value showing the details of the item representing the specifications of the product. The aforementioned retrieval device includes: a similarity learning unit , Based on the aforementioned specification data group held in the product design database, learn how much the two specification item values of the same item are similar to each other for all combinations of specification item values; the input part is used as the aforementioned search condition Input of the value of the specification item; the similar specification retrieval unit searches for specification data similar to the input specification data composed of the specification item value received by the input by the aforementioned similarity learning unit based on the result of learning by the aforementioned similarity learning unit; And the similarity adjustment unit, the similarity adjustment unit adjusts the learning result learned by the similarity learning unit; the similar specification retrieval unit performs the search based on the learning result adjusted by the similarity adjustment unit. The search device described in the first item of the scope of patent application is further equipped with a search result display unit, and the search result display unit displays the search results obtained by the search unit with similar specifications. According to the search device described in the second item of the scope of patent application, the similar specification search unit calculates and displays the specifications accepted by the input unit for all the specification data held in the product design database based on the learning results. To what extent is the item value similar to the specification item value corresponding to the specification data held in the aforementioned product design database? The aforementioned search is performed based on the calculated similarity; the aforementioned search result display unit displays the search result retrieved by the aforementioned similar specification search unit together with the aforementioned similarity corresponding to the search result. As described in the first item of the scope of patent application, the similarity learning unit calculates and displays two specification item values based on the aforementioned specification data and the data of the constituent parts of each product corresponding to each of the aforementioned specification data The degree of similarity between the item values to which degree is similar to each other is used to perform the aforementioned learning; the aforementioned similar specification retrieval unit performs the aforementioned retrieval based on the degree of similarity between the aforementioned item values. According to the search device described in claim 4, the similar specification search unit weights the similarity between the item values, and performs the search for the similarity between the weighted item values. For example, the retrieval device described in item 5 of the scope of patent application further includes a weight data changing unit that changes the weight coefficient used when the similar specification retrieval unit weights the similarity between the item values. According to the search device described in claim 1, wherein the similarity learning unit performs the learning based on the specified specification item value among the specification item values contained in the specification data held in the product design database. For example, the search device described in any one of items 1 to 7 of the scope of patent application further has a special specification data acquisition unit, the aforementioned special specification data acquisition unit obtains text data showing the special specifications of the aforementioned product; the aforementioned similar specification search unit It is to retrieve the specification data similar to the aforementioned input specification data and the aforementioned special specification. According to the search device described in item 8 of the scope of patent application, the similar specification search unit adjusts the learning result based on the text data, and then performs the search based on the adjusted learning result. A retrieval method is a retrieval method executed by a retrieval device. The aforementioned retrieval device retrieves specifications similar to the retrieval conditions from a specification data group composed of more than one specification item value showing the details of the item representing the specifications of the product Data, the aforementioned retrieval method contains the following steps: the similarity learning step is based on the aforementioned specification data group held in the product design database, for the extent to which two specification item values of the same item are similar to each other, for all specification item values Combined learning; the similarity adjustment step is to adjust the learning result learned by the aforementioned similarity learning step; the input step is to accept the input of the value of the specification item used as the aforementioned search condition; and the similar specification search step is based on the aforementioned The adjusted learning result of the similarity adjustment step searches for specification data similar to the input specification data constituted by the specification item values received in the aforementioned input step.

Images