A process case retrieval method, device, equipment and medium

Abstract

The application discloses a process case retrieval method and device, equipment and medium, relates to the technical field of database, and solves the technical problem of low matching precision of the process case retrieved by the existing retrieval method. The method comprises the following steps: acquiring case word information to be retrieved; obtaining a similar case backup screening library and a same theory case backup screening library based on the case word information to be retrieved; performing clustering processing on the similar case backup screening library and the same theory case backup screening library based on keyword information to obtain a similar case output set; obtaining a same theory case output set based on TRIZ theory additional attribute tag information; performing quantity statistics on TRIZ theory additional attribute tag information of all process cases in the similar case backup screening library to obtain a recommended theory output set; and outputting the similar case output set, the same theory case output set and the recommended theory output set to obtain a target retrieval result. Therefore, the application can effectively improve product research and development efficiency and quality.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of database, and particularly relates to a process case retrieval method, device, equipment and medium. BACKGROUND

[0002] As the core of process design software, engineering knowledge plays a crucial role in the performance and generalization of the software. When developing and designing products or materials, developers often need a large amount of engineering knowledge as a research and development basis.

[0003] Due to the limitations of the current retrieval system, there is a lack of intelligent retrieval and recommendation from multiple angles, resulting in a situation that the retrieved process cases are messy and the matching accuracy is low, which affects the research and development quality and efficiency of the developers. SUMMARY

[0004] The main purpose of the present application is to provide a process case retrieval method, device, equipment and medium, which aims to solve the technical problem of low matching accuracy of the retrieved process cases by the existing retrieval method.

[0005] To achieve the above purpose, the present application provides a process case retrieval method, comprising the following steps:

[0006] Obtaining case term information to be retrieved; based on the case term information to be retrieved, obtaining a similar case backup screening library and a same theory case backup screening library; wherein the case term information to be retrieved includes keyword information and TRIZ theory additional attribute label information;

[0007] Based on the keyword information, the similar case backup screening library and the same theory case backup screening library are subjected to clustering processing to obtain a similar case output set;

[0008] Based on the TRIZ theory additional attribute label information, a same theory case output set is obtained; wherein the same theory case output set includes a plurality of first process cases, and each first process case contains all the TRIZ theory additional attribute label information;

[0009] The number of TRIZ theory additional attribute label information of all process cases in the similar case backup screening library is counted to obtain a recommended theory output set;

[0010] The similar case output set, the same theory case output set and the recommended theory output set are output to obtain a target retrieval result.

[0011] Optionally, based on the case term information to be retrieved, the similar case backup screening library and the same theory case backup screening library are obtained, comprising:

[0012] The case entry information to be retrieved is preprocessed to obtain first case entry information to be retrieved; wherein the preprocessing includes string segmentation processing and stop word removal processing;

[0013] Based on the first case entry information to be retrieved, a similar case backup screening library and a same theory case backup screening library are obtained.

[0014] Optionally, based on the keyword entry information, the similar case backup screening library and the same theory case backup screening library are subjected to clustering processing to obtain a similar case output set, including:

[0015] Based on the keyword entry information, a similar case keyword frequency value and a same theory case keyword frequency value are obtained;

[0016] Based on the similar case keyword frequency value, a similar case feature vector is obtained; and based on the same theory case keyword frequency value, a same theory case feature vector is obtained;

[0017] The similar case feature vector and the same theory case feature vector are subjected to clustering processing to obtain a similar case output set.

[0018] Optionally, based on the keyword entry information, a similar case keyword frequency value and a same theory case keyword frequency value are obtained, including:

[0019] Based on the keyword entry information, a similar case total word number value and a similar case keyword entry times value are obtained by preprocessing the process cases in the similar case backup screening library;

[0020] Based on the similar case total word number value and the similar case keyword entry times value, a similar case keyword frequency value is obtained;

[0021] Based on the keyword entry information, a same theory case total word number value and a same theory case keyword entry times value are obtained by preprocessing the process cases in the same theory case backup screening library;

[0022] Based on the same theory case total word number value and the same theory case keyword entry times value, a same theory case keyword frequency value is obtained.

[0023] Optionally, the similar case feature vector and the same theory case feature vector are subjected to clustering processing to obtain a similar case output set, including:

[0024] The keyword entry information is converted into an initial feature vector;

[0025] Input the initial feature vector, the similar case feature vector and the same theory case feature vector into a preset clustering model, perform clustering processing, and obtain a similar case output set.

[0026] Optionally, the obtaining of the same theory case output set based on the TRIZ theory additional attribute tag information comprises:

[0027] obtaining a first same theory case output set based on the TRIZ theory additional attribute tag information; wherein the first same theory case output set comprises a plurality of first process cases.

[0028] performing descending arrangement on all the first process cases in the first same theory case output set based on the publishing time of the first process cases, to obtain the same theory case output set.

[0029] Optionally, the obtaining of the recommended theory output set based on the number of TRIZ theory additional attribute tag information of all the process cases in the similar case standby screening library comprises:

[0030] counting the number of TRIZ theory additional attribute tag information of all the process cases in the similar case standby screening library, to obtain each first number value.

[0031] performing descending arrangement based on the first number value, and taking the TRIZ theory additional attribute tag information arranged in the top N as recommended theories, to obtain the recommended theory output set; wherein 1≤N≤5.

[0032] To solve the above technical problems, the embodiments of the present application further provide a process case retrieval device, comprising:

[0033] an initial screening module configured to acquire case term information to be retrieved, and obtain a similar case standby screening library and a same theory case standby screening library based on the case term information to be retrieved; wherein the case term information to be retrieved comprises keyword information and TRIZ theory additional attribute tag information;

[0034] a first target obtaining module configured to perform clustering processing on the similar case standby screening library and the same theory case standby screening library based on the keyword information, and obtain a similar case output set;

[0035] a second target obtaining module configured to obtain a same theory case output set based on the TRIZ theory additional attribute tag information; wherein the same theory case output set comprises a plurality of first process cases, and each first process case contains all the TRIZ theory additional attribute tag information;

[0036] a third target obtaining module, configured to count the number of TRIZ theory additional attribute tag information of all process cases in the similar case backup screening library, and obtain a recommended theory output set;

[0037] a target output module, configured to output the similar case output set, the same theory case output set and the recommended theory output set, and obtain a target search result.

[0038] To solve the above technical problems, the embodiments of the present application further provide an electronic device, which comprises a memory and a processor, the memory stores a computer program, and the processor executes the computer program to realize the method as described above.

[0039] To solve the above technical problems, the embodiments of the present application further provide a computer readable storage medium, which stores a computer program, and a processor executes the computer program to realize the method as described above.

[0040] The beneficial effects that can be achieved by the present application are as follows:

[0041] The process case retrieval method provided in the embodiments of the present application comprises: obtaining case word information to be retrieved; obtaining a similar case backup screening library and a same theory case backup screening library based on the case word information to be retrieved; wherein the case word information to be retrieved comprises keyword information and TRIZ theory additional attribute label information; performing clustering processing on the similar case backup screening library and the same theory case backup screening library based on the keyword information, to obtain a similar case output set; obtaining a same theory case output set based on the TRIZ theory additional attribute label information; wherein the same theory case output set comprises a plurality of first process cases, and each of the first process cases contains all the TRIZ theory additional attribute label information; counting the number of TRIZ theory additional attribute label information of all process cases in the similar case backup screening library, to obtain a recommended theory output set; and outputting the similar case output set, the same theory case output set and the recommended theory output set, to obtain a target retrieval result. That is, the keyword information and the TRIZ theory additional attribute label information obtained are used to preliminarily screen process cases in a process case knowledge base, to obtain the similar case backup screening library and the same theory case backup screening library. Then, the similar case backup screening library and the same theory case backup screening library are processed and finely screened through different angles such as the keyword information and the TRIZ theory additional attribute label information, to obtain the similar case output set, the same theory case output set and the recommended theory output set. That is, the target retrieval result is obtained through multiple screenings from different angles of the whole scheme and the TRIZ theory, so that the target retrieval result is concise and has high matching precision, which can help to provide accurate design inspiration and new ideas for solving process problems for development designers, and effectively improve product development efficiency and quality. BRIEF DESCRIPTION OF DRAWINGS

[0042] Figure 1 A flowchart of a process case retrieval method related to the embodiments of the present application;

[0043] Figure 2 A structural diagram of a process case retrieval device related to the embodiments of the present application;

[0044] Figure 3 An electronic device structural diagram of a hardware running environment related to the embodiments of the present application.

[0045] The implementation, functional features and advantages of the present application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0046] It should be understood that the specific embodiments described herein are merely intended to explain the present application, and are not intended to limit the present application.

[0047] Engineering knowledge is increasingly widely used in the field of design and manufacturing. Engineering knowledge, as the core of process design software, plays a crucial role in the performance and generalization of the software. When designers are designing products and innovating materials, they often need a large amount of engineering knowledge as the basis for research and development.

[0048] At present, when searching, only the keyword information is usually searched, and similar cases are screened from the knowledge base at one time according to the keyword information, and the similar cases are taken as target search cases. However, the process cases searched by the above method have limitations, cannot be intelligently searched and recommended from multiple angles, and lead to the situation that the process cases searched are messy and have low matching accuracy, affecting the research and development quality and efficiency of the development designers.

[0049] Based on this, the present application provides a solution, that is, a multi-angle process case searching method, which searches from two angles of keyword and TRIZ theory label, respectively obtains a similar case standby screening library and a same theory case standby screening library, and then based on the keyword information and the TRIZ theory additional attribute label information, the two standby screening libraries are clustered to obtain a similar case output set. Subsequently, based on the TRIZ theory additional attribute label information, a first process case containing all the TRIZ theory additional attribute label information is screened from the same theory case standby screening library; based on the keyword information, the number of TRIZ theory additional attribute label information of all process cases in the similar case standby screening library is counted, and the TRIZ theory additional attribute label information with more number is taken as a recommended theory to obtain a recommended theory output set. Finally, the similar case output set, the same theory case output set and the recommended theory output set obtained above are taken as the target search result. Through the above steps, the target search result obtained can be simplified and has high matching accuracy, which can effectively improve the product research and development efficiency and quality.

[0050] It can be understood that before the steps of the process case searching method are performed, the following steps are further included:

[0051] A process case knowledge base is established; the process case knowledge base includes a plurality of process cases, wherein each process case includes at least one of a title, an abstract, a keyword, a publication time and a content text; and each process case further includes at least one TRIZ theory additional attribute label for representing TRIZ theory.

[0052] In the specific implementation process, the method for establishing the process case knowledge base includes:

[0053] Based on the preset keywords, the relevant process cases are obtained by using the conventional means such as web crawler, the process cases are integrated, and the process knowledge base is obtained. The process cases are selected from the literature types such as periodicals, monographs, paper collections, degree papers, patents, technical standards, newspapers and scientific and technological reports. The preset keywords include at least one of material name, part model and part name. The content text of the process case refers to the text content in the process case except the title, abstract, keywords, release time. The TRIZ theory additional attribute label is used to explain the relevant TRIZ theory adopted by the process case, which includes the theory category, category serial number, specific name and name serial number of the TRIZ theory. The theory category of the TRIZ theory mainly includes invention principle, separation principle, scientific effect, etc. In addition to the TRIZ theory additional attribute label of the process case, all the remaining contents of the process case are stored in the form of string, and the TRIZ theory additional attribute label is a category data represented by character variable.

[0054] In addition, in actual application, according to the category of the TRIZ theory additional attribute label of the process case, the process case knowledge base is divided into several knowledge sub-bases.

[0055] In the specific implementation process, the collected process case knowledge is integrated and edited by the process experts in the field, and the TRIZ theory related additional attribute label adopted by each process case is edited, and the specific label type and content are shown in Table 1. After the corresponding key information of each process case is attached, each process case is integrated and stored in the process case knowledge base; wherein, the key information includes title, abstract, keywords, release time, content text and additional attribute label, as shown in Table 2.

[0056] Table 1 TRIZ theory additional attribute label type table

[0057] Class Principle class Class number Operation name Operation name number 1 Invention principle ① Splitting 001 2 Invention principle ① Extraction 002 3 Invention principle ① Local quality 003 …… …… …… …… …… m Separation principle ② Spatial separation principle 078 …… …… …… …… ……

[0058] Table 2 Process case composition table and example

[0059]

[0060]

[0061] The TRIZ theory additional attribute tag composed of "principle category, category serial number, operation name, and operation name serial number" represents the TRIZ theory additional attribute tag information of a process case. For example, the process knowledge case "porous metal mesh absorbs excess solder at the joint" adopts the TRIZ theory of "porous material", and the TRIZ theory additional attribute tag information is "1031". The process cases in the process case knowledge base with the same additional attribute tag are classified into the same knowledge sub-library for pre-classification.

[0062] Based on this, referring to Figure 1 The embodiment of the present application provides a process case retrieval method, which comprises the following steps:

[0063] In step S10, the case word information to be retrieved is obtained, and based on the case word information to be retrieved, a similar case standby screening library and a same theory case standby screening library are obtained; wherein the case word information to be retrieved comprises keyword information and TRIZ theory additional attribute tag information.

[0064] It should be noted that the case word information to be retrieved refers to the keyword information and TRIZ theory additional attribute tag information involved in the process knowledge to be obtained. The keyword information comprises a title and a keyword, and the TRIZ theory additional attribute tag information comprises a TRIZ theory additional attribute tag. The keyword can be a process method, a composite material name, a part model, a part name, a patent name, etc., for example, "large composite material wallboard forming".

[0065] In the specific implementation process, the input keyword information is matched with the process cases in the process case knowledge base, based on the reasoning principle consistent with at least one of the keyword information, a plurality of process cases matched with the keyword information are screened out, and the process cases are integrated to obtain a similar case standby screening library. The input TRIZ theory additional attribute tag information is matched with each knowledge sub-library, based on the reasoning principle consistent with at least one of the TRIZ theory additional attribute tag information, a plurality of knowledge sub-libraries matched with the TRIZ theory additional attribute tag information are screened out, and the process cases in the knowledge sub-libraries are integrated to obtain a same theory case standby screening library.

[0066] In actual application, after the keyword information is input in the search box, an additional search content can be added, that is, the TRIZ theory additional attribute tag information "1001" expected to be adopted is input, and the number of TRIZ theory additional attribute tag information to be added is selected according to the actual situation, that is, a plurality of TRIZ theory additional attribute tag information can be input at the same time for simultaneous search. For example, the search information is "large composite material wallboard forming (1001)".

[0067] Since the obtained to-be-retrieved case term information may contain words without characteristic meaning, the efficiency of the target retrieval result is affected. Therefore, obtaining a similar case backup screening library and a same theory case backup screening library based on the to-be-retrieved case term information comprises:

[0068] The to-be-retrieved case term information is preprocessed to obtain first to-be-retrieved case term information; wherein the preprocessing comprises string tokenization processing and stop word removal processing.

[0069] It should be noted that the first to-be-retrieved case term information is the term information remaining after the to-be-retrieved case term information is preprocessed. The string tokenization processing is a method of dividing a complete to-be-retrieved case term information string into multiple fields according to the retrieval rules. The stop word removal processing is a method of deleting corresponding stop words in the to-be-retrieved case term information according to a pre-constructed stop word list.

[0070] Based on the first to-be-retrieved case term information, a similar case backup screening library and a same theory case backup screening library are obtained.

[0071] It should be noted that after string segmentation and stop word deletion, the first to-be-retrieved case term information with characteristic meaning can be obtained. According to the matching of the first to-be-retrieved case term information and the process case in the process case knowledge base, a similar case backup screening library and a same theory case backup screening library are obtained respectively. Based on the similar case backup screening library and the same theory case backup screening library, further screening is carried out to obtain the relevant output set.

[0072] Next, taking the case retrieval in the process case knowledge base according to the search information "large composite wallboard forming (1001)" as an example, the above technical solutions are described in detail.

[0073] Firstly, the information in the search box is preprocessed, Chinese word segmentation and stop words are removed, and the obtained keywords "composite material, wallboard, forming" are matched with the "title, keyword, abstract, content text" of the process case in the process case knowledge base. According to the reasoning principle of "consistent with the information content of the case keyword to be searched", search for similar process cases in the process case knowledge base. For example, the search results are "large composite material wallboard forming technology", "large fuselage composite material stiffened wallboard manufacturing technology and application", "advanced manufacturing technology and application of large composite material wallboard", and the above process cases are integrated to form a similar case standby screening library. After the similar case standby screening library is formed, based on the same rule reasoning, the input TRIZ theory additional attribute label information is searched in the process case knowledge base, and the process cases in the process case knowledge base consistent with any TRIZ theory additional attribute label information are screened out. For example, the search results are "Analysis of modular furniture design", "Research on process scheme of T8150116 heavy truck split bearing special-shaped ring", "Design and manufacture of local convex die demolding mechanism of composite material forming die", and the above case information is combined to form a same theory case standby screening library.

[0074] Step S20, based on the keyword information, the similar case standby screening library and the same theory case standby screening library are clustered to obtain a similar case output set.

[0075] It should be noted that the clustering processing is a method of dividing the set formed by superimposing the process cases in the similar case standby screening library and the same theory case standby screening library into multiple classes composed of similar process cases. The process cases in each class are similar to each other and different from the process cases of other classes. The similar case output set obtained after clustering processing has a distribution rule, which can better facilitate the management of different types of process cases.

[0076] Specifically, the similar case standby screening library and the same theory case standby screening library are clustered based on the keyword information to obtain a similar case output set, including:

[0077] Based on the keyword information, the similar case keyword frequency value and the same theory case keyword frequency value are obtained.

[0078] In the implementation process, based on the keyword information, the keyword frequency value of each process case in the similar case backup screening library is calculated, that is, the similar case keyword frequency value; the similar case keyword frequency value is the set of keyword frequency values of the keyword information in the title, keyword, abstract and content text of each process case in the similar case backup screening library. The keyword frequency value of each process case in the same theory case backup screening library is calculated, that is, the same theory case keyword frequency value; the same theory case keyword frequency value is the set of keyword frequency values of the keyword information in the title, keyword, abstract and content text of each process case in the same theory case backup screening library.

[0079] It should be noted that in order to improve the retrieval efficiency, each part of the process case needs to be preprocessed. And the keyword frequency value of the keyword information in the title, keyword, abstract and content text of each process case needs to be calculated, so the above steps include:

[0080] Based on the keyword information, the process cases in the similar case backup screening library are preprocessed to obtain the similar case total word number value and the similar case keyword information frequency value.

[0081] It should be noted that the similar case total word number value represents the total number of words remaining after preprocessing each part of each process case in the similar case backup screening library. The similar case keyword information frequency value represents the number of times the keyword information appears in each part of each process case in the similar case backup screening library.

[0082] Based on the similar case total word number value and the similar case keyword information frequency value, the similar case keyword frequency value is obtained.

[0083] In the implementation process, the calculation formula of the similar case keyword frequency value is as follows:

[0084] TF1=N1 / M1;

[0085] Wherein, TF1 is the similar case keyword frequency value; N1 is the total number of words remaining after preprocessing each part of each process case in the similar case backup screening library; M1 is the number of times the keyword information appears in each part of each process case in the similar case backup screening library.

[0086] Based on the keyword information, the process cases in the same theory case backup screening library are preprocessed to obtain the same theory case total word number value and the same theory case keyword information frequency value;

[0087] It should be noted that the same theoretical case total word value represents the total number of words remaining after preprocessing of each part of each process case in the same theoretical case backup screening library. The same theoretical case keyword frequency value represents the number of times the keyword information appears in each part of each process case in the same theoretical case backup screening library.

[0088] Based on the same theoretical case total word value and the same theoretical case keyword frequency value, a same theoretical case keyword frequency value is obtained.

[0089] In the specific implementation process, the calculation formula of the similar case keyword frequency value is as follows:

[0090] TF2=N2 / M2;

[0091] Wherein, TF2 is the same theoretical case keyword frequency value; N2 is the total number of words remaining after preprocessing of each part of each process case in the same theoretical case backup screening library; M2 is the number of times the keyword information appears in each part of each process case in the same theoretical case backup screening library.

[0092] Based on the similar case keyword frequency value, a similar case feature vector is obtained; based on the same theoretical case keyword frequency value, a same theoretical case feature vector is obtained.

[0093] It should be noted that the similar case feature vector is obtained after the similar case keyword frequency value is represented by a vector; the same theoretical case feature vector is obtained after the same theoretical case keyword frequency value is represented by a vector.

[0094] The similar case feature vector and the same theoretical case feature vector are clustered to obtain a similar case output set.

[0095] More specifically, the above steps include:

[0096] Converting the keyword information into an initial feature vector;

[0097] The initial feature vector, the similar case feature vector and the same theoretical case feature vector are input into a preset clustering model for clustering processing to obtain a similar case output set.

[0098] For example, the process case set in the similar case backup screening library is X={x1, x2, x3,…, x i ,…,x n}, where xi is all the information of the i-th process case in the similar case backup screening library, i = 1, 2...n, and n represents the number of process cases in the similar case backup screening library. For example: X = {"Large composite wall panel forming technology", "Large fuselage composite reinforced wall panel manufacturing technology and application", "Large composite wall panel advanced manufacturing technology and application"}. The process case set in the same theoretical case backup screening library is Y = {y1, y2, y3, ..., y j ,…,ym}, where yj represents all the information for the jth process case in the backup screening library for the same theoretical case, j = 1, 2…m, and m represents the number of process cases in the backup screening library for the same theoretical case. For example: Y = {"A Brief Analysis of Modular Furniture Design," "Research on Process Solutions for Special-Shaped Rings of T8150116 Truck Separation Bearings," "Design and Manufacturing of the Local Punch Demolding Mechanism of Composite Material Forming Dies"}. The keyword frequency calculation formula above is used to obtain the keyword frequency value for each process case. x il ={x i1 ,x i2 ,x i3 ,x i4},y jl ={x j1 ,x j2 ,x j3 ,x j4} respectively represent the process cases x in the similar case alternative screening library i , Process case y of the alternative screening library of the same theoretical case i The word frequency values ​​of the title, keywords, abstract and content of the four parts, where l = 1, 2, 3, 4... Then the BERT model is used to vectorize the above keyword frequency values ​​to obtain and The feature vector of the keyword information is then converted into the initial feature vector x 0 ={1,1,1,1}, and input the above feature vector into the clustering model, and use the k-means clustering method to cluster, with the initial feature vector x 0 Take as the cluster center, perform cluster analysis, and obtain the output set of similar cases.

[0099] Step S30: obtaining a same-theory case output set based on the TRIZ theory additional attribute label information; wherein the same-theory case output set includes a plurality of first process cases, and each of the first process cases simultaneously includes all of the TRIZ theory additional attribute label information.

[0100] It should be noted that after obtaining the similar case output set, based on the TRIZ theory additional attribute tag information, the first process case containing all the TRIZ theory additional attribute tag information is screened out in the same theory case standby screening library, and the first process case is integrated to obtain the same theory case output set.

[0101] Because the process scheme will be adjusted according to the actual situation in the enterprise execution process, the closer the time is to the same theory case, the more the process case conforms to the current process demand, so the release time is arranged in descending order. Specifically, the same theory case output set is obtained based on the TRIZ theory additional attribute tag information, including:

[0102] Based on the TRIZ theory additional attribute tag information, a first same theory case output set is obtained; wherein the first same theory case output set includes a plurality of first process cases;

[0103] Based on the release time of the first process case, all the first process cases in the first same theory case output set are arranged in descending order to obtain a same theory case output set.

[0104] It should be noted that the first same theory case output set is a set of first process cases in the same theory case standby screening library containing all the TRIZ theory additional attribute tag information; the same theory case output set is a set of first process cases in the first same theory case output set arranged in descending order.

[0105] In practical application, if the input additional attribute tag is 1001, the retrieved same theory case output set is: "Analysis of modular furniture design", "T8150116 heavy truck split bearing special ring process scheme research", "Design and manufacture of local convex die demolding mechanism of composite material forming die". If multiple additional attribute tags are input at the same time, such as searching while inputting "periodic action 1019" and "rapid action 1021" two search additional attribute tags, then after matching, the process cases with "1019" and "1021" additional attribute tags will be output at the same time and arranged in descending order according to the release time, for example, "THz-TDS based ceramic fiber composite material defect detection analysis", "Preparation and performance of dye-doped luminescent polyurethane composite material", "Preparation and wave absorption performance of MoS2 / biomass carbon composite material".

[0106] Step S40, count the number of TRIZ theory additional attribute tag information of all process cases in the similar case standby screening library to obtain a recommended theory output set.

[0107] It should be noted that each process case in the similar case backup screening library contains one or more TRIZ theory additional attribute tag information which is the same or different, and the number of TRIZ theory additional attribute tag information of each process case in the similar case backup screening library is counted to obtain a recommended theory output set.

[0108] Specifically, the number of TRIZ theory additional attribute tag information of all process cases in the similar case backup screening library is counted to obtain a recommended theory output set, including:

[0109] The number of TRIZ theory additional attribute tag information of all process cases in the similar case backup screening library is counted to obtain each first number value.

[0110] It should be noted that the first number value represents the number value of each TRIZ theory additional attribute tag information contained in all process cases in the similar case backup screening library.

[0111] Based on the first number value, the TRIZ theory additional attribute tag information arranged in the top N is taken as a recommended theory to obtain a recommended theory output set; wherein, 1≤N≤5.

[0112] It should be noted that taking the TRIZ theory additional attribute tag information arranged in the top N as a recommended theory is equivalent to taking the top N TRIZ theories with the most TRIZ theory additional attribute tag information as recommended theories, and then integrating the recommended theories to obtain a recommended theory output set.

[0113] In actual application, the number of TRIZ theory additional attribute tag information of all process cases in the similar case backup screening library is counted, and the five TRIZ theories arranged in the top five according to the number are "1031 (porous material)", "1001 (division)", "1002 (extraction)", "1006 (versatility)", and "1014 (curving)", and the above TRIZ theories are integrated to obtain a recommended theory output set.

[0114] Step S50, output the similar case output set, the same theory case output set and the recommended theory output set to obtain a target retrieval result.

[0115] It should be noted that the target retrieval result is a retrieval result output by the terminal to the research and development designer after the research and development designer inputs the to-be-retrieved case term information in the terminal.

[0116] In actual application, the target retrieval result obtained is:

[0117] Similar case output set: "Large composite panel forming technology", "Large composite fuselage stiffened panel manufacturing technology and application", "Large composite panel advanced manufacturing technology and application";

[0118] Same theory case output set: "Analysis of modular furniture design", "Research on process scheme of special-shaped ring of split bearing of T8150116 truck", "Design and manufacturing of local convex die demolding mechanism of composite forming die";

[0119] Recommended theory output set: "1031 (porous material)", "1001 (split)", "1002 (extraction)", "1006 (versatility)", "1014 (curving)".

[0120] In summary, the embodiments of the present application analyze, screen and intelligently recommend from different angles of the overall scheme and TRIZ theory, so that the obtained target retrieval result is concise and has high matching accuracy, which can help to provide accurate design inspiration and new ideas for solving process problems for development designers, and effectively improve the product research and development efficiency and quality.

[0121] Referring to Figure 2 , based on the same invention idea, the embodiments of the present application also propose a process case retrieval device, comprising:

[0122] An initial screening module is configured to obtain to-be-retrieved case term information; based on the to-be-retrieved case term information, a similar case backup screening library and a same theory case backup screening library are obtained; wherein the to-be-retrieved case term information includes keyword information and TRIZ theory additional attribute tag information;

[0123] A first target obtaining module is configured to perform clustering processing on the similar case backup screening library and the same theory case backup screening library based on the keyword information, and obtain a similar case output set;

[0124] A second target obtaining module is configured to obtain a same theory case output set based on the TRIZ theory additional attribute tag information; wherein the same theory case output set includes a plurality of first process cases, and each first process case contains all the TRIZ theory additional attribute tag information;

[0125] A third target obtaining module is configured to count the number of TRIZ theory additional attribute tag information of all process cases in the similar case backup screening library, and obtain a recommended theory output set;

[0126] A target output module is configured to output the similar case output set, the same theory case output set and the recommended theory output set, and obtain a target retrieval result.

[0127] It should be noted that the modules in the process case retrieval device in this embodiment correspond one by one to the steps in the process case retrieval method in the foregoing embodiments, and therefore the specific embodiments of this embodiment can refer to the embodiments of the foregoing process case retrieval method, which will not be described again here.

[0128] Referring to Figure 3 , Figure 3 The electronic device structure schematic diagram of the hardware running environment involved in the embodiment scheme of the present application.

[0129] As Figure 3 shown, the electronic device can include a processor 1001, such as a central processing unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. The communication bus 1002 is used to realize the connection and communication between the components. The user interface 1003 can include a display, an input unit such as a keyboard, and can also include a standard wired interface, a wireless interface. The network interface 1004 can optionally include a standard wired interface, a wireless interface (such as a wireless fidelity (WIreless-FIdElity, WI-FI) interface). The memory 1005 can be a high-speed random access memory (RAM) memory, and can also be a stable non-volatile memory (Non-Volatile Memory, NVM), such as a disk memory. The memory 1005 can also be a storage device independent of the foregoing processor 1001.

[0130] Those skilled in the art can understand Figure 3 that the structure shown in the foregoing embodiments does not constitute a limitation on the electronic device, and can include more or fewer components than the diagram, or combine certain components, or different component arrangements.

[0131] As Figure 3 shown, the memory 1005 as a storage medium can include an operating system, a network communication module, a user interface module, and an electronic program.

[0132] In Figure 3The electronic device shown, the network interface 1004 is mainly used for data communication with the network server; the user interface 1003 is mainly used for data interaction with the user; the processor 1001 and the memory 1005 in the electronic device of the application can be arranged in the electronic device, the electronic device calls the process case retrieval device stored in the memory 1005 through the processor 1001, and executes the process case retrieval method provided by the application.

[0133] In addition, in an embodiment, the embodiment of the application also provides a computer program product, which, when executed by a processor, implements the method described above.

[0134] In addition, in an embodiment, the embodiment of the application also provides a computer storage medium, which stores a computer program, and the computer program is executed by a processor to implement the steps of the method in the foregoing embodiment.

[0135] In some embodiments, the computer readable storage medium can be FRAM, ROM, PROM, EPROM, EEPROM, flash memory, magnetic surface memory, optical disc, or CD-ROM memory, etc.; it can also be various devices including one or any combination of the above storage devices. The computer can be various computing devices including smart terminals and servers.

[0136] In some embodiments, the executable instructions can be in the form of programs, software, software modules, scripts or codes, written in any form of programming language (including compiled or interpreted languages, or declarative or procedural languages), and can be deployed in any form, including being deployed as independent programs or being deployed as modules, components, subroutines or other units suitable for use in computing environments.

[0137] As an example, the executable instructions can but not necessarily correspond to files in the file system, can be stored in part of a file storing other programs or data, for example, stored in one or more scripts in a hyper text markup language (HTML, Hyper Text Markup Language) document, stored in a single file dedicated to the program in question, or stored in multiple cooperative files (for example, files storing one or more modules, subroutines or code portions).

[0138] As an example, the executable instructions can be deployed to execute on one computing device, or on multiple computing devices located at one site, or on multiple computing devices distributed in multiple sites and interconnected through a communication network.

[0139] It should be noted that, in this document, the terms "comprises", "comprising", or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or system. An element proceeded by "comprises a" does not, without more constraints, exclude the presence of additional identical elements in the process, method, article, or system that comprises the element.

[0140] The above-mentioned sequence numbers of the embodiments of the present application are only for description, and do not represent advantages or disadvantages of the embodiments.

[0141] From the above description of the embodiments, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be realized by means of software and necessary general hardware platforms, and of course can also be realized by hardware, but in many cases the former is a better embodiment. Based on such understanding, the technical solutions of the present application can be embodied in the form of a software product in essence or in the form of a part of the prior art that makes a contribution. The computer software product is stored in a storage medium (such as a read-only memory / random access memory, a magnetic disk, or an optical disk) and includes a plurality of instructions for causing a multimedia terminal device (which can be a mobile phone, a computer, a television receiver, or a network device) to execute the methods described in the various embodiments of the present application.

[0142] The above is only a preferred embodiment of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation, or direct or indirect application in other related technical fields, which is made by using the content of the specification and drawings of the present application, is also included in the patent protection scope of the present application.

Claims

1. A process case retrieval method, characterized in that: The following steps are involved: Get the information of the case entries to be retrieved; Based on the case entry information to be retrieved, a similar case backup screening library and a same theory case backup screening library are obtained; wherein the case entry information to be retrieved includes keyword entry information and TRIZ theory additional attribute tag information; Based on the keyword information, clustering is performed on the similar case backup screening library and the same theoretical case backup screening library to obtain a similar case output set; Based on the TRIZ theory additional attribute label information, an output set of identical theory cases is obtained; wherein the output set of identical theory cases includes a plurality of first process cases, and each of the first process cases simultaneously includes all of the TRIZ theory additional attribute label information; Counting the number of TRIZ theory additional attribute label information of all process cases in the similar case backup screening library to obtain a recommended theory output set; The similar case output set, the same theory case output set and the recommended theory output set are output to obtain target retrieval results.

2. The process case retrieval method according to claim 1, wherein: The method of obtaining a similar case backup screening library and an identical theoretical case backup screening library based on the case entry information to be retrieved includes: Preprocessing the to-be-retrieved case term information to obtain first to-be-retrieved case term information; wherein the preprocessing includes character string segmentation processing and stop word removal processing; Based on the first to-be-retrieved case term information, a similar case backup screening library and an identical theoretical case backup screening library are obtained.

3. The process case retrieval method according to claim 1, wherein: The method of clustering the similar case backup screening library and the same theoretical case backup screening library based on the keyword information to obtain a similar case output set includes: Based on the keyword information, obtaining similar case keyword frequency values ​​and the same theoretical case keyword frequency values; Based on the similar case keyword frequency values, a similar case feature vector is obtained; based on the same theoretical case keyword frequency values, a same theoretical case feature vector is obtained; Clustering is performed on the similar case feature vectors and the same theoretical case feature vectors to obtain a similar case output set.

4. The process case retrieval method according to claim 3, wherein: The obtaining of similar case keyword frequency values ​​and identical theoretical case keyword frequency values ​​based on the keyword entry information includes: Based on the keyword information, pre-processing the process cases in the similar case backup screening library is performed to obtain the total word value of similar cases and the number of keyword entries of similar cases; Obtaining a similar case keyword frequency value based on the similar case total word value and the similar case keyword entry count value; Based on the keyword information, pre-processing the process cases in the same theoretical case backup screening library is performed to obtain the total word value of the same theoretical case and the number of keyword entries of the same theoretical case; Based on the total word value of the same theoretical case and the number of key word entries of the same theoretical case, the frequency value of the key word of the same theoretical case is obtained.

5. The process case retrieval method according to claim 3, wherein: The clustering process of the similar case feature vectors and the identical theoretical case feature vectors to obtain a similar case output set includes: Converting the keyword information into an initial feature vector; The initial feature vector, the similar case feature vector and the same theoretical case feature vector are input into a preset clustering model and clustered to obtain a similar case output set.

6. The process case retrieval method according to claim 1, wherein: The method of obtaining the same theoretical case output set based on the additional attribute tag information of the TRIZ theory includes: Based on the TRIZ theory additional attribute label information, a first identical theory case output set is obtained; wherein the first identical theory case output set includes a plurality of first process cases; Based on the release time of the first process case, all the first process cases in the first identical theoretical case output set are arranged in descending order to obtain an identical theoretical case output set.

7. The process case retrieval method according to claim 1, wherein: The number of TRIZ theory additional attribute label information of all process cases in the similar case backup screening library is counted to obtain a recommended theory output set, including: Counting the number of TRIZ theory additional attribute label information of all process cases in the similar case backup screening library to obtain each first quantity value; Based on the first quantity values, descending order is performed, and the TRIZ theory additional attribute label information arranged in the first N is used as a recommended theory to obtain a recommended theory output set; wherein 1≤N≤5.

8. A process case retrieval device, characterized in that: include: The initial screening module is used to obtain the information of the case entries to be retrieved; Based on the case entry information to be retrieved, a similar case backup screening library and a same theory case backup screening library are obtained; wherein the case entry information to be retrieved includes keyword entry information and TRIZ theory additional attribute tag information; A first target acquisition module is configured to perform clustering processing on the similar case backup screening library and the same theoretical case backup screening library based on the keyword information to obtain a similar case output set; A second target acquisition module is configured to obtain an output set of identical theoretical cases based on the TRIZ theory additional attribute label information; wherein the output set of identical theoretical cases includes a plurality of first process cases, each of which simultaneously includes all of the TRIZ theory additional attribute label information; The third target acquisition module is used to count the number of TRIZ theory additional attribute label information of all process cases in the similar case backup screening library to obtain a recommended theory output set; The target output module is used to output the similar case output set, the same theory case output set and the recommended theory output set to obtain the target retrieval result.

9. An electronic device, characterized in that: The electronic device includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and the processor executes the computer program to implement the method according to any one of claims 1 to 7.

Images