CN110837566A - Dynamic construction method of knowledge graph for CNC (computerized numerical control) machine tool fault diagnosis - Google Patents
Dynamic construction method of knowledge graph for CNC (computerized numerical control) machine tool fault diagnosis Download PDFInfo
- Publication number
- CN110837566A CN110837566A CN201911116876.1A CN201911116876A CN110837566A CN 110837566 A CN110837566 A CN 110837566A CN 201911116876 A CN201911116876 A CN 201911116876A CN 110837566 A CN110837566 A CN 110837566A
- Authority
- CN
- China
- Prior art keywords
- user
- case
- users
- cases
- knowledge graph
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000003745 diagnosis Methods 0.000 title claims abstract description 29
- 238000010276 construction Methods 0.000 title claims abstract description 16
- 238000011156 evaluation Methods 0.000 claims abstract description 10
- 238000000034 method Methods 0.000 claims description 16
- 238000000605 extraction Methods 0.000 claims description 13
- 238000012423 maintenance Methods 0.000 claims description 7
- 230000004927 fusion Effects 0.000 claims description 5
- 230000009469 supplementation Effects 0.000 claims description 3
- 238000012937 correction Methods 0.000 claims description 2
- 230000011218 segmentation Effects 0.000 claims description 2
- 238000012549 training Methods 0.000 claims 1
- 238000004519 manufacturing process Methods 0.000 abstract description 2
- 238000005516 engineering process Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000013136 deep learning model Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 239000013589 supplement Substances 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000008676 import Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000004377 microelectronic Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/20—Administration of product repair or maintenance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/04—Manufacturing
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Human Resources & Organizations (AREA)
- General Engineering & Computer Science (AREA)
- General Business, Economics & Management (AREA)
- Evolutionary Computation (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Health & Medical Sciences (AREA)
- Tourism & Hospitality (AREA)
- Economics (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Manufacturing & Machinery (AREA)
- Computing Systems (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Evolutionary Biology (AREA)
- Primary Health Care (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Animal Behavior & Ethology (AREA)
- Databases & Information Systems (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the invention provides a dynamic construction method of a knowledge graph aiming at CNC (computer numerical control) machine tool fault diagnosis, which comprises the following steps: extracting entities from the structural data and the non-structural data, establishing a relationship between the entities, and establishing a part aiming at the case in the knowledge graph; and obtaining similarity information between users from the case evaluation of the users, and establishing relationship attributes between the users in the knowledge graph. Because machine tool faults mainly occur in a production link and depend on experience more, a layer of relation information between users is added to original case information, and corresponding cases are recommended better through similarity between the users to improve the diagnosis accuracy. According to the technical scheme provided by the invention, the reliability of the knowledge graph can be improved when the knowledge graph is used.
Description
Technical Field
The invention relates to the field of fault diagnosis, in particular to a dynamic construction method of a knowledge graph aiming at CNC machine tool fault diagnosis.
Background
The CNC system integrates advanced computer technology, microelectronic technology, servo control technology and automatic control technology, so that the numerical control machine tool has the electromechanical-hydraulic integration characteristic, the operation capacity of the numerical control machine tool is improved, the failure rate of the numerical control machine tool is higher than that of a common machine tool, and the maintenance difficulty is increased. In actual production processes, CNC systems may have various failure problems, and generally, the related maintenance engineers need a great deal of case experience, accumulation of related knowledge, and ability to quickly find, solve, and practice problems, which substantially increases the difficulty of maintenance work and the time consumed in the maintenance process.
The prior field of mechanical fault diagnosis still has great defects in the aspects of knowledge representation and acquisition, and has great progress space in the aspect of reasoning strategy of fault diagnosis, most of the prior researches lack an effective learning mechanism, and a practical fault diagnosis system not only has a perfect fault identification method, but also contains sufficient abundant, accurate and effective knowledge in a knowledge base, so that the system needs certain learning capacity. Therefore, the knowledge graph tool is applied to the field of fault diagnosis, and user data information except for cases is introduced into the knowledge graph so as to improve the accuracy of fault diagnosis.
Disclosure of Invention
The invention provides a dynamic construction method of a knowledge graph aiming at CNC machine tool fault diagnosis, which realizes the construction method of the knowledge graph of a CNC machine tool diagnosis case by adopting the concept of the knowledge graph, deep learning and a similarity calculation method, and simultaneously provides a method for searching the constructed knowledge graph.
The invention discloses a dynamic construction method of a knowledge graph aiming at CNC machine tool fault diagnosis, which comprises the following specific steps:
step 1) case entity extraction: the step is a first construction method of the fault case, which establishes a triple structure among entities, relations and entities by entity extraction and knowledge fusion of the existing unstructured case data and establishes a foundation for constructing a complete machine tool fault diagnosis knowledge map. The entity types comprise entity information such as case alarm information, machine tool information, a CNC system model, case key part points, case phenomenon description, case solutions and the like, each case basically comprises all the information of the entities, and after extraction, the corresponding relation between the extracted new entities or the existing entities is established according to case items.
Step 2) case supplementation: the partial cases are used for continuously improving the knowledge graph which is preliminarily established in the step 1), cases which are added by users or existing structured case data are directly established into a triple data structure in a classification, programmed addition or batch import mode, and the partial data provide guarantee for the richness and the perfection of the knowledge graph and make up the defect of a single construction mode.
Step 3) case fusion: fusing the triple structures obtained in the steps 1) and 2) to form data of a case part of the complete knowledge map, and simultaneously removing repeated entities to reduce unnecessary space overhead and finish the primary construction of the machine tool fault diagnosis knowledge map on the fault case part.
Step 4), adding user scores: in the step, the information of the user is added into the atlas, and meanwhile, the evaluation information of the user on the case is added into the knowledge atlas constructed in the step, so that the attribute addition of the grading relation of the user on the case is completed.
Step 5) calculating the similarity among users: and calculating the similarity between the users according to the evaluation of the users on the common cases and the evaluation conditions of the users on the cases, and finishing the maintenance of the relationship between the users.
Step 6) dynamically maintaining the knowledge graph: according to the updating of the cases in the knowledge graph in actual use and the feedback of the cases by the users, the similarity information among the users and the related information of the cases are periodically corrected.
Step 7) searching method aiming at the atlas: matching is carried out according to the matching degree of the problems and the phenomena and the similarity between the user and other users, corresponding cases are comprehensively returned, the solution of the cases is solved, and updating is carried out according to the feedback of the user.
The invention provides a dynamic construction method and steps of a knowledge graph aiming at CNC machine tool fault diagnosis, the knowledge graph can well utilize the mutual relation among cases to store and reason the CNC machine tool fault, and in addition, the relations among users and the relations among the users and the cases are added to cooperatively diagnose and reason the fault information. Meanwhile, the invention also provides a retrieval method aiming at the knowledge graph.
In a first aspect of the invention, methods and steps are disclosed for dynamic construction of a knowledge graph for CNC machine fault diagnosis.
In a second aspect of the invention, a way of calculating information similarity between users of knowledge graphs for CNC machine tool fault diagnosis is disclosed to assist in case reasoning and diagnosis.
The third aspect of the invention discloses a knowledge graph retrieval method for CNC machine tool fault diagnosis, which is used for comprehensively helping a user to carry out case reasoning and analysis by adopting two angles of case matching and user matching for the application of the knowledge graph.
Drawings
FIG. 1 is a schematic flow diagram of the present invention;
fig. 2 is a schematic flow diagram of case entity extraction;
FIG. 3 is a schematic diagram of a deep learning model for entity extraction;
FIG. 4 is a partial schematic view of a fault diagnosis knowledge map.
Detailed Description
For better understanding of the technical solutions of the present invention, the following detailed descriptions of the embodiments of the present invention are provided with reference to the accompanying drawings.
As shown in fig. 1, the specific process of the present invention is as follows:
case entity extraction: and performing entity extraction on the existing unstructured case data. The process is shown in fig. 2, firstly, data cleaning is carried out on a case, then word segmentation is carried out on the subsequent data, common part-of-speech marking is carried out on regular entities according to different entity types, a dictionary part-of-speech marking method is adopted irregularly, BIO marking is carried out according to the data with marked part-of-speech, namely the beginning of the entity is marked as B, the rest is marked as I, the non-entity part is marked as O, the marked data is trained by using a BilSTM-CRF deep learning model, the model is shown in fig. 3, the trained model is verified by a test set and has 5802 marks and 359 words, the accuracy can reach 91.71%, and the better model can be selected and used for entity extraction tasks after being trained. The case part of the corresponding knowledge map is established for our case through entity extraction and knowledge fusion, wherein the entity types comprise entity information such as case alarm information, machine tool information, CNC system models, case key part points, case phenomenon description, case solution schemes and the like, each case basically comprises all the information of the entities, and after extraction, the corresponding relation between the extracted new entity or the existing entity is established according to case items.
Case supplementation: the partial cases are to continuously perfect the knowledge graph which is preliminarily established previously, and cases which are added by users or existing structured case data are added into the knowledge graph in different modes according to different data types. And importing a single or a plurality of cases of the user by adopting a filling programmed adding mode, and importing the structured form data by adopting a batch importing programmed mode.
Case fusion: and fusing triple structures obtained by entity extraction and case supplement to form data of the case part of the complete knowledge graph. In the process, repeated entities need to be combined to reduce unnecessary space overhead, and relationship information between the entities needs to be perfected to completely supplement partial incomplete information of the knowledge graph case.
Adding user scores: in the step, the information of the user is added into the atlas, and meanwhile, the evaluation information of the user on the case is added into the knowledge atlas constructed in the step, so that the attribute addition of the grading relation of the user on the case is completed.
Calculating the similarity between users: and calculating the similarity between the users according to the evaluation of the users on the common cases and the evaluation conditions of the users on the cases, and finishing the maintenance of the relationship between the users. The scoring difference between users is calculated according to the following formula:
D(u,v)＝|Cu-Cv|
wherein u represents user u, v represents user v, CuRepresenting a set of scores for user u in a case of common scores for user u and user v, CvThe scoring set of the user v in the scoring case shared by the user u and the user v is represented;
calculating the information entropy between users according to the grading difference of the cases, and then correcting the result of the information entropy according to the case where the difference value intersects with the user to obtain the final similarity between the users, wherein the formula is as follows:
in the formula (d)iRepresents the difference between two users corresponding to case i, and N represents the number of cases evaluated by the corresponding user.
Dynamically maintaining the knowledge graph: according to the updating of the cases in the knowledge graph in actual use and the feedback of the cases by the users, the similarity information between the users and the related information of the cases are corrected through a timing task.
The searching method aiming at the atlas comprises the following steps: matching is carried out according to the matching degree of the problems and the phenomena and the similarity between the user and other users, corresponding cases are comprehensively returned, the solution of the cases is solved, and updating is carried out according to the feedback of the user. Calculating the similarity between the problems and the cases from the knowledge graph according to the entities according to the following formula:
where set I represents a corresponding set of entities (I) extracted from a user input question1,i2,…,in1) The set D represents the set of entities contained in the case in the knowledge map of machine tool diagnosis (D)1,d2,…,dn2)；
Then, using the calculated similarity between users, the similarity sim (I, D) between cases is corrected according to the following formula:
wherein S isiIndicates the degree of similarity with the user i, riRepresents the rating of the user i for the case, and sigma represents the user similarity correction coefficient.
Through the steps, a fault diagnosis knowledge graph aiming at the CNC machine tool can be finally constructed, and the application of the fault diagnosis knowledge graph on the aspect of searching is realized, and FIG. 4 is a local demonstration of the constructed knowledge graph of the invention
By adopting the knowledge graph, the similar latitude of the user can be increased on the content information of the original case, the case with higher possibility can be recommended to the user better in case recommendation with the same possibility, and the knowledge graph is dynamically updated to realize self-optimization of data.
Claims (4)
1. The dynamic construction method of the knowledge graph for CNC machine tool fault diagnosis is characterized by comprising the following specific steps:
1) case entity extraction: the first construction method for the fault case: establishing entities, relations and triple structures among the entities by extracting the entities and fusing knowledge of the existing unstructured case data to establish a foundation for constructing a complete machine tool fault diagnosis knowledge map;
2) case supplementation: the second construction method for the fault case is as follows: cases added by a user or existing structured case data are directly established into a triple data structure in a classification mode, and the data provide guarantee for the richness and the perfection of a knowledge graph;
3) case fusion: fusing the triple structures obtained in the steps 1) and 2) to complete the primary construction of the machine tool fault diagnosis knowledge graph in the fault case part;
4) adding user scores: adding the evaluation information of the user on the case into the established knowledge graph to complete the attribute addition of the grading relation of the user on the case;
5) calculating the similarity between users: calculating the similarity between users according to the evaluation of the users on the common cases and the evaluation conditions of the users on the cases, and finishing the maintenance of the relationship between the users;
6) dynamically maintaining the knowledge graph: according to the updating of the cases in the knowledge graph in actual use and the feedback of the cases by the users, the similarity information between the users and the related information of the cases are regularly corrected;
7) the searching method aiming at the atlas comprises the following steps: matching is carried out according to the matching degree of the problems and the phenomena and the similarity between the user and other users, corresponding cases are comprehensively returned, the solution of the cases is solved, and updating is carried out according to the feedback of the user.
2. The method of claim 1, wherein step 1) comprises: performing word segmentation, part-of-speech tagging, fault dictionary tagging and BIO tagging on unstructured case data, performing data training by adopting a Bi-LSTM-CRF model according to tagged data, and finally putting the model with optimal comprehensive performance into a system for use.
3. The method as claimed in claim 1, wherein said step 5) comprises: calculating similarity data among users through the scoring relation among the cases among the users, wherein the similarity data includes:
the scoring difference between users is calculated according to the following formula:
D(u,v)＝|Cu-Cv|
wherein u represents user u, v represents user v, CuRepresenting a set of scores for user u in a case of common scores for user u and user v, CvThe scoring set of the user v in the scoring case shared by the user u and the user v is represented;
calculating the information entropy between users according to the grading difference of the cases, and then correcting the result of the information entropy according to the case where the difference value intersects with the user to obtain the final similarity between the users, wherein the formula is as follows:
in the formula (d)iRepresents the difference between two users corresponding to case i, and N represents the number of cases evaluated by the corresponding user.
4. The method as claimed in claim 1, wherein said step 7) comprises: corresponding entities in user problems are obtained by utilizing entity extraction, and the similarity between the problems and the cases is calculated from the knowledge graph according to the entities from the following formula:
where set I represents a corresponding set of entities (I) extracted from a user input question1,i2,…,in1) The set D represents the set of entities contained in the case in the knowledge map of machine tool diagnosis (D)1,d2,…,dn2)；
Then, using the similarity between users calculated in claim 2, the similarity sim (I, D) between cases is corrected according to the following formula:
wherein S isiIndicates the degree of similarity with the user i, riRepresents the rating of the user i for the case, and sigma represents the user similarity correction coefficient.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911116876.1A CN110837566B (en) | 2019-11-15 | 2019-11-15 | Dynamic construction method of knowledge graph for CNC (computerized numerical control) machine tool fault diagnosis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911116876.1A CN110837566B (en) | 2019-11-15 | 2019-11-15 | Dynamic construction method of knowledge graph for CNC (computerized numerical control) machine tool fault diagnosis |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110837566A true CN110837566A (en) | 2020-02-25 |
| CN110837566B CN110837566B (en) | 2022-05-13 |
Family
ID=69575034
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911116876.1A Active CN110837566B (en) | 2019-11-15 | 2019-11-15 | Dynamic construction method of knowledge graph for CNC (computerized numerical control) machine tool fault diagnosis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110837566B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111400505A (en) * | 2020-03-13 | 2020-07-10 | 国家电网有限公司 | Method and system for matching fault elimination scheme of power consumption information acquisition system |
| CN111738445A (en) * | 2020-05-26 | 2020-10-02 | 山东大学 | Design knowledge fusion reasoning method supporting product rapid innovation |
| CN116360388A (en) * | 2023-01-18 | 2023-06-30 | 北京控制工程研究所 | Reasoning method and device of performance-fault relation map based on graph neural network |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102765643A (en) * | 2012-05-31 | 2012-11-07 | 天津大学 | Elevator fault diagnosis and early-warning method based on data drive |
| CN107783973A (en) * | 2016-08-24 | 2018-03-09 | 慧科讯业有限公司 | The methods, devices and systems being monitored based on domain knowledge spectrum data storehouse to the Internet media event |
| CN108509483A (en) * | 2018-01-31 | 2018-09-07 | 北京化工大学 | The mechanical fault diagnosis construction of knowledge base method of knowledge based collection of illustrative plates |
| CN110008288A (en) * | 2019-02-19 | 2019-07-12 | 武汉烽火技术服务有限公司 | The construction method in the knowledge mapping library for Analysis of Network Malfunction and its application |
-
2019
- 2019-11-15 CN CN201911116876.1A patent/CN110837566B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102765643A (en) * | 2012-05-31 | 2012-11-07 | 天津大学 | Elevator fault diagnosis and early-warning method based on data drive |
| CN107783973A (en) * | 2016-08-24 | 2018-03-09 | 慧科讯业有限公司 | The methods, devices and systems being monitored based on domain knowledge spectrum data storehouse to the Internet media event |
| CN108509483A (en) * | 2018-01-31 | 2018-09-07 | 北京化工大学 | The mechanical fault diagnosis construction of knowledge base method of knowledge based collection of illustrative plates |
| CN110008288A (en) * | 2019-02-19 | 2019-07-12 | 武汉烽火技术服务有限公司 | The construction method in the knowledge mapping library for Analysis of Network Malfunction and its application |
Non-Patent Citations (2)
| Title |
|---|
| SHIGANG ZHANG等: ""Optimization of a Dynamic Fault Diagnosis Model Based on Machine Learning"", 《IEEE ACCESS》 * |
| 刘绪忠 等: ""基于知识图谱的故障智能诊断手段研究"", 《山东通信技术》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111400505A (en) * | 2020-03-13 | 2020-07-10 | 国家电网有限公司 | Method and system for matching fault elimination scheme of power consumption information acquisition system |
| CN111738445A (en) * | 2020-05-26 | 2020-10-02 | 山东大学 | Design knowledge fusion reasoning method supporting product rapid innovation |
| CN116360388A (en) * | 2023-01-18 | 2023-06-30 | 北京控制工程研究所 | Reasoning method and device of performance-fault relation map based on graph neural network |
| CN116360388B (en) * | 2023-01-18 | 2023-09-08 | 北京控制工程研究所 | Reasoning method and device of performance-fault relation map based on graph neural network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110837566B (en) | 2022-05-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109902171B (en) | Text relation extraction method and system based on hierarchical knowledge graph attention model | |
| CN110837566B (en) | Dynamic construction method of knowledge graph for CNC (computerized numerical control) machine tool fault diagnosis | |
| CN107273490B (en) | Combined wrong question recommendation method based on knowledge graph | |
| CN110880019B (en) | Method for adaptively training target domain classification model through unsupervised domain | |
| CN108334638B (en) | Project score prediction method based on long-term and short-term memory neural network and interest migration | |
| CN104598611B (en) | The method and system being ranked up to search entry | |
| CN107562863A (en) | Chat robots reply automatic generation method and system | |
| CN112307153B (en) | Automatic construction method and device of industrial knowledge base and storage medium | |
| US20220327398A1 (en) | Technology maturity judgment method and system based on science and technology data | |
| CN111914550B (en) | Knowledge graph updating method and system oriented to limited field | |
| CN111709244A (en) | Deep learning method for identifying causal relationship of contradictory dispute events | |
| CN116719917A (en) | Knowledge question-answering system, method and storage medium combining large model with external knowledge | |
| CN115391553B (en) | Method for automatically searching time sequence knowledge graph completion model | |
| CN115114421A (en) | Question-answer model training method | |
| CN109508864A (en) | A kind of method for building up of enterprise's default risk model based on xgboost | |
| CN117093260B (en) | Fusion model website structure analysis method based on decision tree classification algorithm | |
| CN109597901B (en) | Data analysis method based on biological data | |
| CN117313262A (en) | Intelligent recommendation method for mechanical part milling strategy | |
| CN114240539B (en) | Commodity recommendation method based on Tucker decomposition and knowledge graph | |
| CN112416920B (en) | MES-oriented data cleaning method and system | |
| CN111708896B (en) | Entity relationship extraction method applied to biomedical literature | |
| CN113468995A (en) | Crowd counting method based on density grade perception | |
| CN110163498B (en) | Courseware originality scoring method and device, storage medium and processor | |
| CN112348175A (en) | Method for performing feature engineering based on reinforcement learning | |
| CN111898961A (en) | Error checking method suitable for same field of standing book data of similar power equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |