CN111459973B - Case type retrieval method and system based on case situation triple information - Google Patents
Case type retrieval method and system based on case situation triple information Download PDFInfo
- Publication number
- CN111459973B CN111459973B CN202010548781.3A CN202010548781A CN111459973B CN 111459973 B CN111459973 B CN 111459973B CN 202010548781 A CN202010548781 A CN 202010548781A CN 111459973 B CN111459973 B CN 111459973B
- Authority
- CN
- China
- Prior art keywords
- case
- triple
- model
- database
- entity
- 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.)
- Active
Links
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
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/242—Query formulation
- G06F16/243—Natural language query formulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2455—Query execution
-
- 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—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/18—Legal services; Handling legal documents
Abstract
The invention discloses a case retrieval method and a system based on case triple information, wherein the method comprises the following steps: marking case samples in the case sample library to obtain a marked sample library; training an input model of the labeled sample library to obtain a triple set, wherein triples in the triple set comprise case entities and elements corresponding to the case entities; setting a triple scoring model for extracting case situations; when searching the class case, inputting the information of the case to be searched into the trained model to obtain a corresponding triple set; and scoring the triple sets of the case in the case database by using the triple scoring model, and taking the case corresponding to one or more triple sets with the highest score as a class retrieval result. The invention relates to a case retrieval method and a case retrieval system based on case triple information, which construct a key measurement standard for retrieving a case by utilizing a pre-training language model and a sorting algorithm in the legal field and realize accurate retrieval of the case from the perspective of legal profession.
Description
Technical Field
The invention relates to a natural language processing technology, in particular to a case retrieval method and a system based on case triple information.
Background
At present, key word technology and tf-idf technology are mainly used for class retrieval. Because the case situation fact part is long in length, the current technology is semantic matching of paragraphs at chapter level (consisting of a plurality of sentences) and not semantic matching at sentence level; meanwhile, the description modes of the same entity type and the same case dispute point are various. For the reasons, the conventional method cannot extract effective information of case situations, and the accuracy and recall rate of the class case retrieval are not ideal.
Disclosure of Invention
The invention aims to solve the technical problems that the accuracy rate and the recall rate of case retrieval in the prior art are not ideal, and aims to provide a case retrieval method and a system based on case triple information to solve the problems.
The invention is realized by the following technical scheme:
a case retrieval method based on case triple information comprises the following steps:
s1: selecting a case sample from a case database to obtain a case sample library, marking the case sample in the case sample library to obtain a marked sample library, wherein marked content comprises a case entity and elements corresponding to the case entity;
s2: training an input model of the labeled sample library to obtain a case entity recognition model and a case element recognition model;
s3: identifying case entities of cases in a case database by using a case entity identification model; identifying the case elements in the case database by using the case element identification model; obtaining a triple set of each case according to the case entity of the case in the case database and the corresponding element of the case entity; the elements of the triplets in the triple set comprise the case entity and the elements corresponding to the case entity;
s4: storing the case and the triple set corresponding to the case in the case database, and setting a triple scoring model for extracting the case;
s5: when searching the class case, inputting the case to be searched into the trained model to obtain a triple set of the case to be searched; grading the matching degree of the triple set of the case and the triple set of the case to be retrieved in the case database by using a triple grading model; and taking the case corresponding to the triple set with the score meeting the preset requirement in the case database as a class retrieval result.
When the method is applied, firstly, a sample is marked, the marked sample is used for training a model, and the marked main content comprises a case entity and elements, wherein the elements correspond to the case entity. The case entity in the invention refers to the entity in the legal case, such as the quilt notice, married students, children, property, etc.; an element refers to an element that affects the result. And the data are corresponded to facilitate the learning and training of the subsequent model. By learning these data, a triple set can be obtained, which is the basis for class retrieval. Meanwhile, the triple set also corresponds to a scoring model for optimizing the triple set. And during case retrieval, extracting the information of the case to be retrieved through the same operation, obtaining a triple set of the case to be retrieved through the same rule processing, and comparing the triple set with the triple set in the case database to realize the case retrieval.
Further, all the triples in the triple set are provided with weight values;
and setting the weight value of the triple according to the importance degree of the element represented by the triple in the case sample.
Further, a triple scoring model is set according to the weight values.
Further, the model adopts regular and conditional random fields to extract entities; adopting a pre-training model and a sequencing algorithm in the legal field to realize multi-label classification; triple elements result from specific rule-based entity and multi-tag combinations.
When the invention is applied, the specific rule means that the entity and the label are in matching correspondence, for example:
firstly, an entity and a label are extracted, and a specific rule refers to the combination of the entity and the label; for example, if an entity is a defendant and a tag is a family violence is extracted from a sentence, the generated triplets are a defendant, a tag has and a family violence; in the invention, the relationship between the entity and the label is configured in the configuration file.
A kind of case retrieval system based on case triple information, including:
a sample labeling unit: the system comprises a case condition database, a case condition sample labeling database and a case condition sample labeling database, wherein the case condition sample database is used for selecting a case condition sample from the case condition database to obtain a case condition sample database, the case condition sample in the case condition sample database is labeled to obtain a labeled sample database, and labeled contents comprise case condition entities and elements corresponding to the case condition entities;
a model training unit: the case condition entity recognition model and the case condition element recognition model are obtained by training the input model of the labeled sample library;
a triple set construction unit: identifying case entities of cases in a case database by using a case entity identification model; identifying the case elements in the case database by using the case element identification model; obtaining a triple set of each case according to the case entity of the case in the case database and the corresponding element of the case entity; the elements of the triplets in the triple set comprise the case entity and the elements corresponding to the case entity;
a storage unit: the triple scoring model is used for storing the case and the triple set corresponding to the case in the case database and setting the triple scoring model for extracting the case;
a retrieval unit: when the method is used for case retrieval, inputting the case to be retrieved into the trained model to obtain a triple set of the case to be retrieved; grading the matching degree of the triple set of the case in the case database and the triple set of the case to be retrieved by utilizing a triple grading model; and taking the case corresponding to the triple set with the score meeting the preset requirement in the case database as a class retrieval result.
Further, all the triples in the triple set are provided with weight values;
and setting the weight value of the triple according to the importance degree of the element represented by the triple in the case sample.
Further, a triple scoring model is set according to the weight values.
Further, the model adopts regular and conditional random fields to extract entities; the training model adopts a pre-training model and a sequencing algorithm in the legal field to realize multi-label classification; triple elements result from specific rule-based entity and multi-tag combinations.
Compared with the prior art, the invention has the following advantages and beneficial effects:
the invention relates to a case retrieval method and a system based on case triple information, which utilize a pre-training language model and a sequencing algorithm in the legal field to construct a key measurement standard for case retrieval: and the case triple elements realize accurate retrieval of the class case from the perspective of legal profession.
Drawings
The accompanying drawings, which are included to provide a further understanding of the embodiments of the invention and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the invention and together with the description serve to explain the principles of the invention. In the drawings:
fig. 1 is a schematic diagram of a case retrieval process based on case triple information according to the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail below with reference to examples and accompanying drawings, and the exemplary embodiments and descriptions thereof are only used for explaining the present invention and are not meant to limit the present invention.
Examples
As shown in fig. 1, the invention relates to a case retrieval method based on case triple information, which comprises the following steps:
s1: selecting a case sample from a case database to obtain a case sample library, marking the case sample in the case sample library to obtain a marked sample library, wherein marked content comprises a case entity and elements corresponding to the case entity;
s2: training an input model of the labeled sample library to obtain a case entity recognition model and a case element recognition model;
s3: identifying case entities of cases in a case database by using a case entity identification model; identifying the case elements in the case database by using the case element identification model; obtaining a triple set of each case according to the case entity of the case in the case database and the corresponding element of the case entity; the elements of the triplets in the triple set comprise the case entity and the elements corresponding to the case entity;
s4: storing the case and the triple set corresponding to the case in the case database, and setting a triple scoring model for extracting the case;
s5: when searching the class case, inputting the case to be searched into the trained model to obtain a triple set of the case to be searched; grading the matching degree of the triple set of the case and the triple set of the case to be retrieved in the case database by using a triple grading model; and taking the case corresponding to the triple set with the score meeting the preset requirement in the case database as a class retrieval result.
In the implementation of this embodiment, a sample is labeled first, the labeled sample is used for training a model, and the labeled main content includes a case entity and an element, where the element corresponds to the case entity. The case entity in the invention refers to the entity in the legal case, such as the quilt notice, married students, children, property, etc.; an element refers to an element that affects the result. And the data are corresponded to facilitate the learning and training of the subsequent model. By learning these data, a triple set can be obtained, which is the basis for class retrieval. Meanwhile, the triple set also corresponds to a scoring model for optimizing the triple set. And during case retrieval, extracting the information of the case to be retrieved through the same operation, obtaining a triple set of the case to be retrieved through the same rule processing, and comparing the triple set with the triple set in the case database to realize the case retrieval.
For further explaining the working process of the embodiment, all the triples in the triplet set are provided with weight values;
and setting the weight value of the triple according to the importance degree of the element represented by the triple in the case sample.
To further illustrate the operation of the embodiment, the triple scoring model is set according to the weight value.
To further illustrate the working process of the embodiment, the model adopts regular and conditional random fields to extract entities; adopting a pre-training model and a sequencing algorithm in the legal field to realize multi-label classification; triple elements result from specific rule-based entity and multi-tag combinations.
A kind of case retrieval system based on case triple information, including:
a sample labeling unit: the system comprises a case condition database, a case condition sample labeling database and a case condition sample labeling database, wherein the case condition sample database is used for selecting a case condition sample from the case condition database to obtain a case condition sample database, the case condition sample in the case condition sample database is labeled to obtain a labeled sample database, and labeled contents comprise case condition entities and elements corresponding to the case condition entities;
a model training unit: the case condition entity recognition model and the case condition element recognition model are obtained by training the input model of the labeled sample library;
triple set unit: identifying case entities of cases in a case database by using a case entity identification model; identifying the case elements in the case database by using the case element identification model; obtaining a triple set of each case according to the case entity of the case in the case database and the corresponding element of the case entity; the elements of the triplets in the triple set comprise the case entity and the elements corresponding to the case entity;
a storage unit: the triple scoring model is used for storing the case and the triple set corresponding to the case in the case database and setting the triple scoring model for extracting the case;
a retrieval unit: when the method is used for case retrieval, inputting the case to be retrieved into the trained model to obtain a triple set of the case to be retrieved; grading the matching degree of the triple set of the case and the triple set of the case to be retrieved in the case database by using a triple grading model; and taking the case corresponding to the triple set with the score meeting the preset requirement in the case database as a class retrieval result.
For further explaining the working process of the embodiment, all the triples in the triplet set are provided with weight values;
and setting the weight value of the triple according to the importance degree of the element represented by the triple in the case sample.
To further illustrate the operation of the embodiment, the triple scoring model is set according to the weight value.
To further illustrate the working process of the embodiment, the model adopts regular and conditional random fields to extract entities; adopting a pre-training model and a sequencing algorithm in the legal field to realize multi-label classification; triple elements result from specific rule-based entity and multi-tag combinations.
To further illustrate the working process of the present embodiment, a specific example is illustrated:
as shown in FIG. 1, when retrieving a case, the case retrieval system of the present invention uses the case retrieval method of the present invention based on the case facts of the current case to display the most relevant cases of the same kind to the judge; the whole class case retrieval method comprises the following main steps:
based on specific case, a professional legal team defines triples of case facts of the case, namely key elements of the case, the triples are measurement standards for class retrieval, and each triplet has a corresponding weight value and indicates the importance degree of the element in the case; elements such as divorce table are exemplified as follows: the original advertisement has married children, the married children always follow the life of the original advertisement, the quilt has family violence, and the like;
the professional legal team marks the case situation fact part of the referee document, and marks the entity including the case situation: such as original comforter, married children, property, etc.; labeling the label corresponding to each sentence in the description of the case, such as elements influencing the division of children and women, elements influencing the division of property, elements influencing the feelings of couples and wives, and the like, which is a label with multiple labels;
constructing a natural language processing algorithm model, and training the model by using the labeled data; and applying the trained model to a massive document library to obtain a series of the triple elements for each referee document. The specific algorithm is as follows: extracting regular and conditional random field entities; adopting a pre-training model and a sequencing algorithm in the legal field to realize multi-label classification; the triple elements are obtained by combining an entity based on a specific rule and multiple labels;
and storing the referee document and the corresponding triple information by using an ElasticSearch, and setting a scoring algorithm based on triple extraction (fully utilizing the weight of the triple).
During class retrieval, the input case facts are subjected to the algorithm model to obtain corresponding triples, then the triples are matched with cases in the elastic search, and some cases with the highest score are displayed.
The above-mentioned embodiments are intended to illustrate the objects, technical solutions and advantages of the present invention in further detail, and it should be understood that the above-mentioned embodiments are merely exemplary embodiments of the present invention, and are not intended to limit the scope of the present invention, and any modifications, equivalent substitutions, improvements and the like made within the spirit and principle of the present invention should be included in the scope of the present invention.
Claims (4)
1. A case retrieval method based on case triple information is characterized by comprising the following steps:
s1: selecting a case sample from a case database to obtain a case sample library, marking the case sample in the case sample library to obtain a marked sample library, wherein marked content comprises a case entity and elements corresponding to the case entity;
s2: inputting the labeled sample library into a natural language processing algorithm model for training to obtain a case entity recognition model and a case element recognition model;
extracting entities by a natural language processing algorithm model by adopting regular and conditional random fields; adopting a pre-training model and a sequencing algorithm in the legal field to realize multi-label classification; the elements of the triples are obtained by combining entities and multiple labels based on specific rules;
the specific rule means that the entity and the label are in matching correspondence;
s3: identifying case entities of cases in a case database by using a case entity identification model; identifying the case elements in the case database by using the case element identification model; obtaining a triple set of each case according to the case entity of the case in the case database and the corresponding element of the case entity; the elements of the triplets in the triple set comprise the case entity and the elements corresponding to the case entity;
s4: storing the case and the triple set corresponding to the case in the case database, and setting a triple scoring model for extracting the case;
s5: when searching the class case, inputting the case to be searched into the trained model to obtain a triple set of the case to be searched; grading the matching degree of the triple set of the case and the triple set of the case to be retrieved in the case database by using a triple grading model; taking the case corresponding to the triple set with the score meeting the preset requirement in the case database as a category retrieval result;
the triples in the triple set are all provided with weight values;
setting the weight value of the triad according to the importance degree of the triad in the case sample;
step S1 includes the following substeps:
the professional legal team defines a triple of the case facts, namely key elements of the case, the triple is a measurement standard of class retrieval, and each triple has a corresponding weight value which indicates the importance degree of the triple in the case;
professional legal teams label the case situation fact part of the referee document and label the entity including the case situation.
2. The case retrieval method based on case triple information as claimed in claim 1, wherein the triple scoring model is set according to weight values.
3. A kind of case retrieval system based on the triple information of case situation, characterized by that, including:
a sample labeling unit: the system comprises a case condition database, a case condition sample labeling database and a case condition sample labeling database, wherein the case condition sample database is used for selecting a case condition sample from the case condition database to obtain a case condition sample database, the case condition sample in the case condition sample database is labeled to obtain a labeled sample database, and labeled contents comprise case condition entities and elements corresponding to the case condition entities;
a model training unit: the system comprises a case entity recognition model, a case element recognition model, a natural language processing algorithm model, a case element recognition model and a case element recognition model, wherein the case entity recognition model is used for recognizing case elements;
extracting entities by a natural language processing algorithm model by adopting regular and conditional random fields; adopting a pre-training model and a sequencing algorithm in the legal field to realize multi-label classification; the elements of the triples are obtained by combining entities and multiple labels based on specific rules;
the specific rule means that the entity and the label are in matching correspondence;
a triple set construction unit: identifying case entities of cases in a case database by using a case entity identification model; identifying the case elements in the case database by using the case element identification model; obtaining a triple set of each case according to the case entity of the case in the case database and the corresponding element of the case entity; the elements of the triplets in the triple set comprise the case entity and the elements corresponding to the case entity;
a storage unit: the triple scoring model is used for storing the case and the triple set corresponding to the case in the case database and setting the triple scoring model for extracting the case;
a retrieval unit: when the method is used for case retrieval, inputting the case to be retrieved into the trained model to obtain a triple set of the case to be retrieved; grading the matching degree of the triple set of the case and the triple set of the case to be retrieved in the case database by using a triple grading model; taking the case corresponding to the triple set with the score meeting the preset requirement in the case database as a category retrieval result;
the triples in the triple set are all provided with weight values;
setting the weight value of the triad according to the importance degree of the triad in the case sample;
the professional legal team defines a triple of the case facts, namely key elements of the case, the triple is a measurement standard of class retrieval, and each triple has a corresponding weight value which indicates the importance degree of the triple in the case;
professional legal teams label the case situation fact part of the referee document and label the entity including the case situation.
4. The case retrieval system based on case triple information as claimed in claim 3, wherein the triple scoring model is set according to weight values.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010548781.3A CN111459973B (en) | 2020-06-16 | 2020-06-16 | Case type retrieval method and system based on case situation triple information |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010548781.3A CN111459973B (en) | 2020-06-16 | 2020-06-16 | Case type retrieval method and system based on case situation triple information |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111459973A CN111459973A (en) | 2020-07-28 |
| CN111459973B true CN111459973B (en) | 2020-10-23 |
Family
ID=71683952
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010548781.3A Active CN111459973B (en) | 2020-06-16 | 2020-06-16 | Case type retrieval method and system based on case situation triple information |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111459973B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112632226B (en) * | 2020-12-29 | 2021-10-26 | 天津汇智星源信息技术有限公司 | Semantic search method and device based on legal knowledge graph and electronic equipment |
| CN114528409A (en) * | 2022-04-18 | 2022-05-24 | 北京北大软件工程股份有限公司 | Method and device for evaluating extraction result of element information of letter and visit article |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105701253A (en) * | 2016-03-04 | 2016-06-22 | 南京大学 | Chinese natural language interrogative sentence semantization knowledge base automatic question-answering method |
| WO2018096514A1 (en) * | 2016-11-28 | 2018-05-31 | Thomson Reuters Global Resources | System and method for finding similar documents based on semantic factual similarity |
| CN108647194A (en) * | 2018-04-28 | 2018-10-12 | 北京神州泰岳软件股份有限公司 | information extraction method and device |
| CN110413732A (en) * | 2019-07-16 | 2019-11-05 | 扬州大学 | The knowledge searching method of software-oriented defect knowledge |
| CN110795926A (en) * | 2020-01-03 | 2020-02-14 | 四川大学 | Judgment document similarity judgment method and system based on legal knowledge graph |
| CN111079433A (en) * | 2019-11-29 | 2020-04-28 | 北京奇艺世纪科技有限公司 | Event extraction method and device and electronic equipment |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109145293B (en) * | 2018-08-06 | 2021-05-28 | 中国地质大学（武汉） | Case-oriented keyword extraction method and system |
| CN109886270B (en) * | 2019-01-17 | 2022-03-01 | 大连理工大学 | Case element identification method for electronic file record text |
-
2020
- 2020-06-16 CN CN202010548781.3A patent/CN111459973B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105701253A (en) * | 2016-03-04 | 2016-06-22 | 南京大学 | Chinese natural language interrogative sentence semantization knowledge base automatic question-answering method |
| WO2018096514A1 (en) * | 2016-11-28 | 2018-05-31 | Thomson Reuters Global Resources | System and method for finding similar documents based on semantic factual similarity |
| CN108647194A (en) * | 2018-04-28 | 2018-10-12 | 北京神州泰岳软件股份有限公司 | information extraction method and device |
| CN110413732A (en) * | 2019-07-16 | 2019-11-05 | 扬州大学 | The knowledge searching method of software-oriented defect knowledge |
| CN111079433A (en) * | 2019-11-29 | 2020-04-28 | 北京奇艺世纪科技有限公司 | Event extraction method and device and electronic equipment |
| CN110795926A (en) * | 2020-01-03 | 2020-02-14 | 四川大学 | Judgment document similarity judgment method and system based on legal knowledge graph |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111459973A (en) | 2020-07-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110502621B (en) | Question answering method, question answering device, computer equipment and storage medium | |
| US7421418B2 (en) | Method and apparatus for fundamental operations on token sequences: computing similarity, extracting term values, and searching efficiently | |
| CN110765257A (en) | Intelligent consulting system of law of knowledge map driving type | |
| AU2010208523B2 (en) | Methods and systems for matching records and normalizing names | |
| Khusro et al. | On methods and tools of table detection, extraction and annotation in PDF documents | |
| CN111475623A (en) | Case information semantic retrieval method and device based on knowledge graph | |
| CN106651696B (en) | Approximate question pushing method and system | |
| CN112632989B (en) | Method, device and equipment for prompting risk information in contract text | |
| CN101354705B (en) | Apparatus and method for processing document image | |
| CN114036281B (en) | Knowledge graph-based citrus control question-answering module construction method and question-answering system | |
| JP2013016172A (en) | System and method for processing data | |
| CN111459973B (en) | Case type retrieval method and system based on case situation triple information | |
| AU2018411565B2 (en) | System and methods for generating an enhanced output of relevant content to facilitate content analysis | |
| CN110795932A (en) | Geological report text information extraction method based on geological ontology | |
| CN104636487A (en) | Advertising information management method | |
| CN110704615B (en) | Internet financial non-dominant advertisement identification method and device | |
| CN113282729A (en) | Question-answering method and device based on knowledge graph | |
| Tkaczyk et al. | Extracting contextual information from scientific literature using CERMINE system | |
| CN111274354B (en) | Referee document structuring method and referee document structuring device | |
| CN117216221A (en) | Intelligent question-answering system based on knowledge graph and construction method | |
| CN112015907A (en) | Method and device for quickly constructing discipline knowledge graph and storage medium | |
| Poornima et al. | Multi-modal features and correlation incorporated Naive Bayes classifier for a semantic-enriched lecture video retrieval system | |
| CN110532538B (en) | Key entity extraction algorithm for property dispute referee document | |
| CN114842982B (en) | Knowledge expression method, device and system for medical information system | |
| CN111538832A (en) | Apparatus and method for event annotation of document and recording medium |
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 | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 610000 No. 24 south part of Wuhou District first ring road, Chengdu, Sichuan. Patentee after: SICHUAN University Patentee after: Chengdu shuzhilian Technology Co., Ltd Address before: 610000 No. 24 south part of Wuhou District first ring road, Chengdu, Sichuan. Patentee before: SICHUAN University Patentee before: Chengdu shuzhilian Technology Co., Ltd |
|
| CP01 | Change in the name or title of a patent holder |