CN111625626A - Multi-user case retrieval system based on multi-dimensional semantic combined modeling - Google Patents
Multi-user case retrieval system based on multi-dimensional semantic combined modeling Download PDFInfo
- Publication number
- CN111625626A CN111625626A CN202010747457.4A CN202010747457A CN111625626A CN 111625626 A CN111625626 A CN 111625626A CN 202010747457 A CN202010747457 A CN 202010747457A CN 111625626 A CN111625626 A CN 111625626A
- Authority
- CN
- China
- Prior art keywords
- referee document
- referee
- user
- case
- document
- 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
- 238000000605 extraction Methods 0.000 claims abstract description 25
- 238000010276 construction Methods 0.000 claims abstract description 12
- 238000002372 labelling Methods 0.000 claims description 13
- 238000000034 method Methods 0.000 claims description 13
- 238000007781 pre-processing Methods 0.000 claims description 7
- 239000011159 matrix material Substances 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 5
- 238000012216 screening Methods 0.000 claims description 5
- 238000012163 sequencing technique Methods 0.000 claims description 4
- 230000002457 bidirectional effect Effects 0.000 claims description 3
- 238000012546 transfer Methods 0.000 claims description 2
- 238000010586 diagram Methods 0.000 description 4
- 230000011218 segmentation Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000006467 substitution reaction 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
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/3332—Query translation
- G06F16/3334—Selection or weighting of terms from queries, including natural language queries
-
- 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
- G06F16/31—Indexing; Data structures therefor; Storage structures
- G06F16/313—Selection or weighting of terms for indexing
-
- 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
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
-
- 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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- 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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- 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
- G06Q50/10—Services
- G06Q50/18—Legal services
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Artificial Intelligence (AREA)
- Business, Economics & Management (AREA)
- Databases & Information Systems (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Evolutionary Computation (AREA)
- Tourism & Hospitality (AREA)
- Life Sciences & Earth Sciences (AREA)
- Technology Law (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention relates to a multi-user case retrieval system based on multi-dimensional semantic combined modeling, which comprises a knowledge construction module: extracting information of the existing referee document, building a referee document library, and generating a referee document image library and a case and case personnel image library; an element identification module: extracting element characteristics of the text provided by the user by using an element characteristic extraction model, and extracting the user character characteristics of the consulting user; a matching query module: and inquiring and matching the extracted element features in the element recognition module and the proposed user character features in a referee document image library and a case personnel image library generated by the knowledge construction module respectively. The technical scheme provided by the invention can be suitable for the retrieval requirements of common people on the legal cases, meets the requirement of retrieval accuracy to a certain extent, and improves the retrieval accuracy of lawyers or common people for retrieving similar cases or retrieving whether corresponding similar referee documents exist according to self situations.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a multi-user case retrieval system based on multi-dimensional semantic combined modeling.
Background
With the coming of the information age, channels for people to acquire information are wider, and meanwhile, the requirements on the quality of acquired information are also improved, especially in the field of legal case retrieval.
At present, people tend to seek related cases on the network and check judgment conditions when encountering legal disputes, however, most of case retrieval systems in the current market are based on keyword retrieval modes and cannot fully capture the main body targets of users; secondly, most case retrieval systems are oriented to users with legal professional knowledge, and ordinary people cannot well relate to the retrieval systems; even a retrieval system related to specific semantic analysis only captures shallow semantics and cannot mine deep information of a user, so that the retrieval accuracy of the method is low.
Disclosure of Invention
In order to solve the problems, the invention provides a multi-user case retrieval system based on multi-dimensional semantic combined modeling, which can improve the retrieval accuracy of lawyers or common people in retrieving similar cases or retrieving whether corresponding similar referee documents exist according to self situations.
The technical scheme of the invention is as follows:
a multi-user case retrieval system based on multi-dimensional semantic joint modeling comprises the following components:
a knowledge construction module: extracting information of the existing referee document, building a referee document library, and generating a referee document image library and a case personnel image library according to the referee document library;
an element identification module: extracting element characteristics of the text provided by the user by using an element characteristic extraction model, and extracting the user character characteristics of the consulting user;
a matching query module: and inquiring and matching the extracted element features in the element recognition module and the proposed user character features in a referee document image library and a case personnel image library generated by the knowledge construction module respectively.
Preferably, the construction process of the referee library comprises:
s1: based on the existing referee document, performing structured hierarchical operation on the referee document; the layering operation specifically comprises the following steps: the referee document is layered by the information subareas of the original telling, the told telling, the fact finding, the courtyard thinking and the judgment, the case personnel information and the like in the referee document;
s2: marking the referee document based on the field of the existing referee document;
s3: and establishing a referee text library based on the information.
Preferably, the process of constructing the referee document portrait database comprises:
s1: designing a referee document portrait model in advance based on the referee document library;
s2: extracting each layer of information of the structured referee document based on the referee document portrait model;
s3: and constructing a referee document portrait database according to the extracted referee document portrait.
Preferably, the construction process of the case personnel portrait library is as follows:
s1: designing a case personnel portrait model in advance based on the existing user character characteristics;
s2: based on the user portrait model, portrait information extraction is carried out on case personnel in the judgment document;
s3: and preliminarily establishing a case personnel photo library based on the information.
Preferably, the extraction process of the official document portrait specifically comprises:
s1: aiming at the linguistic data in the referee text library, performing element labeling on part of the linguistic data in the referee text library by professional personnel related to law, and generating a hierarchical element labeling set;
s2: aiming at original reported information in the referee document, extracting referee document case-related figure information, constructing case-related figure images, and generating a referee document case-related figure image database;
s3: performing word embedding training on the corpus by using a BERT (belief transfer) aiming at the judging document element labeling set and existing labeling data of a judging document case-related human figure image database, generating a word weight matrix by using a bidirectional LSTM (least squares metric) model based on an attention mechanism, and training a judging document element feature extraction model based on the word weight matrix;
s4: and (3) aiming at the referee document element labeling set and the referee document case-related person feature image database, using a referee document element feature extraction model to extract element features of the linguistic data in the whole database, generating referee document images according to different elements, and further generating a referee document image database.
Preferably, the step of implementing the function by the element identification module is as follows:
s1: performing data preprocessing on an input text of a user, specifically, dividing the input text into a list consisting of single characters;
s2: aiming at the result of data preprocessing, using a trained element feature extraction model BERT + BilSTM + Attention to carry out element identification to obtain an identification result set (namely an output vector of a classification layer) of various elements;
s3: finally, determining the elements with the threshold value higher than the screening value as the elements extracted from the input text in the element identification result set;
s4: the extraction of the user character feature of the consulting user is performed according to the information structure of the case personnel portrait.
Preferably, the steps of the matching query module for implementing functions are as follows:
s1: element identification and sequencing, namely performing portrait matching on elements extracted from a text input by a consultation user, acquiring a plurality of referee documents in the front of the element identification and sequencing from a referee document portrait library, and calculating the similarity sim _ label of the referee document portrait;
s2: calculating similarity sim _ person between the user and the portrait in the prior referee document and case personnel portrait library according to the user portrait of the consultant user;
s3: according to the referee document labels and the figure image labels extracted by the element feature extraction model, calculating the similarity sim _ score between the referee document labels and all documents in the referee document library by using an ElasticSearch database through a joint calculation formula, and returning a plurality of corresponding referee documents according to the sim _ score descending order;
s4: and returning a plurality of referee documents with top scores in the similarity sim _ score in the matching process to the user. Preferably, the joint calculation formula is specifically:
the invention has the beneficial effects that: the technical scheme provided by the invention can be suitable for the retrieval requirements of common people on the legal cases, meets the requirement of retrieval accuracy to a certain extent, and improves the retrieval accuracy of lawyers or common people for retrieving similar cases or retrieving whether corresponding similar referee documents exist according to self situations.
Drawings
FIG. 1 is an overall architecture diagram of an embodiment of the present invention.
FIG. 2 is an architecture diagram of a knowledge building block in an embodiment of the invention.
Fig. 3 is an architecture diagram of an element recognition module according to an embodiment of the present invention.
FIG. 4 is an architecture diagram of a match query module in an embodiment of the invention.
Detailed Description
The embodiments of the present invention will be described in detail below with reference to the accompanying drawings.
As shown in FIG. 1, the multi-user case retrieval system based on multi-dimensional semantic joint modeling comprises three modules, namely knowledge construction, element identification and matching query.
As shown in fig. 2, the knowledge building module includes the following steps:
step 1: and (5) constructing a structured referee document library.
In this step, the structured referee library is constructed mainly by the following means:
the hierarchical information of the referee document is extracted according to the referee document, for example, the general referee document has the following parts: the original telling and telling, the institute found out, the institute thought that the judgement is as follows, and the case personnel information, firstly, we extract the information from the original referee document library according to the rule, and convert the original referee document into the layered referee document.
Step 2: and extracting the feature of the referee document.
In this step, the following method is mainly adopted:
because the information confidence degrees of different parts in the hierarchical information of the referee document are different, wherein the original declaration and the told declaration are classified into one type of information, the court considers that the court finds out and judges the other type of information as follows. The former has a certain subjectivity, the latter is objective, and in the element feature extraction model, element feature extraction is performed according to the two types of information, such as:
the ' release labor contract agreement ' signed by the upper complainer and the upper complainer is voluntarily agreed by both parties, does not violate the mandatory regulations of laws and administrative laws, is legal and effective, and the upper complainer pays 134040 yuan of the upper complainer's economic reimbursement according to the agreed agreement.
In this section of official document, we can extract the following elements: there is a release labor contract agreement, both parties agree to release, both parties have an agreement on the economic indemnity or compensation, and the court supports the proposition of requesting payment of the economic indemnity.
The steps finally form a judge document and document portrait library, and the concrete process is as follows:
s1: aiming at the linguistic data in the structured referee text library, performing element labeling on part of the linguistic data in the referee text library by legal related professionals to generate a hierarchical element labeling set;
s2, extracting the referee document case-related character information aiming at the original reported information in the referee document, constructing case-related character images and generating a referee document case-related character image database;
s3, aiming at the annotation set of referee document elements and existing annotation data of the referee document case-related human figure image database, carrying out word embedding training on the corpus by using BERT, generating a word weight matrix by using a bidirectional LSTM model based on an attention mechanism, and training a referee document element feature extraction model based on the word weight matrix;
s4, aiming at the referee document element labeling set, using the referee document element feature extraction model to extract the element features of the linguistic data in the whole database with the referee document case character image database, generating referee document images according to different elements, and further generating the referee document image database.
And step 3: extracting the case personnel image.
In this step, the following method is mainly adopted:
the method comprises the steps of presetting required character image characteristics, extracting information of case personnel according to the preset characteristics, wherein the preset characteristics comprise the following steps: gender, identity, native place, age group, academic horizon, etc.
And generating a case staff image library corresponding to the referee document from the referee document library according to the preset characteristics.
After the hierarchical referee text library and the case personnel image library are established, the element identification module provided by the embodiment is entered.
The element recognition module shown in fig. 3 includes the following steps:
step 1: and (4) preprocessing data.
Because the BERT input is based on characters, word segmentation operation is not needed, and the defect of poor performance of Chinese word segmentation is avoided. In this step, the text is divided into a list of individual words. If the text in the referee document is "both parties have a factual labor relationship", it is converted into such a representation [ 'double', 'square', 'existing', 'thing', 'real', 'labor', 'dynamic', 'off', 'system' ]afterdata preprocessing.
Step 2: and (5) element extraction.
And aiming at the result of data preprocessing, using a trained element feature extraction model BERT + BilSTM + Attention to extract elements, and finally considering the elements with the threshold value higher than the screening value in the vector output by the classification layer as the elements for extracting the text lock. Namely: the element of 'existence of factual labor relationship' contained behind 'two parties have factual labor relationship' in the above example is extracted.
And step 3: and extracting the character features of the user.
And extracting the corresponding character features of the user by using the user feature extraction model according to the information provided by the user and the structure (such as gender, region, age level, annual income level and the like) of the case personnel image library.
After obtaining the elements corresponding to the user input, the module shown in fig. 4 is entered, and the module is a matching query module provided in this embodiment, and includes the following steps:
step 1: and (5) screening candidate cases by the referee document elements.
Acquiring element information extracted from a user text aiming at an element feature extraction model, and using a weighting-word matching algorithm: as in the above example, when searching for the "both have factual labor relationship" element, for one matching referee document, if the element image of the referee document is [ a, B, C, D, E, there is factual labor relationship ], the element matching degree is 1/(1 +5 + 0.1+ 0) = 0.67, where 1 is the weight of the element extracted from the user text and 0.1 is the weight of the element not matched in the referee document image. And the case with the matching degree of 20 is listed as a candidate case.
Step 2: and screening candidate cases with case personnel pictures.
And calculating the similarity of the case and the case personnel of each case by using the same weighting-word matching algorithm for the 20 screened candidate cases and the case personnel images of the cases.
And step 3: the joint calculation reorders the candidate cases.
In order to ensure that the identified similar cases are more consistent with the statements of the user, the identified similar cases are reordered in the step by using the element matching similarity sim _ label and the case personnel image similarity sim _ person obtained in the previous example, so that the requirement of accurate matching is met to a certain extent. The reordering formula is:
for example, with the text "both parties have a factual labor relationship", the score is 0.66 in the element identification module. However, in another case, the similarity of the possible elements is 0.8, and at this time, the situation and the situation person and the counselor are required to be more consistent by comparing the case person images, so that the accuracy of the similar case is improved, and thus, the entity reordering operation is required.
Finally, it should be noted that: the above-mentioned embodiments are only specific embodiments of the present invention, which are used for illustrating the technical solutions of the present invention and not for limiting the same, and the protection scope of the present invention is not limited thereto, although the present invention is described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: any person skilled in the art can modify or easily conceive the technical solutions described in the foregoing embodiments or equivalent substitutes for some technical features within the technical scope of the present disclosure; such modifications, changes or substitutions do not depart from the spirit and scope of the present invention in its spirit and scope. Are intended to be covered by the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (8)
1. A multi-user case retrieval system based on multi-dimensional semantic joint modeling is characterized by comprising:
a knowledge construction module: extracting information of the existing referee document, building a referee document library, and generating a referee document image library and a case personnel image library according to the referee document library;
an element identification module: extracting element characteristics of the text provided by the user by using an element characteristic extraction model, and extracting the user character characteristics of the consulting user;
a matching query module: and inquiring and matching the extracted element features in the element recognition module and the proposed user character features in a referee document image library and a case personnel image library generated by the knowledge construction module respectively.
2. The multi-user case retrieval system based on multi-dimensional semantic union modeling according to claim 1, wherein the construction process of the referee text library is as follows:
s1: based on the existing referee document, performing structured hierarchical operation on the referee document;
s2: marking the referee document based on the field of the existing referee document;
s3: and establishing a referee text library based on the information.
3. The multi-user case retrieval system based on multi-dimensional semantic union modeling according to claim 1, wherein the construction process of the referee document portrait database is as follows:
s1: designing a referee document portrait model in advance based on the referee document library;
s2: extracting each layer of information of the structured referee document based on the referee document portrait model;
s3: and constructing a referee document portrait database according to the extracted referee document portrait.
4. The multi-user case retrieval system based on multi-dimensional semantic combined modeling according to claim 1, wherein the construction process with case personnel portrait library is as follows:
s1: designing a case personnel portrait model in advance based on the existing user character characteristics;
s2: based on the user portrait model, portrait information extraction is carried out on case personnel in the judgment document;
s3: and preliminarily establishing a case personnel photo library based on the information.
5. The multi-user case retrieval system based on multi-dimensional semantic combined modeling as claimed in claim 3, wherein the extraction process of the referee document portrait is specifically:
s1: aiming at the linguistic data in the referee text library, performing element labeling on part of the linguistic data in the referee text library by professional personnel related to law, and generating a hierarchical element labeling set;
s2: aiming at original reported information in the referee document, extracting referee document case-related figure information, constructing case-related figure images, and generating a referee document case-related figure image database;
s3: performing word embedding training on the corpus by using a BERT (belief transfer) aiming at the judging document element labeling set and existing labeling data of a judging document case-related human figure image database, generating a word weight matrix by using a bidirectional LSTM (least squares metric) model based on an attention mechanism, and training a judging document element feature extraction model based on the word weight matrix;
s4; and (3) aiming at the referee document element labeling set and the referee document case-related person feature image database, using a referee document element feature extraction model to extract element features of the linguistic data in the whole database, generating referee document images according to different elements, and further generating a referee document image database.
6. The multi-user case retrieval system based on multi-dimensional semantic union modeling according to claim 4, wherein the element identification module implements functions by the steps of:
s1: performing data preprocessing on an input text of a user, specifically, dividing the input text into a list consisting of single characters;
s2: aiming at the result of data preprocessing, using a trained element feature extraction model BERT + BilSTM + Attention to carry out element identification to obtain an identification result set of various elements;
s3: finally, determining the elements with the threshold value higher than the screening value as the elements extracted from the input text in the element identification result set;
s4: the extraction of the user character feature of the consulting user is performed according to the information structure of the case personnel portrait.
7. The multi-user case retrieval system based on multi-dimensional semantic union modeling according to claim 1, wherein the matching query module implements functions as follows:
s1: element identification and sequencing, namely performing portrait matching on elements extracted from a text input by a consultation user, acquiring a plurality of referee documents in the front of the element identification and sequencing from a referee document portrait library, and calculating the similarity sim _ label of the referee document portrait;
s2: calculating similarity sim _ person between the user and the portrait in the prior referee document and case personnel portrait library according to the user portrait of the consultant user;
s3: according to the referee document labels and the figure image labels extracted by the element feature extraction model, calculating the similarity sim _ score between the referee document labels and all documents in the referee document library by using an ElasticSearch database through a joint calculation formula, and returning a plurality of corresponding referee documents according to the sim _ score descending order;
s4: and returning a plurality of referee documents with top scores in the similarity sim _ score in the matching process to the user.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010747457.4A CN111625626B (en) | 2020-07-30 | 2020-07-30 | Multi-user case retrieval system based on multi-dimensional semantic combined modeling |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010747457.4A CN111625626B (en) | 2020-07-30 | 2020-07-30 | Multi-user case retrieval system based on multi-dimensional semantic combined modeling |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111625626A true CN111625626A (en) | 2020-09-04 |
CN111625626B CN111625626B (en) | 2020-12-18 |
Family
ID=72271528
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010747457.4A Active CN111625626B (en) | 2020-07-30 | 2020-07-30 | Multi-user case retrieval system based on multi-dimensional semantic combined modeling |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111625626B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113177106A (en) * | 2021-05-19 | 2021-07-27 | 苏州黑云智能科技有限公司 | Similar decision book retrieval method based on element extraction |
CN113486158A (en) * | 2021-09-08 | 2021-10-08 | 中国司法大数据研究院有限公司 | Case situation comparison-based case retrieval method, device, equipment and storage medium |
CN117290384A (en) * | 2023-11-27 | 2023-12-26 | 同方赛威讯信息技术有限公司 | Graphic and text retrieval system and method based on combination of big data and computer vision |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104636430A (en) * | 2014-12-30 | 2015-05-20 | 东软集团股份有限公司 | Case knowledge base representation and case similarity obtaining method and system |
CN107247743A (en) * | 2017-05-17 | 2017-10-13 | 安徽富驰信息技术有限公司 | A kind of judicial class case search method and system |
CN108038091A (en) * | 2017-10-30 | 2018-05-15 | 上海思贤信息技术股份有限公司 | A kind of similar calculating of judgement document's case based on figure and search method and system |
CN110019697A (en) * | 2017-08-29 | 2019-07-16 | 北京国双科技有限公司 | A kind of method for pushing and device of criminal document |
CN111340555A (en) * | 2020-02-29 | 2020-06-26 | 重庆百事得大牛机器人有限公司 | Suggestion decision-making system and method based on user portrait model in legal field |
-
2020
- 2020-07-30 CN CN202010747457.4A patent/CN111625626B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104636430A (en) * | 2014-12-30 | 2015-05-20 | 东软集团股份有限公司 | Case knowledge base representation and case similarity obtaining method and system |
CN107247743A (en) * | 2017-05-17 | 2017-10-13 | 安徽富驰信息技术有限公司 | A kind of judicial class case search method and system |
CN110019697A (en) * | 2017-08-29 | 2019-07-16 | 北京国双科技有限公司 | A kind of method for pushing and device of criminal document |
CN108038091A (en) * | 2017-10-30 | 2018-05-15 | 上海思贤信息技术股份有限公司 | A kind of similar calculating of judgement document's case based on figure and search method and system |
CN111340555A (en) * | 2020-02-29 | 2020-06-26 | 重庆百事得大牛机器人有限公司 | Suggestion decision-making system and method based on user portrait model in legal field |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113177106A (en) * | 2021-05-19 | 2021-07-27 | 苏州黑云智能科技有限公司 | Similar decision book retrieval method based on element extraction |
CN113177106B (en) * | 2021-05-19 | 2024-04-12 | 苏州黑云智能科技有限公司 | Similar decision book retrieval method based on element extraction |
CN113486158A (en) * | 2021-09-08 | 2021-10-08 | 中国司法大数据研究院有限公司 | Case situation comparison-based case retrieval method, device, equipment and storage medium |
CN113486158B (en) * | 2021-09-08 | 2021-12-14 | 中国司法大数据研究院有限公司 | Case situation comparison-based case retrieval method, device, equipment and storage medium |
CN117290384A (en) * | 2023-11-27 | 2023-12-26 | 同方赛威讯信息技术有限公司 | Graphic and text retrieval system and method based on combination of big data and computer vision |
CN117290384B (en) * | 2023-11-27 | 2024-02-02 | 同方赛威讯信息技术有限公司 | Graphic and text retrieval system and method based on combination of big data and computer vision |
Also Published As
Publication number | Publication date |
---|---|
CN111625626B (en) | 2020-12-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111625626B (en) | Multi-user case retrieval system based on multi-dimensional semantic combined modeling | |
CN110119786B (en) | Text topic classification method and device | |
CN109271529B (en) | Method for constructing bilingual knowledge graph of Xilier Mongolian and traditional Mongolian | |
Srihari | Automatic indexing and content-based retrieval of captioned images | |
CN113283551A (en) | Training method and training device of multi-mode pre-training model and electronic equipment | |
CN111382565B (en) | Emotion-reason pair extraction method and system based on multiple labels | |
CN107463658B (en) | Text classification method and device | |
CN110334213B (en) | Method for identifying time sequence relation of Hanyue news events based on bidirectional cross attention mechanism | |
US8190538B2 (en) | Methods and systems for matching records and normalizing names | |
RU2760471C1 (en) | Methods and systems for identifying fields in a document | |
CN111222305A (en) | Information structuring method and device | |
CN102663129A (en) | Medical field deep question and answer method and medical retrieval system | |
Bordes et al. | Incorporating visual semantics into sentence representations within a grounded space | |
CN107679070B (en) | Intelligent reading recommendation method and device and electronic equipment | |
CN111858896B (en) | Knowledge base question-answering method based on deep learning | |
Benavent et al. | Multimedia information retrieval based on late semantic fusion approaches: Experiments on a wikipedia image collection | |
CN110532480B (en) | Knowledge graph construction method for recommending human-read threat information and threat information recommendation method | |
Luo et al. | Who’s doing what: Joint modeling of names and verbs for simultaneous face and pose annotation | |
CN109446399A (en) | A kind of video display entity search method | |
CN114997288A (en) | Design resource association method | |
CN114881043A (en) | Deep learning model-based legal document semantic similarity evaluation method and system | |
CN113761192A (en) | Text processing method, text processing device and text processing equipment | |
CN115879901B (en) | Intelligent personnel self-service platform | |
CN117216187A (en) | Semantic intelligent retrieval method for constructing legal knowledge graph based on terms | |
CN114707615B (en) | Ancient character similarity quantification method based on duration Chinese character knowledge graph |
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 | ||
CB03 | Change of inventor or designer information |
Inventor after: Liu Guangfeng Inventor after: Lu Sifan Inventor after: Xia Fei Inventor before: Liu Guangfeng Inventor before: Lu Sifan |
|
CB03 | Change of inventor or designer information |