CN113220888A - Case clue element extraction method and system based on Ernie model - Google Patents
Case clue element extraction method and system based on Ernie model Download PDFInfo
- Publication number
- CN113220888A CN113220888A CN202110609811.1A CN202110609811A CN113220888A CN 113220888 A CN113220888 A CN 113220888A CN 202110609811 A CN202110609811 A CN 202110609811A CN 113220888 A CN113220888 A CN 113220888A
- Authority
- CN
- China
- Prior art keywords
- clue
- elements
- illegal
- ernie model
- text
- 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 title claims abstract description 40
- 230000009471 action Effects 0.000 claims abstract description 26
- 238000000034 method Methods 0.000 claims abstract description 15
- 230000008520 organization Effects 0.000 claims description 10
- 239000003814 drug Substances 0.000 claims description 6
- 229940079593 drug Drugs 0.000 claims description 5
- 230000010354 integration Effects 0.000 claims description 5
- 230000000694 effects Effects 0.000 abstract description 6
- 230000006399 behavior Effects 0.000 abstract description 5
- 230000007547 defect Effects 0.000 abstract description 4
- 230000006870 function Effects 0.000 description 10
- 238000010586 diagram Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000003058 natural language processing Methods 0.000 description 2
- 239000013598 vector Substances 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 210000000056 organ Anatomy 0.000 description 1
- 230000002035 prolonged effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012216 screening 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
- G06F16/35—Clustering; Classification
-
- 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
- G06F16/335—Filtering based on additional data, e.g. user or group profiles
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
- G06F40/295—Named entity recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- 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)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Business, Economics & Management (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Tourism & Hospitality (AREA)
- Databases & Information Systems (AREA)
- Technology Law (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Machine Translation (AREA)
Abstract
The invention provides a case clue element extraction method and system based on an Ernie model, relating to the technical field of computers, and the method comprises the following steps: inputting the clue text into a clue classification unit to obtain the clue type of the clue text; the clue classification unit is formed by matching an Ernie model and keywords, adjusts the weight output by the Ernie model and classifies according to the weight; dividing the clue text into a single sentence set S1, sequentially inputting the single sentence set S1 into a named entity identification unit, and identifying entities in the named entity identification unit; sequentially inputting elements in the simple sentence set S1 into an illegal action and illegal consequence extraction unit to obtain illegal action elements and illegal consequence elements; and organizing and integrating information according to the elements, organizing and integrating the clue type, the entity, the illegal action elements and the illegal consequence elements, and acquiring an element extraction result. The method and the device can overcome the defects that the prior art is low in extraction precision and easy in missing clue elements or extracting clue elements by mistake, and can improve the extraction effect of illegal behaviors and illegal consequence elements.
Description
Technical Field
The invention relates to the technical field of computers, in particular to a case clue element extraction method and system based on an Ernie model.
Background
With the great increase of the number of cases in the public welfare actions, the number of clues in the public welfare actions is also greatly increased, so that the requirement for case handling is difficult to meet only by means of the analysis of clues by inspection officers. On one hand, the public welfare litigation cases have various types, and the efficiency of manually analyzing clues is low, so that a detection organ cannot intervene in time, and the damage time is prolonged. On the other hand, the accuracy of manual analysis of clues is not sufficient, and misjudgment is easily caused. Therefore, it is necessary to introduce information technology means such as artificial intelligence and natural language processing to assist the inspection institution in analyzing the clues of fair litigation.
At present, the main methods for extracting the clue elements of the public welfare litigation also focus on non-intelligent means such as keywords and the like. The prior Chinese patent with publication number CN112270633A discloses a public welfare litigation clue judging system and method based on big data drive, which comprises the following steps: aiming at different source channels and data characteristics, formulating corresponding data acquisition schemes, automatically and dynamically acquiring case source information related to the fair litigation from a plurality of channels, and integrating, cleaning and converting the case source information to form a fair litigation case source library; based on technologies such as big data and natural language processing, a public welfare lition clue studying and judging model is constructed, the acquired case source information is automatically and accurately classified, analyzed and studied and judged, a clue studying and judging index is calculated, and clues larger than a preset threshold value are actively pushed to an inspector to be handled. The method realizes the acquisition, treatment, analysis, research and judgment and early warning of mass case source data, effectively expands the way of the fair litigation case source, improves the quality effect of screening the fair litigation clues from the mass case source data, and enhances the timeliness and accuracy of finding the fair litigation clues.
The method takes keyword matching as a means for acquiring factors such as the types, subjects and illegal consequences of the guild litigation clues, and cannot dig out deep semantic features of clue texts, so that the method cannot adapt to complex natural language text corpora, has low extraction precision of clue factors, and is easy to omit the factors in the clues or extract the factors by mistake.
Disclosure of Invention
Aiming at the defects in the prior art, the invention provides a case clue element extraction method and system based on an Ernie model.
According to the case clue element extraction method and system based on the Ernie model, the scheme is as follows:
in a first aspect, a case clue element extraction method based on an Ernie model is provided, and the method includes:
step 1: inputting the clue text into a clue classification unit to obtain the clue type of the clue text; the clue classification unit is formed by matching an Ernie model and keywords, adjusts the weight output by the Ernie model according to the times of matching the keywords in the clue text, and classifies the keywords according to the weight;
step 2: dividing the clue text into a single sentence set S1;
and step 3: sequentially inputting elements in the single sentence set S1 into a named entity recognition unit, and recognizing entities in the line text;
and 4, step 4: sequentially inputting the elements in the single sentence set S1 into an illegal action and illegal consequence extraction unit, and acquiring illegal action elements and illegal consequence elements in a clue text;
and 5: and organizing and integrating the clue type, the entity, the illegal action element and the illegal consequence element according to element organization and integration information to obtain an element extraction result.
Preferably, the categories of the thread types in step 1 include ecological environment, state conservation, food safety and drug safety.
Preferably, the entities in step 3 are names of persons, addresses, company names and organizations.
Preferably, the input of the Ernie model in step 4 includes the position characteristics of the single sentence.
Preferably, the Loss function of the Ernie model in step 4 is a multi-class Focal-local, which enables the Ernie model to focus more on samples that are difficult to classify.
In a second aspect, a case clue element extraction system based on an Ernie model is provided, the system includes:
module 1: inputting the clue text into a clue classification unit to obtain the clue type of the clue text; the clue classification unit is formed by matching an Ernie model and keywords, adjusts the weight output by the Ernie model according to the times of matching the keywords in the clue text, and classifies the keywords according to the weight;
and (3) module 2: dividing the clue text into a single sentence set S1;
and a module 3: sequentially inputting elements in the single sentence set S1 into a named entity recognition unit, and recognizing entities in the line text;
and (4) module: sequentially inputting the elements in the single sentence set S1 into an illegal action and illegal consequence extraction unit, and acquiring illegal action elements and illegal consequence elements in a clue text;
and a module 5: and organizing and integrating the clue type, the entity, the illegal action element and the illegal consequence element according to element organization and integration information to obtain an element extraction result.
Preferably, the types of threads in the module 1 include ecological environment, state protection, food safety and drug safety.
Preferably, the entities in the module 3 are a person name, an address, a company name and an organization name.
Preferably, the input of the Ernie model in the module 4 comprises the position characteristics of the single sentence.
Preferably, the Loss function of the Ernie model in the module 4 is a multi-class Focal-local, which enables the Ernie model to focus more on samples that are difficult to classify.
Compared with the prior art, the invention has the following beneficial effects:
1. the method has the advantages that the pre-trained Ernie model is subjected to fine adjustment by adopting the text in the public welfare suit clue field, so that the model can dig out deep semantic features of clue texts, clue elements can be more accurately extracted from complex public welfare suit clue text corpora, and the defects that the extraction precision is low, the clue elements are easy to miss or the clue elements are extracted mistakenly in the prior art are overcome;
2. the model is optimized by adopting various means, and the classification effect of samples difficult to classify is improved by combining with keyword matching when text classification is carried out; when named entity recognition is carried out, a conditional random field is connected behind the Ernie model, so that the precision of named entity recognition is improved; when illegal behaviors and illegal consequences are extracted, the input layer of the Ernie model is modified, so that the Ernie model can obtain position feature vectors of sentences, and the multi-classification Focal-Loss function is used as a Loss function, so that the extraction effect of illegal behaviors and illegal consequence elements is improved.
Drawings
Other features, objects and advantages of the invention will become more apparent upon reading of the detailed description of non-limiting embodiments with reference to the following drawings:
FIG. 1 is a flow chart of element extraction in an embodiment of the present invention;
FIG. 2 is a diagram of a thread classification unit according to an embodiment of the present application;
FIG. 3 is a diagram illustrating a named entity recognition unit according to an embodiment of the present application;
FIG. 4 is a schematic diagram of an illegal action and illegal consequence extracting unit in the embodiment of the present application.
Detailed Description
The present invention will be described in detail with reference to specific examples. The following examples will assist those skilled in the art in further understanding the invention, but are not intended to limit the invention in any way. It should be noted that it would be obvious to those skilled in the art that various changes and modifications can be made without departing from the spirit of the invention. All falling within the scope of the present invention.
The embodiment of the invention provides a case clue element extraction method based on an Ernie model, which is shown in a reference figure 1 and a reference figure 2 and comprises the following specific steps:
firstly, a clue text is input into a clue classification unit, and the clue type of the clue text is obtained; the clue classification unit is formed by matching an Ernie model and keywords, adjusts the weight output by the Ernie model according to the times of matching the keywords in the clue text, and classifies the keywords according to the weight; dividing the clue text into a single sentence set S1; the cable type in this embodiment is classified into five categories, i.e., ecological environment, national resource protection, homeland protection, food safety, and drug safety.
Secondly, referring to fig. 3 and 4, the elements in the single sentence set S1 are sequentially input into the named entity recognition unit to recognize the entities in the line text; the named entity recognition unit is composed of an Ernie model and a conditional random field, and the conditional random field finds the most appropriate label from the output result of the Ernie model according to the part of speech rule; the entities in this embodiment are a person name, an address, a company name, and an organization name.
Subsequently, the elements in the simple sentence set S1 are sequentially input into an illegal action and illegal consequence extraction unit, and illegal action elements and illegal consequence elements in the clue text are obtained; the illegal action and illegal consequence extraction unit is composed of an Ernie model and a full connection layer. Wherein the input of the Ernie model includes the position characteristics of the single sentence; the Loss function of the Ernie model is a multi-class Focal-local that allows the Ernie model to focus more on samples that are difficult to classify.
And finally, organizing and integrating information according to the elements, organizing and integrating the clue type, the entity, the illegal action element and the illegal consequence element, and acquiring an element extraction result.
The invention also provides a case clue element extraction system based on the Ernie model, which specifically comprises the following modules:
module 1: inputting the clue text into a clue classification unit to obtain the clue type of the clue text; the clue classification unit is formed by matching an Ernie model and keywords, adjusts the weight output by the Ernie model according to the times of matching the keywords in the clue text, and classifies the keywords according to the weight.
And (3) module 2: dividing the clue text into a single sentence set S1;
and a module 3: sequentially inputting elements in the single sentence set S1 into a named entity recognition unit, and recognizing entities in the line text; the named entity recognition unit is composed of an Ernie model and a conditional random field, and the conditional random field finds the most appropriate label from the output result of the Ernie model according to the part of speech rule.
And (4) module: sequentially inputting elements in the simple sentence set S1 into an illegal action and illegal consequence extraction unit, and acquiring illegal action elements and illegal consequence elements in the clue text; the illegal action and illegal consequence extraction unit is composed of an Ernie model and a full connection layer.
And a module 5: and organizing and integrating the clue type, the entity, the illegal action element and the illegal consequence element according to element organization and integration information to obtain an element extraction result.
Further, in the module 1, the thread types are classified into five categories, such as ecological environment, national resource protection, homeland protection, food safety, and medicine safety. The entities mentioned in module 3 are the name of a person, address, company name and organization name.
The input of the Ernie model in the module 4 includes the position characteristics of a single sentence, and the Loss function of the Ernie model is multi-class Focal-local, which enables the Ernie model to pay more attention to samples which are difficult to classify.
The embodiment of the invention provides a case clue element extraction method and system based on an Ernie model, which are used for finely adjusting a text in a public litigation clue field of the pretrained Ernie model, so that the model can dig out deep semantic features of clue texts, clue elements can be more accurately extracted from complex public litigation clue text corpora, and the defects that the extraction precision is low, the clue elements are easy to omit or the clue elements are extracted mistakenly in the prior art are overcome; meanwhile, a model is optimized by adopting various means, and the classification effect of samples difficult to classify is improved by combining with keyword matching when text classification is carried out; when named entity recognition is carried out, a conditional random field is connected behind the Ernie model, and the named entity recognition accuracy is improved. When illegal behaviors and illegal consequences are extracted, the input layer of the Ernie model is modified, so that the Ernie model can obtain position feature vectors of sentences, and the multi-classification Focal-Loss function is used as a Loss function, so that the extraction effect of illegal behaviors and illegal consequence elements is improved.
Those skilled in the art will appreciate that, in addition to implementing the system and its various devices, modules, units provided by the present invention as pure computer readable program code, the system and its various devices, modules, units provided by the present invention can be fully implemented by logically programming method steps in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers and the like. Therefore, the system and various devices, modules and units thereof provided by the invention can be regarded as a hardware component, and the devices, modules and units included in the system for realizing various functions can also be regarded as structures in the hardware component; means, modules, units for performing the various functions may also be regarded as structures within both software modules and hardware components for performing the method.
The foregoing description of specific embodiments of the present invention has been presented. It is to be understood that the present invention is not limited to the specific embodiments described above, and that various changes or modifications may be made by one skilled in the art within the scope of the appended claims without departing from the spirit of the invention. The embodiments and features of the embodiments of the present application may be combined with each other arbitrarily without conflict.
Claims (10)
1. A case clue element extraction method based on an Ernie model is characterized by comprising the following steps:
step 1: inputting the clue text into a clue classification unit to obtain the clue type of the clue text; the clue classification unit is formed by matching an Ernie model and keywords, adjusts the weight output by the Ernie model according to the times of matching the keywords in the clue text, and classifies the keywords according to the weight;
step 2: dividing the clue text into a single sentence set S1;
and step 3: sequentially inputting elements in the single sentence set S1 into a named entity recognition unit, and recognizing entities in the line text;
and 4, step 4: sequentially inputting the elements in the single sentence set S1 into an illegal action and illegal consequence extraction unit, and acquiring illegal action elements and illegal consequence elements in a clue text;
and 5: and organizing and integrating the clue type, the entity, the illegal action element and the illegal consequence element according to element organization and integration information to obtain an element extraction result.
2. The method for extracting case thread elements based on Ernie model as claimed in claim 1, wherein the categories of thread types in step 1 include ecological environment, state conservation, food safety and drug safety.
3. The method as claimed in claim 1, wherein the entities in step 3 are names of people, addresses, company names and organization names.
4. The method for extracting case clue elements based on Ernie model as claimed in claim 1, wherein the input of Ernie model in step 4 includes the position features of a single sentence.
5. The method for extracting case clue elements based on the Ernie model as claimed in claim 1, wherein the Loss function of the Ernie model in step 4 is a multi-class Focal-local that allows the Ernie model to focus more on samples that are difficult to classify.
6. A case clue element extraction system based on an Ernie model is characterized by comprising the following components:
module 1: inputting the clue text into a clue classification unit to obtain the clue type of the clue text; the clue classification unit is formed by matching an Ernie model and keywords, adjusts the weight output by the Ernie model according to the times of matching the keywords in the clue text, and classifies the keywords according to the weight;
and (3) module 2: dividing the clue text into a single sentence set S1;
and a module 3: sequentially inputting elements in the single sentence set S1 into a named entity recognition unit, and recognizing entities in the line text;
and (4) module: sequentially inputting the elements in the single sentence set S1 into an illegal action and illegal consequence extraction unit, and acquiring illegal action elements and illegal consequence elements in a clue text;
and a module 5: and organizing and integrating the clue type, the entity, the illegal action element and the illegal consequence element according to element organization and integration information to obtain an element extraction result.
7. The Ernie model-based case clue element extraction system as claimed in claim 6, wherein the category of clue type in the module 1 includes ecological environment, state conservation, food safety and drug safety.
8. The Ernie model-based case clue element extraction system as claimed in claim 6, wherein the entities in the module 3 are names of people, addresses, company names and organization names.
9. The Ernie model-based case clue element extraction system as claimed in claim 6, wherein the input of Ernie model in module 4 includes the position feature of a single sentence.
10. The system for extracting case clue elements based on Ernie model as claimed in claim 6, wherein the Loss function of Ernie model in module 4 is multi-class Focal-local, which can make Ernie model focus more on samples that are difficult to classify.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110609811.1A CN113220888B (en) | 2021-06-01 | 2021-06-01 | Case clue element extraction method and system based on Ernie model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110609811.1A CN113220888B (en) | 2021-06-01 | 2021-06-01 | Case clue element extraction method and system based on Ernie model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113220888A true CN113220888A (en) | 2021-08-06 |
| CN113220888B CN113220888B (en) | 2022-12-13 |
Family
ID=77082228
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110609811.1A Active CN113220888B (en) | 2021-06-01 | 2021-06-01 | Case clue element extraction method and system based on Ernie model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113220888B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117668499A (en) * | 2024-01-31 | 2024-03-08 | 平潭综合实验区智慧岛投资发展有限公司 | Ocean public service litigation thread research and judgment method based on machine learning |
Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120290950A1 (en) * | 2011-05-12 | 2012-11-15 | Jeffrey A. Rapaport | Social-topical adaptive networking (stan) system allowing for group based contextual transaction offers and acceptances and hot topic watchdogging |
| CN105956797A (en) * | 2016-06-21 | 2016-09-21 | 中国***关岭布依族苗族自治县纪律检查委员会 | Big data-based party style and clean government construction informatization management system |
| CN111160017A (en) * | 2019-12-12 | 2020-05-15 | 北京文思海辉金信软件有限公司 | Keyword extraction method, phonetics scoring method and phonetics recommendation method |
| CN111475542A (en) * | 2020-03-26 | 2020-07-31 | 中国***江安县纪律检查委员会 | Quadruple linkage violation information interaction platform and monitoring method |
| CN111522955A (en) * | 2020-04-29 | 2020-08-11 | 深圳市华云中盛科技股份有限公司 | Litigation case classification method and device, computer equipment and storage medium |
| CN111783420A (en) * | 2020-06-19 | 2020-10-16 | 上海交通大学 | Anti-complaint book element extraction method, system, medium and device based on BERT model |
| CN112163405A (en) * | 2020-09-08 | 2021-01-01 | 北京百度网讯科技有限公司 | Question generation method and device |
| CN112380993A (en) * | 2020-11-12 | 2021-02-19 | 青岛以萨数据技术有限公司 | Intelligent illegal behavior detection system and method based on target real-time tracking information |
| CN112417880A (en) * | 2020-11-30 | 2021-02-26 | 太极计算机股份有限公司 | Court electronic file oriented case information automatic extraction method |
| CN112434535A (en) * | 2020-11-24 | 2021-03-02 | 上海浦东发展银行股份有限公司 | Multi-model-based factor extraction method, device, equipment and storage medium |
| CN112507099A (en) * | 2020-12-18 | 2021-03-16 | 北京百度网讯科技有限公司 | Training method, device, equipment and storage medium of dialogue understanding model |
| CN112507118A (en) * | 2020-12-22 | 2021-03-16 | 北京百度网讯科技有限公司 | Information classification and extraction method and device and electronic equipment |
| US10957310B1 (en) * | 2012-07-23 | 2021-03-23 | Soundhound, Inc. | Integrated programming framework for speech and text understanding with meaning parsing |
| CN112711700A (en) * | 2019-10-24 | 2021-04-27 | 富驰律法(北京)科技有限公司 | Method and system for recommending case for fair litigation |
-
2021
- 2021-06-01 CN CN202110609811.1A patent/CN113220888B/en active Active
Patent Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120290950A1 (en) * | 2011-05-12 | 2012-11-15 | Jeffrey A. Rapaport | Social-topical adaptive networking (stan) system allowing for group based contextual transaction offers and acceptances and hot topic watchdogging |
| US10957310B1 (en) * | 2012-07-23 | 2021-03-23 | Soundhound, Inc. | Integrated programming framework for speech and text understanding with meaning parsing |
| CN105956797A (en) * | 2016-06-21 | 2016-09-21 | 中国***关岭布依族苗族自治县纪律检查委员会 | Big data-based party style and clean government construction informatization management system |
| CN112711700A (en) * | 2019-10-24 | 2021-04-27 | 富驰律法(北京)科技有限公司 | Method and system for recommending case for fair litigation |
| CN111160017A (en) * | 2019-12-12 | 2020-05-15 | 北京文思海辉金信软件有限公司 | Keyword extraction method, phonetics scoring method and phonetics recommendation method |
| CN111475542A (en) * | 2020-03-26 | 2020-07-31 | 中国***江安县纪律检查委员会 | Quadruple linkage violation information interaction platform and monitoring method |
| CN111522955A (en) * | 2020-04-29 | 2020-08-11 | 深圳市华云中盛科技股份有限公司 | Litigation case classification method and device, computer equipment and storage medium |
| CN111783420A (en) * | 2020-06-19 | 2020-10-16 | 上海交通大学 | Anti-complaint book element extraction method, system, medium and device based on BERT model |
| CN112163405A (en) * | 2020-09-08 | 2021-01-01 | 北京百度网讯科技有限公司 | Question generation method and device |
| CN112380993A (en) * | 2020-11-12 | 2021-02-19 | 青岛以萨数据技术有限公司 | Intelligent illegal behavior detection system and method based on target real-time tracking information |
| CN112434535A (en) * | 2020-11-24 | 2021-03-02 | 上海浦东发展银行股份有限公司 | Multi-model-based factor extraction method, device, equipment and storage medium |
| CN112417880A (en) * | 2020-11-30 | 2021-02-26 | 太极计算机股份有限公司 | Court electronic file oriented case information automatic extraction method |
| CN112507099A (en) * | 2020-12-18 | 2021-03-16 | 北京百度网讯科技有限公司 | Training method, device, equipment and storage medium of dialogue understanding model |
| CN112507118A (en) * | 2020-12-22 | 2021-03-16 | 北京百度网讯科技有限公司 | Information classification and extraction method and device and electronic equipment |
Non-Patent Citations (7)
| Title |
|---|
| JIAHONG YUAN等: "Pauses for Detection of Alzheimer"s Disease", 《FRONTIERS IN COMPUTER SCIENCE》, 18 April 2021 (2021-04-18) * |
| JIAN YUAN等: "Overview of SMP-CAIL2020-Argmine:The Interactive Argument-Pair Extraction in Judgement Document Challenge", 《DATA INTELLIGENCE》, 15 May 2021 (2021-05-15) * |
| LINGYUN ZHAO等: "A BERT based Sentiment Analysis and Key Entity Detection Approach for Online Financial Texts", 《2021 IEEE 24TH INTERNATIONAL CONFERENCE ON COMPUTER SUPPORTED COOPERATIVE WORK IN DESIGN (CSCWD)》, 28 May 2021 (2021-05-28) * |
| 刘文臻: "中文文本多标签分类算法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》, 15 July 2020 (2020-07-15) * |
| 张月国: "基于知网语义相似度计算的文本特征提取", 《信息安全与通信保密》, 10 May 2009 (2009-05-10) * |
| 李凌云: "从损害控制到风险预防：野生动物保护公益诉讼的优化进路", 《中国环境管理》 * |
| 李凌云: "从损害控制到风险预防：野生动物保护公益诉讼的优化进路", 《中国环境管理》, no. 05, 25 October 2020 (2020-10-25) * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117668499A (en) * | 2024-01-31 | 2024-03-08 | 平潭综合实验区智慧岛投资发展有限公司 | Ocean public service litigation thread research and judgment method based on machine learning |
| CN117668499B (en) * | 2024-01-31 | 2024-05-14 | 平潭综合实验区智慧岛投资发展有限公司 | Ocean public service litigation thread research and judgment method, system, equipment and medium based on machine learning |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113220888B (en) | 2022-12-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Hopp et al. | The extended Moral Foundations Dictionary (eMFD): Development and applications of a crowd-sourced approach to extracting moral intuitions from text | |
| Rubin et al. | Certainty identification in texts: Categorization model and manual tagging results | |
| CN107491435B (en) | Method and device for automatically identifying user emotion based on computer | |
| Isa et al. | Indobert for indonesian fake news detection | |
| Kumar et al. | Random forest-based sarcastic tweet classification using multiple feature collection | |
| Elayidom et al. | Text classification for authorship attribution analysis | |
| WO2016036345A1 (en) | External resource identification | |
| Maynard et al. | Multimodal sentiment analysis of social media | |
| Alqurashi | Arabic sentiment analysis for twitter data: a systematic literature review | |
| CN113220888B (en) | Case clue element extraction method and system based on Ernie model | |
| Ibrahim et al. | An explainable AI model for hate speech detection on Indonesian twitter | |
| Joharee et al. | Sentiment Analysis and Text Classification for Depression Detection | |
| Ndabula et al. | Detection of hate speech code mix involving english and other nigerian languages | |
| Thanos et al. | Combined deep learning and traditional NLP approaches for fire burst detection based on twitter posts | |
| Lasri | Suicide ideation detection on social networks: short literature review | |
| Perez-Martin et al. | Semantic search of memes on twitter | |
| Abdulkadhar et al. | Recurrent convolution neural networks for classification of protein-protein interaction articles from biomedical literature | |
| Sadek et al. | Learning causality for Arabic-proclitics | |
| Shawkat | Evaluation of Different Machine Learning, Deep Learning and Text Processing Techniques for Hate Speech Detection | |
| Shupta et al. | An Adaptive Approach to Detecting Fake News Based on Generalized Text Features. | |
| Wang et al. | Can Generative AI Eliminate Speech Harms? A Study on Detection of Abusive and Hate Speech during the COVID-19 Pandemic | |
| Ramadhanti et al. | Topic Detection on Twitter Using Deep Learning Method with Feature Expansion GloVe | |
| Awwad et al. | Non-vocalised Arabic word classifications based on mining affixes features | |
| Diesenreiter et al. | ProperBERT-Proactive Recognition of Offensive Phrasing for Effective Regulation | |
| Uthayashangar et al. | Taxonomy of keyword extraction in Facebook using Decision Tree algorithm in NLP |
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 |