CN-112507110-B - Method, device and storage medium for recommending related information of complaint cases

CN112507110BCN 112507110 BCN112507110 BCN 112507110BCN-112507110-B

Abstract

The embodiment of the specification discloses a related information recommendation method, device and storage medium for a prosecution case, wherein the method comprises the steps of inputting a prosecution book of the prosecution case, dividing the content of the prosecution book, structurally analyzing case information and crime fact information of the prosecution case, screening case information identical to the case information from a case knowledge base, calculating first similarity of the crime fact information and the case information, and taking the case information with the first similarity higher than a first preset threshold value as first recommendation information of the prosecution case. The scheme can reduce the inquiry workload of the related information of the complaint case, improves the efficiency of case handling of the inspection officer, and can also improve the accuracy of information recommendation.

Inventors

LIU LEI
ZHANG LI
WANG XUCHU
CAI NA

Assignees

北京计算机技术及应用研究所

Dates

Publication Date: 20260512
Application Date: 20201215

Claims (3)

1. The related information recommendation method for the complaint case is characterized by comprising the following steps: Inputting a prosecution book of prosecution cases; dividing the content of the prosecution book, and structurally analyzing the case information and crime facts of the prosecution case; screening case information which is the same as the case information from a case knowledge base; calculating a first similarity between the crime fact information and the case information; Taking the case information with the first similarity higher than a first preset threshold value as first recommendation information of the complaint case; Screening laws and regulations which are the same as the information of the case from a case knowledge base; Calculating a second similarity of the crime facts information to the laws and regulations; taking laws and regulations with the second similarity higher than a second preset threshold value as second recommendation information of the prosecution case; screening expert views identical to the case information from a case knowledge base; Calculating a third similarity of the crime facts information to the expert's view; Taking expert views with the third similarity higher than a third preset threshold value as third recommendation information of the complaint case; selecting the dispute focus and the answer points which are the same as the case information from a case knowledge base; Calculating a fourth similarity of the crime facts information to the dispute focus; Taking the dispute focus and the answering point with the fourth similarity higher than a fourth preset threshold value as fourth recommendation information of the complaint case; The case information comprises a case ID, case information and crime facts, wherein the crime facts and the case information are in one-to-one relation; the laws and regulations comprise a laws and regulations ID, case-by-information and contents of laws and regulations, wherein the case-by-information and the laws and regulations are in one-to-many relation; the expert view comprises an expert view ID, a case information and expert view content, wherein the case information and the expert view content are in one-to-many relation; the dispute focus comprises a dispute focus ID, a counter information, dispute focus content and answer point content, wherein the counter information and the dispute focus content are in one-to-many relationship, and the dispute focus content and the answer point content are in one-to-one relationship.
2. A related information recommending apparatus for a prosecution case, the apparatus comprising: The input module is used for inputting a prosecution book of a prosecution case; The analysis module is used for dividing the content of the prosecution book and structurally analyzing the case information and crime fact information of the prosecution case; the screening module is used for screening case information which is the same as the case information from the case knowledge base; A similarity calculation module, configured to calculate a first similarity between the crime fact information and the case information; The recommending module is used for taking the case information with the first similarity higher than a first preset threshold value as first recommending information of the complaint case; Screening laws and regulations which are the same as the information of the case from a case knowledge base; Calculating a second similarity of the crime facts information to the laws and regulations; taking laws and regulations with the second similarity higher than a second preset threshold value as second recommendation information of the prosecution case; screening expert views identical to the case information from a case knowledge base; Calculating a third similarity of the crime facts information to the expert's view; Taking expert views with the third similarity higher than a third preset threshold value as third recommendation information of the complaint case; selecting the dispute focus and the answer points which are the same as the case information from a case knowledge base; Calculating a fourth similarity of the crime facts information to the dispute focus; Taking the dispute focus and the answering point with the fourth similarity higher than a fourth preset threshold value as fourth recommendation information of the complaint case; The case information comprises a case ID, case information and crime facts, wherein the crime facts and the case information are in one-to-one relation; the laws and regulations comprise a laws and regulations ID, case-by-information and contents of laws and regulations, wherein the case-by-information and the laws and regulations are in one-to-many relation; the expert view comprises an expert view ID, a case information and expert view content, wherein the case information and the expert view content are in one-to-many relation; the dispute focus comprises a dispute focus ID, a counter information, dispute focus content and answer point content, wherein the counter information and the dispute focus content are in one-to-many relationship, and the dispute focus content and the answer point content are in one-to-one relationship.
3. A storage medium having stored thereon computer readable instructions executable by a processor to implement the related information recommendation method of a prosecution case of claim 1.

Description

Method, device and storage medium for recommending related information of complaint cases Technical Field The present application relates to the field of computer technologies, and in particular, to a method and apparatus for recommending relevant information of a complaint case, and a storage medium. Background In the process of checking the case, the inspector needs to build a court-out plan before court trial. In the process of constructing the out-of-court plan, contents such as laws and regulations, expert views, disputes, answering points and the like applicable to the case need to be prepared. At present, similar cases can be generally obtained through a retrieval mode in various case handling auxiliary tools. However, similar case information lacks relevant information such as legal regulations, expert views, disputed focuses, answering points and the like, and when a inspector builds a court plan, the inspector still needs to review relevant information to arrange the information, which takes a lot of time and cannot guarantee accuracy. Disclosure of Invention In view of this, the embodiment of the application provides a method, a device and a storage medium for recommending relevant information of a complaint case, which are used for reducing the inquiry workload of the relevant information, improving the accuracy and improving the efficiency of case handling of a inspector. In order to solve the above technical problems, the embodiments of the present specification are implemented as follows: a related information recommendation method for a prosecution case, the method comprising: Inputting a prosecution book of prosecution cases; dividing the content of the prosecution book, and structurally analyzing the case information and crime facts of the prosecution case; screening case information which is the same as the case information from a case knowledge base; calculating a first similarity between the crime fact information and the case information; and taking the case information with the first similarity higher than a first preset threshold value as first recommendation information of the complaint case. Optionally, the method further comprises: Screening laws and regulations which are the same as the information of the case from a case knowledge base; Calculating a second similarity of the crime facts information to the laws and regulations; And taking laws and regulations with the second similarity higher than a second preset threshold value as second recommendation information of the prosecution case. Optionally, the method further comprises: screening expert views identical to the case information from a case knowledge base; Calculating a third similarity of the crime facts information to the expert's view; and taking the expert point of which the third similarity is higher than a third preset threshold value as third recommendation information of the complaint case. Optionally, the method further comprises: selecting the dispute focus and the answer points which are the same as the case information from a case knowledge base; Calculating a fourth similarity of the crime facts information to the dispute focus; and taking the dispute focus and the answering point with the fourth similarity higher than a fourth preset threshold value as fourth recommendation information of the complaint case. Optionally, the case information comprises a case ID, case information and crime facts, wherein the crime facts and the case information are in one-to-one relationship. Optionally, the laws and regulations comprise a laws and regulations ID, a case by information and a laws and regulations content, wherein the case by information and the laws and regulations are in one-to-many relation. Optionally, the expert view includes an expert view ID, a case by information, and an expert view content, wherein the case by information and the expert view content are in one-to-many relationship. Optionally, the dispute focus comprises a dispute focus ID, a case and a dispute focus content and a debate point content, wherein the case and the dispute focus content are in one-to-many relation, and the dispute focus content and the debate point content are in one-to-one relation. A related information recommending apparatus for a prosecution case, the apparatus comprising: The input module is used for inputting a prosecution book of a prosecution case; The analysis module is used for dividing the content of the prosecution book and structurally analyzing the case information and crime fact information of the prosecution case; the screening module is used for screening case information which is the same as the case information from the case knowledge base; A similarity calculation module, configured to calculate a first similarity between the crime fact information and the case information; and the recommending module is used for taking the case information with the first similarity higher than a first preset threshold va