CN110377632B - Litigation result prediction method, litigation result prediction device, litigation result prediction computer device and litigation result prediction storage medium - Google Patents
Litigation result prediction method, litigation result prediction device, litigation result prediction computer device and litigation result prediction storage medium Download PDFInfo
- Publication number
- CN110377632B CN110377632B CN201910520319.XA CN201910520319A CN110377632B CN 110377632 B CN110377632 B CN 110377632B CN 201910520319 A CN201910520319 A CN 201910520319A CN 110377632 B CN110377632 B CN 110377632B
- Authority
- CN
- China
- Prior art keywords
- case
- evidence
- type
- prediction
- focus
- 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
- 238000000034 method Methods 0.000 title claims abstract description 44
- 239000013598 vector Substances 0.000 claims abstract description 88
- 238000012549 training Methods 0.000 claims description 24
- 238000004590 computer program Methods 0.000 claims description 12
- 238000000605 extraction Methods 0.000 claims description 5
- 238000010276 construction Methods 0.000 claims description 4
- 239000000463 material Substances 0.000 claims description 3
- 239000000203 mixture Substances 0.000 claims description 3
- 238000010801 machine learning Methods 0.000 abstract description 3
- 238000004458 analytical method Methods 0.000 description 46
- 238000013145 classification model Methods 0.000 description 7
- 230000014509 gene expression Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 238000012512 characterization method Methods 0.000 description 3
- 238000012790 confirmation Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 230000001427 coherent effect Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000005484 gravity Effects 0.000 description 2
- 238000003032 molecular docking Methods 0.000 description 2
- 230000000306 recurrent effect Effects 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2462—Approximate or statistical queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/248—Presentation of query results
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- 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)
- Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Probability & Statistics with Applications (AREA)
- Health & Medical Sciences (AREA)
- Tourism & Hospitality (AREA)
- Databases & Information Systems (AREA)
- Human Resources & Organizations (AREA)
- Molecular Biology (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Business, Economics & Management (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Strategic Management (AREA)
- Primary Health Care (AREA)
- Marketing (AREA)
- Economics (AREA)
- Fuzzy Systems (AREA)
- Technology Law (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application relates to a litigation result prediction method, a litigation result prediction device, a litigation result prediction computer device and a litigation result prediction storage medium based on machine learning. The method comprises the following steps: receiving a litigation result prediction request sent by a terminal; the litigation result prediction request carries case information and evidence combination of the current case; determining a evidence type corresponding to each evidence item in the evidence combination; generating an evidence feature vector of the current case based on the evidence type; determining the case type and legal relation of the current case according to the case information; and acquiring a target prediction model corresponding to the case type and legal relation, and inputting the evidence feature vector into the target prediction model to obtain the probability of complaint of the current case. By adopting the method, the litigation result prediction efficiency and accuracy can be improved.
Description
Technical Field
The present invention relates to the field of computer technologies, and in particular, to a litigation result prediction method, a litigation result prediction device, a computer device, and a storage medium.
Background
Along with the rapid spread of legal information, the legal consciousness of people is gradually improved, and the legal requirements are also gradually increased. However, the general public generally has little knowledge about legal knowledge with strong professionals, so that the difficulty of searching and analyzing legal related information by themselves is increased. For example, to avoid spending a lot of time and effort to complain in the event that the evidence strength is insufficient to support the complaint, the user desires to know in advance the probability of complaint for the current case based on the evidence currently mastered. However, at present, a technology for automatically predicting the case complaint probability is lacking, so that the case complaint probability prediction is low in efficiency and the objective accuracy of the prediction result is low.
Disclosure of Invention
In view of the above, it is necessary to provide a litigation result prediction method, device, computer device, and storage medium that can improve the litigation result prediction efficiency and accuracy.
A litigation outcome prediction method, the method comprising: receiving a litigation result prediction request sent by a terminal; the litigation result prediction request carries case information and evidence combination of the current case; determining a evidence type corresponding to each evidence item in the evidence combination; generating an evidence feature vector of the current case based on the evidence type; determining the case type and legal relation of the current case according to the case information; and acquiring a target prediction model corresponding to the case type and legal relation, and inputting the evidence feature vector into the target prediction model to obtain the probability of complaint of the current case.
In one embodiment, the method further comprises: receiving an evidence chain guide request sent by a terminal; the evidence chain guide request carries a case identification; acquiring the case information of the current case according to the case identifier; extracting the dispute focus of the current case from the case information; determining a focus type of the dispute focus by fuzzy matching; acquiring one or more associated evidence items and decision weights corresponding to each evidence item according to the case type, legal relation and focus type of the current case; generating an evidence label corresponding to the corresponding evidence item according to the decision weight; an evidence chain is generated based on a plurality of evidence items with evidence labels, and the evidence chain is returned to the terminal.
In one embodiment, before the obtaining the target prediction model corresponding to the case type and the target legal relation, the method further includes: acquiring a plurality of historical cases and case factors corresponding to each historical case; the case factors comprise case types and legal relations; grouping a plurality of historical cases according to the case types and legal relations; coding the case factors to obtain feature vectors of corresponding historical cases; constructing a training set corresponding to the corresponding case type and legal relation based on the feature vector of each group of historical cases; and training the basic prediction model based on different training sets to obtain target prediction models corresponding to different case types and legal relations.
In one embodiment, the case factor includes evidence combinations; the obtaining a plurality of historical cases and case factors corresponding to each historical case includes: acquiring case files of a plurality of historical cases; extracting one or more evidence description sentences from the case file through regular matching; inputting the evidence description sentences into a preset semantic understanding model to obtain one or more evidence items corresponding to each evidence description sentence; and identifying the evidence type corresponding to each evidence item, and generating the evidence combination of the corresponding historical case based on the evidence type.
In one embodiment, the inputting the evidence feature vector into the target prediction model includes: extracting the dispute focus of the current case from the case information; determining a focus type of the dispute focus by fuzzy matching; generating a focus characteristic vector of the current case based on the focus type; splicing the focus feature vector and the evidence feature vector to obtain a case feature vector; and inputting the case feature vector into the target prediction model.
In one embodiment, the target prediction model includes a first prediction model and a second prediction model; the obtaining a target prediction model corresponding to the case type and legal relation, inputting the evidence feature vector into the target prediction model to obtain the probability of complaint of the current case, comprising: acquiring a corresponding first prediction model according to the case type and legal relation, and inputting the evidence feature vector into the first prediction model to obtain a first prediction value; extracting a dispute focus of the current case from the case information, determining a focus type of the dispute focus, and acquiring a corresponding second prediction model according to the focus type; generating a focus characteristic vector of the current case based on the focus type, and inputting the focus characteristic vector into the second prediction model to obtain a second prediction value; determining the prediction weights respectively corresponding to the first prediction model and the second prediction model according to the case types and the legal relation; and carrying out preset logic operation on the first predicted value and the second predicted value according to the predicted weight to obtain the complaint probability of the current case.
A litigation outcome prediction device, the device comprising: the case prediction request module is used for receiving a litigation result prediction request sent by the terminal; the litigation result prediction request carries case information and evidence combination of the current case; the case feature extraction module is used for determining the evidence type corresponding to each evidence item in the evidence combination; generating an evidence feature vector of the current case based on the evidence type; the litigation result prediction module is used for determining the case type and legal relation of the current case according to the case information; and acquiring a target prediction model corresponding to the case type and legal relation, and inputting the evidence feature vector into the target prediction model to obtain the probability of complaint of the current case.
In one embodiment, the device further comprises a evidence chain guiding module, which is used for receiving the evidence chain guiding request sent by the terminal; the evidence chain guide request carries a case identification; acquiring the case information of the current case according to the case identifier; extracting the dispute focus of the current case from the case information; determining a focus type of the dispute focus by fuzzy matching; acquiring one or more associated evidence items and decision weights corresponding to each evidence item according to the case type, legal relation and focus type of the current case; generating an evidence label corresponding to the corresponding evidence item according to the decision weight; an evidence chain is generated based on a plurality of evidence items with evidence labels, and the evidence chain is returned to the terminal.
A computer device comprising a memory storing a computer program and a processor implementing the steps of the litigation outcome prediction method provided in any one of the embodiments of the application when the computer program is executed.
A computer readable storage medium having stored thereon a computer program which when executed by a processor implements the steps of the litigation result prediction method provided in any one of the embodiments of the present application.
According to the litigation result prediction method, device, computer equipment and storage medium, as the user only needs to provide the mastered evidence combination of the current case, the evidence type related to the evidence combination is automatically identified, and the winning probability prediction is automatically carried out based on the preset litigation result prediction model, so that the prediction efficiency can be improved; in addition, different litigation result prediction models are respectively preset aiming at the cases of different case types and legal relations, so that the pertinence of the models can be improved, and the accuracy of the prediction results can be further improved.
Drawings
FIG. 1 is an application scenario diagram of a litigation outcome prediction method in one embodiment;
FIG. 2 is a flow chart of a litigation outcome prediction method in one embodiment;
FIG. 3 is a flow diagram of the steps of evidence chain guidance in one embodiment;
FIG. 4 is a block diagram of a litigation outcome prediction device in one embodiment;
fig. 5 is an internal structural diagram of a computer device in one embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application will be further described in detail with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the present application.
The litigation result prediction method provided by the application can be applied to an application environment shown in fig. 1. Wherein the terminal 102 communicates with the server 104 via a network. The terminal 102 may be, but not limited to, various personal computers, notebook computers, smartphones, tablet computers, and portable wearable devices, and the server 104 may be implemented by a stand-alone server or a server cluster composed of a plurality of servers. When the user needs to know the probability of success or failure of the current case, a litigation outcome prediction request may be sent to server 104 through terminal 102. The litigation result prediction request carries case information and evidence combination of the current case. The case information may be a case file or the like. The evidence composition includes one or more evidence items. The server 104 identifies the type of evidence for each item of evidence. The server 104 generates an evidence feature vector for the current case based on the evidence types of the plurality of evidence items. Based on the case information of the current case, server 104 determines the case type and legal relationship of the current case. Server 104 correspondingly presets different litigation result prediction models for cases of different case types and legal relationships. Server 104 marks the litigation result prediction model corresponding to the case type and legal relation of the current case as a target prediction model, and inputs the evidence feature vector into the target prediction model to obtain the probability of complaint of the current case. According to the litigation result prediction process, a user only needs to provide the mastered evidence combination of the current case, the evidence type related to the evidence combination is automatically identified, and the winning probability prediction is automatically carried out based on the preset litigation result prediction model, so that the prediction efficiency can be improved; in addition, different litigation result prediction models are respectively preset aiming at the cases of different case types and legal relations, so that the pertinence of the models can be improved, and the accuracy of the prediction results can be further improved.
In one embodiment, as shown in fig. 2, a litigation result prediction method is provided, and the method is applied to the server in fig. 1 for illustration, and includes the following steps:
When the user needs to know the winning or losing probability of the current case, the case information of the current case and the evidence combination mastered at present can be filled in based on a litigation service platform on the terminal. The litigation service platform generates a litigation result prediction request based on the filled case information and evidence combination, and sends the litigation result prediction request to the server. Case information may include litigation request books, applicant information, and the like. The applicant information includes not only the name, certificate, etc. of the applicant, but also litigation facts and reasons, evidence, a target amount, a target object, etc.
The evidence composition includes a plurality of evidence items. The evidence item may be a principal statement, a document, a material evidence, audiovisual material, a witness's witness, electronic data, an authentication conclusion, a survey transcript, or the like. The evidence item may be a representation phrase of evidence, for example, in an evidence file of an divorce case, may contain a pre-wedding agreement, a real estate certificate, a double-party banking record, etc. The technician can identify the evidence items contained in the case information by means of semantic identification and the like. The evidence type is a type classified based on the identified evidence item, for example, in the evidence file of the divorce case, the identified pre-marriage agreement may be classified into agreement type evidence, and the real property evidence and the two-party banking record may be classified into property type evidence, and so on.
And the server acquires a preset evidence sequence. The evidence sequence includes a plurality of evidence types in an ordered arrangement. The server obtains evidence feature vectors by setting the mark positions of the corresponding evidence types in the evidence sequence as 1 and the mark positions of the rest evidence types as 0 according to the evidence types related to the current case, and takes the evidence feature vectors as case feature vectors.
The case type can be financial borrowing disputes, contract disputes, inheritance right disputes and the like. Legal relationships may be lending relationships, vouch-for relationships, common repayment relationships, and the like.
The server presets corresponding target prediction models aiming at cases with different case types and legal relations. For example, a corresponding first prediction model may be set for a case with a case type of "financial borrowing disputes" and a legal relationship of "borrowing relationships"; setting a corresponding second prediction model and the like for a case with a case type of financial borrowing disputes and a legal relationship of guarantee relationship. The target prediction model is used for predicting case complaint probability based on case information. The target prediction model may be obtained by machine learning case information of a large number of real historical cases. The server inputs the case feature vector of the current case into the obtained target prediction model, and outputs the probability of the complaint of the current case.
In one embodiment, before obtaining the target prediction model corresponding to the case type and the target legal relation, the method further comprises: acquiring a plurality of historical cases and case factors corresponding to each historical case; the case factors include case type and legal relationship; grouping a plurality of historical cases according to case types and legal relations; coding the case factors to obtain feature vectors of corresponding historical cases; constructing a training set corresponding to the corresponding case type and legal relation based on the feature vector of each group of historical cases; and training the basic prediction model based on different training sets to obtain target prediction models corresponding to different case types and legal relations.
The server acquires case files of a plurality of historical cases and records the case files as historical case files. The historical case file may be a litigation request book, a referee document, or the like. The server extracts the case factors of the corresponding historical cases from the historical case file. Case factors include case clearance, legal relationships, evidence combinations, focus type, and arbitration perspective, among others. The server may determine the case type of the corresponding history case according to the case route. The server groups a number of historical cases according to case type and legal relationship.
And the server performs single-heat coding on the case factors to obtain the feature vector of each historical case. The server builds a plurality of training sets based on the feature vectors of each set of historical case factors. The server trains the basic prediction model based on different training sets to obtain a winner rate prediction model with accuracy reaching a threshold value corresponding to each case type and legal relation. The base prediction model may be an X-GBoost model or the like. In addition, the basic prediction models corresponding to different case types and legal relations can be different, and the method is not limited.
In this embodiment, according to a litigation result prediction request sent by a terminal, case information and evidence combination of a current case can be obtained; the evidence characteristic vector of the current case can be obtained by identifying the evidence type corresponding to each evidence item in the evidence combination; according to the case information, the case type and legal relation of the current case can be determined; based on a preset target prediction model corresponding to the case type and legal relation and the evidence feature vector obtained through calculation, the probability of complaint of the current case can be predicted. Because the user only needs to provide the mastered evidence combination of the current case, the evidence type related to the evidence combination is automatically identified, and the winning probability prediction is automatically carried out based on the preset litigation result prediction model, so that the prediction efficiency can be improved; in addition, different litigation result prediction models are respectively preset aiming at the cases of different case types and legal relations, so that the pertinence of the models can be improved, and the accuracy of the prediction results can be further improved.
In one embodiment, the case factors include evidence combinations; acquiring a plurality of historical cases and case factors corresponding to each historical case, including: acquiring case files of a plurality of historical cases; extracting one or more evidence description sentences from the case file through regular matching; inputting the evidence description sentences into a preset semantic understanding model to obtain one or more evidence items corresponding to each evidence description sentence; and identifying the evidence type corresponding to each evidence item, and generating the evidence combination of the corresponding historical case based on the evidence type.
The extraction mode of different case factors can be different. For the information content directly recorded in the plaintext in the case file, the factor value of the corresponding case factor, such as legal relation, can be obtained by utilizing keyword matching or regular matching. However, for the factor value of the case factor which is not explicitly described in the case file, it is required to refine the factor value based on a pre-trained semantic understanding model.
The semantic understanding model can be obtained based on case file training of a large number of sample cases and is used for extracting factor values of target case factors and recording the factor values as target factor values. Specifically, the server screens the description sentences related to the target case factors in the case files of the historical cases through regular matching, and records the description sentences as factor description sentences. And the server marks the target factor value of the screened factor description statement. Different factor description sentences and corresponding target factor values respectively form different samples. Training the initial model to be trained based on a large number of samples to obtain a semantic understanding model. The initial model to be trained may be an X-GBoost model or the like.
The server presets one or more template description statements for each target case factor. Each template description statement is associated with a corresponding reference factor value. The server marks the reference factor value associated with the template description statement with the highest semantic similarity and reaching the threshold value as the target factor value of the corresponding target case factor based on the semantic similarity between the factor description statement extracted by semantic understanding calculation and the corresponding template description statement.
The target case factor may be evidence combination, focus type, etc. The factor description statement may be an evidence description statement, a focus description statement, or the like, depending on the target case factor. Specifically, if the target case factor is an evidence combination, the server describes the associated reference evidence item according to the template description statement with the highest semantic similarity, and takes the reference evidence item as the evidence item related to the current case. The server identifies the evidence type corresponding to each evidence item of the current case, and generates the evidence combination of the corresponding historical case based on the evidence type. If the target case factor is the focus type, the server describes the focus type associated with the statement according to the template with the highest semantic similarity, and takes the focus type as the focus type of the current case disputed focus.
In the embodiment, the case files of a large number of cases are automatically deconstructed based on the semantic understanding model, required case factors are extracted, and therefore sample processing efficiency can be improved, and model training efficiency is further improved.
In one embodiment, as shown in fig. 3, the method further includes the step of evidence chain guidance, including:
step 302, receiving an evidence chain guide request sent by a terminal; the evidence chain directs that the request carries the case identification.
In step 308, fuzzy matching determines the focus type of the disputed focus.
And step 312, generating evidence labels corresponding to the corresponding evidence items according to the decision weights.
And step 314, generating an evidence chain based on the plurality of evidence items with the evidence labels, and returning the evidence chain to the terminal.
When the user desires to know which evidence needs to be provided for the current case has a higher probability of success, the case identification of the current case can also be filled in based on a litigation service platform on the terminal. The litigation service platform generates an evidence chain guide request based on the filled case identification, and sends the evidence chain guide request to the server. And the server pulls the case information of the corresponding case from the database corresponding to the litigation service platform according to the case identification. The server and litigation service platform directly realize data docking. If the data docking is not realized, the user can submit the case information of the current case to the server through the terminal.
The server presets a case statistics table. The case statistics table records case information of a plurality of history cases. The case information may include a case identification and a plurality of case factors deconstructed from the history case file. The case factors may be extracted as described above. The case statistics may be dynamically updated. The case statistics table may be as shown in table 1 below.
TABLE 1
The server trains an evidence chain guide model in advance based on the case statistical table obtained by deconstructing, is used for counting evidence items under legal relation/under the rule relation and supporting conditions thereof in the history cases of different case, calculates the decision weight of each evidence item, and forms a high-supporting-rate and simplified evidence chain corresponding to each case and the legal relation.
And the server compares whether the decision weight of each evidence item reaches a preset value, if so, marks the evidence item as core evidence, and otherwise marks the evidence item as non-core evidence. The preset value can be freely defined according to the requirement, and is not limited.
In the embodiment, not only the prediction requirement of the user is met, but also the evidence chain guide is carried out on the user, so that the litigation threshold is reduced; in addition, the importance ranking and classification are carried out on the evidence items, so that a user can quickly know the importance degree of each evidence item on the current case.
In one embodiment, the method further comprises the step of analyzing big data of litigation results, and specifically comprises the following steps: receiving a litigation analysis request sent by a terminal; the litigation analysis request carries a retrieval analysis statement; acquiring a case statistical table and corresponding table information; generating a target vector according to the search analysis statement and the table information; inputting the target vector into a preset sequence model to obtain a plurality of analysis intention expressions; inputting the target vector into a preset intention classification model to obtain a target SQL template; filling the analysis intention expression into the target SQL template to obtain an SQL retrieval analysis statement; inquiring related cases in the case statistics table based on the SQL search analysis statement, carrying out statistics analysis on case information of the related cases, and returning an analysis result to the terminal.
The search analysis statement may be one or more phrases formed in natural language. For example, "support rate of financial borrowing cases in Guangdong area", "specific gravity of dispute cases in 2018 Guangdong area contract release", "where cases related to lending disputes are generally distributed", and the like. The search analysis statement may be a statement that has a grammatical error and is semantically non-coherent. For example, "case interpretation trend of overdue repayment of loan in the last five years", such as Guangdong court ", and the like. For the search analysis statement with grammar errors and incoherent semantics, the server performs semantic analysis on the search analysis statement to generate one or more corresponding search intention statements with coherent semantics, generates a search intention confirmation prompt based on the search intention statement, and returns the search intention confirmation prompt to the terminal. The user can select one of the search intention sentences based on the search intention confirmation prompt, and the terminal transmits the selected information to the server. The server performs search analysis based on the search intention sentence selected by the user according to the method provided by the embodiment. In another embodiment, the search analysis statement may be a plurality of search fields, which is not limited thereto.
The server obtains the table information of the case statistics table. The table information comprises a table name, a plurality of table heads and a plurality of field enumeration values corresponding to each table head. Each header may be a case factor. For example, in table 1, each field of the first row is a header, and the fields of the remaining rows of each column are a plurality of field enumeration values corresponding to the corresponding header.
The server calculates the characterization vectors corresponding to the retrieval analysis statement and the table information respectively, and splices the characterization vectors corresponding to the retrieval analysis statement and the characterization vectors corresponding to the table information to obtain the target vector.
The server pre-trains the sequence model based on case information of a large number of real historical cases. The sequence model is used for identifying the retrieval analysis intention of the user, namely mining potential information of analysis dimensions, analysis conditions and the like which can reflect the expectations of the user in the retrieval analysis statement. The analysis intention expression may be in the form of Key-value Key value pairs, for example, the analysis intention expression corresponding to the search analysis statement "support rate of financial borrowing cases in guangdong region" may be "case by=financial borrowing disputes", "region=guangdong", "litigation result=support", "intention=victory proportion". For another example, the corresponding analysis intent expressions of the search analysis statement "where cases related to lending disputes are generally distributed" may be "case by= lending disputes OR × borrowing disputes", "intent=region", respectively.
The server presets a variety of SQL templates. Different SQL templates are used to satisfy the user's intent to analyze based on different dimensions and conditions. The server training trains the intent classification model. The intention classification model is used for determining which SQL template to choose according to the current user retrieval analysis intention. The intention classification model can be obtained by performing supervised training on the basic classification model based on a large number of simulated search analysis sentences and target SQL templates correspondingly marked by each search analysis sentence. The basic training model may be an RNN model (Recurrent neural network, recurrent neural network model). And the server sequentially inputs a plurality of target vectors corresponding to the retrieval analysis sentences into the intention classification model to obtain a target SQL template.
The manner in which different SQL templates are populated may be different. And the server fills the analysis intention expression into the target SQL template according to the filling mode of the target SQL template, so that the SQL retrieval analysis statement can be obtained.
Based on different SQL search analysis sentences, the data inquiry of different dimensionalities such as judge time, region, case, and the like can be realized; and the data statistics of different conditions such as case specific gravity, case number, support rate and the like can be realized. And the server performs data query and statistical analysis in the case statistics table based on the SQL search analysis statement, and returns an analysis result to the terminal.
For the information query processing of related cases in the past, the traditional mode directly judges the search intention of the user through the mode of presetting word lists, and does not support the user to search in a natural language mode. In addition, the preset vocabulary not only requires a lot of labor, but also has difficulty in ensuring the coverage rate of the vocabulary information, and once a certain search keyword input by a user is not covered in the vocabulary, search analysis fails. And the application supports the retrieval of the user in a natural language manner. It is easy to understand that natural language can express the search intention of the user more accurately than the single search keyword, so that the search analysis intention of the user can be mined more accurately based on the search analysis statement. The method and the device can further quickly and accurately identify the search analysis intention of the user through the machine learning pre-training sequence model and the intention classification model, and compared with a preset word list, the method and the device can reduce manual participation and realize end-to-end judgment information search analysis in a true sense. By combining a case statistical table which is pre-deconstructed, the retrieval and analysis efficiency of the judging information can be improved, and different retrieval and analysis intentions of a user can be responded quickly.
In this embodiment, in addition to simply predicting the complaint probability, the method also supports the user to search and analyze the processing situation of the related case based on the natural language, so that the complaint situation of the current case can be known in advance from multiple dimensions.
In one embodiment, inputting the evidence feature vector into the target prediction model includes: extracting the dispute focus of the current case from the case information; fuzzy matching is carried out to determine the focus type of the disputed focus; generating a focus characteristic vector of the current case based on the focus type; splicing the focus feature vector and the evidence feature vector to obtain a case feature vector; the case feature vector is input into the target prediction model.
In order to provide accuracy of the prediction result, the embodiment combines the dispute focus of the current case to perform specific case specific analysis. Specifically, the server determines the focus type of the current case disputed focus according to the mode, and calculates a focus characteristic vector corresponding to the focus type. And the server splices the data type characteristic vector and the focus type characteristic vector. The evidence feature vector and the focus feature vector can be respectively converted, the converted evidence feature vector and the converted focus feature vector are spliced, and the spliced vector is used as a case feature vector of the current case. The server respectively presets corresponding litigation result prediction models aiming at cases of different case types, legal relations and focus types. In other words, the server counts the legal relations of subdivisions in the historical cases of different cases, the evidence items under the disputed focus and the supporting conditions thereof in advance, calculates the decision weight of each evidence item, and obtains the reference statistical result. The server predicts the complaint probability of the current case based on the reference statistics result and the case feature vector of the current case.
In another embodiment, the target prediction model includes a first prediction model and a second prediction model; the method comprises the steps of obtaining a target prediction model corresponding to a case type and legal relation, inputting an evidence feature vector into the target prediction model to obtain the complaint probability of the current case, and comprising the following steps: acquiring a corresponding first prediction model according to the case type and legal relation, and inputting the evidence feature vector into the first prediction model to obtain a first prediction value; extracting a dispute focus of the current case from the case information, determining a focus type of the dispute focus, and acquiring a corresponding second prediction model according to the focus type; generating a focus characteristic vector of the current case based on the focus type, and inputting the focus characteristic vector into a second prediction model to obtain a second prediction value; determining the prediction weights respectively corresponding to the first prediction model and the second prediction model according to the case types and legal relations; and carrying out preset logic operation on the first predicted value and the second predicted value according to the predicted weight to obtain the probability of the complaint of the current case.
The server combines the combination of credentials with the predictors of the dispute focus in different ways. Specifically, the evidence feature vector and the focus feature vector are not required to be spliced, but corresponding litigation result prediction models are preset for different prediction factors respectively, such as corresponding first prediction models are preset for different case types and cases of legal relation respectively; and respectively presetting corresponding second prediction models for cases with different focus types. The server inputs the evidence feature vector into a first prediction model to obtain a first predicted value, and inputs the focus feature vector into a second prediction model to obtain a second predicted value.
The server inputs the case type, legal relation and focus type of the current case into a preset classifier to obtain decision weights corresponding to the evidence combination and the focus type prediction factors respectively. The classifier may also be an X-GBoost model. The server performs preset logic operation on the first predicted value and the second predicted value, and takes an operation result as the complaint probability of the current case. The preset logic operation may be a superposition operation based on decision weights, etc. It is easy to understand that the first predicted value or the second predicted value may be used as the probability of complaint of the current case according to the decision weight.
In the embodiment, in addition to considering the case type, legal relation and evidence combination, the case characteristics of the current case are fully considered in combination with the dispute focus of the current case, so that specific analysis of specific cases is realized, and the accuracy of the prediction result can be provided.
It should be understood that, although the steps in the flowcharts of fig. 2 and 3 are shown in order as indicated by the arrows, these steps are not necessarily performed in order as indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in fig. 2 and 3 may include multiple sub-steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, which are not necessarily performed sequentially, but may be performed alternately or alternately with other steps or at least a portion of the sub-steps or stages of other steps.
In one embodiment, as shown in fig. 4, there is provided a litigation result prediction apparatus, comprising: a case forecast request module 402, a case feature extraction module 404, and a litigation result forecast module 406, wherein:
a case prediction request module 402, configured to receive a litigation result prediction request sent by a terminal; the litigation result prediction request carries case information and evidence combination of the current case.
A case feature extraction module 404, configured to determine a evidence type corresponding to each evidence item in the evidence combination; an evidence feature vector for the current case is generated based on the evidence type.
A litigation result prediction module 406, configured to determine a case type and legal relation of the current case according to the case information; and acquiring a target prediction model corresponding to the case type and legal relation, and inputting the evidence feature vector into the target prediction model to obtain the complaint probability of the current case.
In one embodiment, the apparatus further includes a evidence chain guiding module 408, configured to receive an evidence chain guiding request sent by the terminal; the evidence chain guiding request carries a case identification; acquiring case information of a current case according to the case identification; extracting the dispute focus of the current case from the case information; fuzzy matching is carried out to determine the focus type of the disputed focus; acquiring one or more associated evidence items and decision weights corresponding to each evidence item according to the case type, legal relation and focus type of the current case; generating an evidence label corresponding to the corresponding evidence item according to the decision weight; and generating an evidence chain based on a plurality of evidence items with evidence labels, and returning the evidence chain to the terminal.
In one embodiment, the apparatus includes a prediction model construction module 410 configured to obtain a plurality of historical cases and a case factor corresponding to each historical case; the case factors include case type and legal relationship; grouping a plurality of historical cases according to case types and legal relations; coding the case factors to obtain feature vectors of corresponding historical cases; constructing a training set corresponding to the corresponding case type and legal relation based on the feature vector of each group of historical cases; and training the basic prediction model based on the characteristic directions of the historical cases in different training sets to obtain target prediction models corresponding to different case types and legal relations.
In one embodiment, the case factors include evidence combinations; the prediction model construction module 410 is further configured to obtain case files of a plurality of historical cases; extracting one or more evidence description sentences from the case file through regular matching; inputting the evidence description sentences into a preset semantic understanding model to obtain one or more evidence items corresponding to each evidence description sentence; and identifying the evidence type corresponding to each evidence item, and generating the evidence combination of the corresponding historical case based on the evidence type.
In one embodiment, litigation result prediction module 406 is further configured to extract a dispute focus for the current case in the case information; fuzzy matching is carried out to determine the focus type of the disputed focus; generating a focus characteristic vector of the current case based on the focus type; splicing the focus feature vector and the evidence feature vector to obtain a case feature vector; the case feature vector is input into the target prediction model.
In one embodiment, the target prediction model includes a first prediction model and a second prediction model; the litigation result prediction module 406 is further configured to obtain a corresponding first prediction model according to the case type and the legal relationship, and input the evidence feature vector into the first prediction model to obtain a first prediction value; extracting a dispute focus of the current case from the case information, determining a focus type of the dispute focus, and acquiring a corresponding second prediction model according to the focus type; generating a focus characteristic vector of the current case based on the focus type, and inputting the focus characteristic vector into a second prediction model to obtain a second prediction value; determining the prediction weights respectively corresponding to the first prediction model and the second prediction model according to the case types and legal relations; and carrying out preset logic operation on the first predicted value and the second predicted value according to the predicted weight to obtain the probability of the complaint of the current case.
For specific limitations on litigation outcome prediction means, reference is made to the limitations on litigation outcome prediction methods hereinabove, and no further description is given here. The respective modules in the litigation result prediction apparatus described above may be implemented in whole or in part by software, hardware, or a combination thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.
In one embodiment, a computer device is provided, which may be a server, the internal structure of which may be as shown in fig. 5. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the computer device is used for storing the case information of the current case and the historical case. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program when executed by a processor implements a litigation outcome prediction method.
It will be appreciated by those skilled in the art that the structure shown in fig. 5 is merely a block diagram of some of the structures associated with the present application and is not limiting of the computer device to which the present application may be applied, and that a particular computer device may include more or fewer components than shown, or may combine certain components, or have a different arrangement of components.
A computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of the litigation result prediction method provided in any one of the embodiments of the present application.
Those skilled in the art will appreciate that implementing all or part of the above-described methods in accordance with the embodiments may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the various embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), memory bus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.
The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The foregoing examples represent only a few embodiments of the present application, which are described in more detail and are not to be construed as limiting the scope of the invention. It should be noted that it would be apparent to those skilled in the art that various modifications and improvements could be made without departing from the spirit of the present application, which would be within the scope of the present application. Accordingly, the scope of protection of the present application is to be determined by the claims appended hereto.
Claims (10)
1. A litigation outcome prediction method, the method comprising:
receiving a litigation result prediction request sent by a terminal; the litigation result prediction request carries case information and evidence combination of the current case;
determining a evidence type corresponding to each evidence item in the evidence combination;
generating an evidence feature vector of the current case based on the evidence type;
Determining the case type and legal relation of the current case according to the case information;
acquiring a corresponding first prediction model according to the case type and legal relation, and inputting the evidence feature vector into the first prediction model to obtain a first prediction value;
extracting a dispute focus of the current case from the case information, determining a focus type of the dispute focus through fuzzy matching, and acquiring a corresponding second prediction model according to the focus type;
generating a focus characteristic vector of the current case based on the focus type, and inputting the focus characteristic vector into the second prediction model to obtain a second prediction value;
determining the prediction weights respectively corresponding to the first prediction model and the second prediction model according to the case types and the legal relation;
performing preset logic operation on the first predicted value and the second predicted value according to the predicted weight to obtain the complaint probability of the current case;
the method further comprises the steps of:
receiving an evidence chain guide request sent by a terminal; the evidence chain guide request carries a case identification;
acquiring the case information of the current case according to the case identifier;
extracting the dispute focus of the current case from the case information;
Determining a focus type of the dispute focus by fuzzy matching;
acquiring one or more associated evidence items and decision weights corresponding to each evidence item according to the case type, legal relation and focus type of the current case;
generating an evidence label corresponding to the corresponding evidence item according to the decision weight;
an evidence chain is generated based on a plurality of evidence items with evidence labels, and the evidence chain is returned to the terminal.
2. The method of claim 1, wherein prior to obtaining the target prediction model corresponding to the case type and target legal relationship, further comprising:
acquiring a plurality of historical cases and case factors corresponding to each historical case; the case factors comprise case types and legal relations;
grouping a plurality of historical cases according to the case types and legal relations;
coding the case factors to obtain feature vectors of corresponding historical cases;
constructing a training set corresponding to the corresponding case type and legal relation based on the feature vector of each group of historical cases;
and training the basic prediction model based on different training sets to obtain target prediction models corresponding to different case types and legal relations.
3. The method of claim 2, wherein the case factor comprises a combination of evidence; the obtaining a plurality of historical cases and case factors corresponding to each historical case includes:
acquiring case files of a plurality of historical cases;
extracting one or more evidence description sentences from the case file through regular matching;
inputting the evidence description sentences into a preset semantic understanding model to obtain one or more evidence items corresponding to each evidence description sentence;
and identifying the evidence type corresponding to each evidence item, and generating the evidence combination of the corresponding historical case based on the evidence type.
4. The method of claim 1, wherein the case types include financial borrowing disputes, contract disputes, inheritance rights disputes; the legal relationships include debit relationships, guaranty relationships, and common repayment relationships.
5. The method of claim 1, wherein the evidence composition comprises a plurality of evidence items including principal statements, certificates, physical evidence, audiovisual material, witness letters, electronic data, authentication conclusions, and survey notes.
6. A litigation outcome prediction device for performing the method of claim 1, the device comprising:
the case prediction request module is used for receiving a litigation result prediction request sent by the terminal; the litigation result prediction request carries case information and evidence combination of the current case;
the case feature extraction module is used for determining the evidence type corresponding to each evidence item in the evidence combination; generating an evidence feature vector of the current case based on the evidence type;
the litigation result prediction module is used for determining the case type and legal relation of the current case according to the case information; acquiring a corresponding first prediction model according to the case type and legal relation, and inputting the evidence feature vector into the first prediction model to obtain a first prediction value; extracting a dispute focus of the current case from the case information, determining a focus type of the dispute focus through fuzzy matching, and acquiring a corresponding second prediction model according to the focus type; generating a focus characteristic vector of the current case based on the focus type, and inputting the focus characteristic vector into the second prediction model to obtain a second prediction value; determining the prediction weights respectively corresponding to the first prediction model and the second prediction model according to the case types and the legal relation; performing preset logic operation on the first predicted value and the second predicted value according to the predicted weight to obtain the complaint probability of the current case;
The device also comprises an evidence chain guide module, which is used for receiving an evidence chain guide request sent by the terminal; the evidence chain guide request carries a case identification; acquiring the case information of the current case according to the case identifier; extracting the dispute focus of the current case from the case information; determining a focus type of the dispute focus by fuzzy matching; acquiring one or more associated evidence items and decision weights corresponding to each evidence item according to the case type, legal relation and focus type of the current case; generating an evidence label corresponding to the corresponding evidence item according to the decision weight; an evidence chain is generated based on a plurality of evidence items with evidence labels, and the evidence chain is returned to the terminal.
7. The apparatus of claim 6, further comprising a predictive model construction module configured to obtain a plurality of historical cases and a case factor corresponding to each of the historical cases; the case factors comprise case types and legal relations; grouping a plurality of historical cases according to the case types and legal relations; coding the case factors to obtain feature vectors of corresponding historical cases; constructing a training set corresponding to the corresponding case type and legal relation based on the feature vector of each group of historical cases; and training the basic prediction model based on different training sets to obtain target prediction models corresponding to different case types and legal relations.
8. The apparatus of claim 7, wherein the case factor comprises a combination of evidence; the prediction model construction module is also used for acquiring case files of a plurality of historical cases; extracting one or more evidence description sentences from the case file through regular matching; inputting the evidence description sentences into a preset semantic understanding model to obtain one or more evidence items corresponding to each evidence description sentence; and identifying the evidence type corresponding to each evidence item, and generating the evidence combination of the corresponding historical case based on the evidence type.
9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any one of claims 1 to 5 when the computer program is executed.
10. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910520319.XA CN110377632B (en) | 2019-06-17 | 2019-06-17 | Litigation result prediction method, litigation result prediction device, litigation result prediction computer device and litigation result prediction storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910520319.XA CN110377632B (en) | 2019-06-17 | 2019-06-17 | Litigation result prediction method, litigation result prediction device, litigation result prediction computer device and litigation result prediction storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110377632A CN110377632A (en) | 2019-10-25 |
| CN110377632B true CN110377632B (en) | 2023-06-20 |
Family
ID=68248914
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910520319.XA Active CN110377632B (en) | 2019-06-17 | 2019-06-17 | Litigation result prediction method, litigation result prediction device, litigation result prediction computer device and litigation result prediction storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110377632B (en) |
Families Citing this family (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112825174A (en) * | 2019-11-15 | 2021-05-21 | 阿里云计算有限公司 | Litigation prediction method, device, system and computer storage medium |
| CN112818671A (en) * | 2019-11-15 | 2021-05-18 | 阿里巴巴集团控股有限公司 | Text information processing method and device, storage medium and processor |
| CN111191871A (en) * | 2019-11-21 | 2020-05-22 | 深圳壹账通智能科技有限公司 | Project baseline data generation method and device, computer equipment and storage medium |
| CN111047092A (en) * | 2019-12-11 | 2020-04-21 | 深圳前海环融联易信息科技服务有限公司 | Dispute case victory rate prediction method and device, computer equipment and storage medium |
| CN111160801A (en) * | 2019-12-31 | 2020-05-15 | 重庆百事得大牛机器人有限公司 | Electronic evidence risk judgment method |
| CN111339379A (en) * | 2020-02-29 | 2020-06-26 | 重庆百事得大牛机器人有限公司 | Electronic evidence analysis system |
| CN111353079B (en) * | 2020-02-29 | 2023-05-05 | 重庆百事得大牛机器人有限公司 | Electronic evidence analysis suggestion system and method |
| CN112836891B (en) * | 2021-02-23 | 2023-05-16 | 支付宝(杭州)信息技术有限公司 | Case processing method and device |
| JP7047231B1 (en) * | 2021-06-25 | 2022-04-05 | 株式会社Robot Consulting | Information processing systems, computer systems and programs |
| CN113792127B (en) * | 2021-09-15 | 2023-12-26 | 平安国际智慧城市科技股份有限公司 | Rule recognition method and device based on big data, electronic equipment and medium |
| CN115796285B (en) * | 2023-02-13 | 2023-05-09 | 上海百事通法务信息技术有限公司浙江分公司 | Litigation case pre-judging method and device based on engineering model and electronic equipment |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109272156A (en) * | 2018-09-12 | 2019-01-25 | 河海大学 | A kind of super short-period wind power probability forecasting method |
| CN109359175A (en) * | 2018-09-07 | 2019-02-19 | 平安科技（深圳）有限公司 | Electronic device, the method for lawsuit data processing and storage medium |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2014036027A1 (en) * | 2012-08-27 | 2014-03-06 | Datacert, Inc. | Predictive modeling: litigation decision analysis and optimization |
| US9336533B2 (en) * | 2013-03-13 | 2016-05-10 | Salesforce.Com, Inc. | Systems, methods, and apparatuses for implementing a similar command with a predictive query interface |
| US20140279583A1 (en) * | 2013-03-14 | 2014-09-18 | Lex Machina, Inc. | Systems and Methods for Classifying Entities |
| US20190130476A1 (en) * | 2017-04-25 | 2019-05-02 | Yada Zhu | Management System and Predictive Modeling Method for Optimal Decision of Cargo Bidding Price |
| CN109410096B (en) * | 2018-10-18 | 2024-02-06 | 上海右云信息技术有限公司 | Method and equipment for providing litigation strategy information of infringing cases |
-
2019
- 2019-06-17 CN CN201910520319.XA patent/CN110377632B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109359175A (en) * | 2018-09-07 | 2019-02-19 | 平安科技（深圳）有限公司 | Electronic device, the method for lawsuit data processing and storage medium |
| CN109272156A (en) * | 2018-09-12 | 2019-01-25 | 河海大学 | A kind of super short-period wind power probability forecasting method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110377632A (en) | 2019-10-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110377632B (en) | Litigation result prediction method, litigation result prediction device, litigation result prediction computer device and litigation result prediction storage medium | |
| CN110021439B (en) | Medical data classification method and device based on machine learning and computer equipment | |
| CN111160017B (en) | Keyword extraction method, phonetics scoring method and phonetics recommendation method | |
| CN110209764B (en) | Corpus annotation set generation method and device, electronic equipment and storage medium | |
| WO2021169111A1 (en) | Resume screening method and apparatus, computer device and storage medium | |
| US11907274B2 (en) | Hyper-graph learner for natural language comprehension | |
| CN109829629B (en) | Risk analysis report generation method, apparatus, computer device and storage medium | |
| CN109543516A (en) | Signing intention judgment method, device, computer equipment and storage medium | |
| US11164564B2 (en) | Augmented intent and entity extraction using pattern recognition interstitial regular expressions | |
| CN110362798B (en) | Method, apparatus, computer device and storage medium for judging information retrieval analysis | |
| CN108491406B (en) | Information classification method and device, computer equipment and storage medium | |
| CN111309881A (en) | Method and device for processing unknown questions in intelligent question answering, computer equipment and medium | |
| CN110377631A (en) | Case information processing method, device, computer equipment and storage medium | |
| CN111190946A (en) | Report generation method and device, computer equipment and storage medium | |
| CN110532229B (en) | Evidence file retrieval method, device, computer equipment and storage medium | |
| CN112288279A (en) | Business risk assessment method and device based on natural language processing and linear regression | |
| CN110309279A (en) | Based on language model, method, apparatus and computer equipment are practiced in speech therapy | |
| Beltrán et al. | ClaimHunter: An Unattended Tool for Automated Claim Detection on Twitter. | |
| CN110377618B (en) | Method, device, computer equipment and storage medium for analyzing decision result | |
| CN110362592B (en) | Method, device, computer equipment and storage medium for pushing arbitration guide information | |
| Zhang et al. | Enabling rapid large-scale seismic bridge vulnerability assessment through artificial intelligence | |
| US20220366344A1 (en) | Determining section conformity and providing recommendations | |
| US20170293863A1 (en) | Data analysis system, and control method, program, and recording medium therefor | |
| CN112905763A (en) | Session system development method, device, computer equipment and storage medium | |
| AU2019290658B2 (en) | Systems and methods for identifying and linking events in structured proceedings |
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 |