CN110895568B

CN110895568B - Method and system for processing court trial records

Info

Publication number: CN110895568B
Application number: CN201811069881.7A
Authority: CN
Inventors: 周鑫; 张雅婷; 李泉志; 孙常龙; 刘晓钟; 司罗
Original assignee: Alibaba Group Holding Ltd
Current assignee: Alibaba Group Holding Ltd
Priority date: 2018-09-13
Filing date: 2018-09-13
Publication date: 2023-07-21
Anticipated expiration: 2038-09-13
Also published as: CN110895568A

Abstract

The invention discloses a method and a system for processing court trial records. Wherein the method comprises the following steps: acquiring a court trial record recorded in a court trial process, wherein the court trial record comprises at least one theme module, and the theme module at least records legal information generated in different court trial stages in the court trial process; determining an information extraction model corresponding to each topic module based on the identification information of different topic modules and corresponding preset parameters; and extracting information elements of each topic module from the court trial record based on the information extraction model, wherein the information elements are used for initializing legal knowledge maps. The invention solves the technical problem that in the prior art, the abstract of the strokes of the Internet court is difficult to acquire.

Description

Method and system for processing court trial records

Technical Field

The invention relates to the legal field, in particular to a method and a system for processing court trial records.

Background

In the trial flow of a whole flow, the method is divided into three stages of court front, court trial and court back. The court needs to submit prosecution and evidence through original notice, examine the case, and be submitted with dialectical evidence; during the court trial, the two parties of the court organization debate about the dispute focus, submit new evidence and even modify complaints; the court officer can complete the judgment through the conclusions. In the existing Internet court mode, the whole process of judgment is carried out on line, and all relevant files such as complaints, evidences and the like are input electronically and structurally so that an algorithm can extract information and judge reasoning.

In the Internet court mode, the method has the defects of single case type, simple court trial process and low information in court trial records. In the process of popularizing the mode of the Internet court to the traditional trial, related files before the court trial and related files after the court trial are important contents in intelligent trial, and court trial records are very complex and important files generated in the court trial process and are also very important links in intelligent trial.

The related files before and after the court are easy to process electronically and structurally, but the court trial records are difficult to accurately arrange, wherein one important reason is that in the court trial process of the off-line court, a bookend can make abstract records according to the court trial process, an on-line Internet court is archived by using a court trial process audio-video, the court trial pen-record abstract of the bookend is not available, a large amount of irrelevant information is generated after voice in the video is converted into characters, and therefore effective and accurate content in the court trial records is difficult to provide for an intelligent trial algorithm.

Aiming at the problem that in the prior art, the abstract of the strokes of the Internet court is difficult to acquire, no effective solution is proposed at present.

Disclosure of Invention

The embodiment of the invention provides a method and a system for processing court trial records, which at least solve the technical problem that in the prior art, the abstract of the stroke records of an online Internet court is difficult to acquire.

According to one aspect of the embodiment of the invention, there is provided a method for processing a court trial record, including: acquiring a court trial record recorded in a court trial process, wherein the court trial record comprises at least one theme module, and the theme module at least records legal information generated in different court trial stages in the court trial process; determining an information extraction model corresponding to each topic module based on the identification information of different topic modules and corresponding preset parameters; and extracting information elements of each topic module from the court trial record based on the information extraction model, wherein the information elements are used for initializing legal knowledge maps.

According to another aspect of the embodiment of the present invention, there is also provided a method for obtaining a referee result of a case, including: displaying a court trial record recorded in a court trial process, wherein the court trial record comprises at least one theme module, and the theme module at least records legal information generated in different court trial stages in the court trial process; and outputting an information extraction model corresponding to each topic module, and extracting information elements of each topic module from the court trial record, wherein the information elements are used for initializing legal knowledge maps, and the information extraction model is obtained based on identification information of different topic modules and corresponding preset parameters.

According to another aspect of the embodiment of the present invention, there is also provided an early warning system for referee results, including: a processor; and a memory, coupled to the processor, for providing instructions to the processor to process the steps of: acquiring a court trial record recorded in a court trial process, wherein the court trial record comprises at least one theme module, and the theme module at least records legal information generated in different court trial stages in the court trial process; determining an information extraction model corresponding to each topic module based on the identification information of different topic modules and corresponding preset parameters; and extracting information elements of each topic module from the court trial record based on the information extraction model, wherein the information elements are used for initializing legal knowledge maps.

In the embodiment of the invention, a court trial record recorded in a court trial process is obtained, wherein the court trial record comprises at least one theme module, and the theme module at least records legal information generated in different court trial stages in the court trial process; determining an information extraction model corresponding to each topic module based on the identification information of different topic modules and corresponding preset parameters; and extracting information elements of each topic module from the court trial record based on the information extraction model, wherein the information elements are used for initializing legal knowledge maps. Through the scheme, the online Internet court does not need a bookman to make a summary record of the court trial process, and the information elements can be automatically extracted from the court trial record according to the court trial record of the court trial process, so that the technical problem that the abstract of the pen record of the online Internet court is difficult to obtain in the prior art is solved, and the Internet court can be popularized to a traditional trial mode.

Therefore, the technical problem that in the prior art, the abstract of the pen records of the Internet court is difficult to acquire is solved.

Drawings

The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiments of the invention and together with the description serve to explain the invention and do not constitute a limitation on the invention. In the drawings:

FIG. 1 shows a block diagram of the hardware architecture of a computer terminal (or mobile device) for implementing a method of acquiring referee results for a case;

FIG. 2 is a flow chart of a method of processing a court trial record according to embodiment 1 of the present application;

FIG. 3 is a schematic illustration of a legal knowledge graph according to example 1 of the present application;

FIG. 4 is a schematic illustration of an alternative legal indicator map according to example 1 of the present application;

FIG. 5 is a schematic diagram of an interactive interface presented to a judge by an algorithm according to embodiment 1 of the present application;

FIG. 6 is a schematic diagram of a forensic feedback risk point according to example 1 of the present application;

FIG. 7 is a flow chart of a legal case referee method according to embodiment 1 of the present application;

FIG. 8 is a schematic diagram of an apparatus for processing a court trial recording according to embodiment 2 of the present application;

FIG. 9 is a flowchart of a method of acquiring referee results for a case according to embodiment 3 of the present application;

FIG. 10 is a schematic diagram of an apparatus for processing a court trial recording according to embodiment 4 of the present application; and

fig. 11 is a block diagram of a computer terminal according to an embodiment of the present invention.

Detailed Description

In order that those skilled in the art will better understand the present invention, a technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, shall fall within the scope of the present invention.

It should be noted that the terms "first," "second," and the like in the description and the claims of the present invention and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the invention described herein may be implemented in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

Example 1

In accordance with an embodiment of the present invention, there is also provided an embodiment of a method of obtaining referee results for a case, it being noted that the steps shown in the flowchart of the figures may be performed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be performed in an order other than that shown.

The method embodiment provided in the first embodiment of the present application may be executed in a mobile terminal, a computer terminal or a similar computing device. Fig. 1 shows a block diagram of a hardware configuration of a computer terminal (or mobile device) for implementing a method of acquiring referee results of cases. As shown in fig. 1, the computer terminal 10 (or mobile device 10) may include one or more (shown as 102a, 102b, … …,102 n) processors 102 (the processors 102 may include, but are not limited to, a microprocessor MCU, a programmable logic device FPGA, etc. processing means), a memory 104 for storing data, and a transmission means 106 for communication functions. In addition, the method may further include: a display, an input/output interface (I/O interface), a Universal Serial Bus (USB) port (which may be included as one of the ports of the I/O interface), a network interface, a power supply, and/or a camera. It will be appreciated by those of ordinary skill in the art that the configuration shown in fig. 1 is merely illustrative and is not intended to limit the configuration of the electronic device described above. For example, the computer terminal 10 may also include more or fewer components than shown in FIG. 1, or have a different configuration than shown in FIG. 1.

It should be noted that the one or more processors 102 and/or other data processing circuits described above may be referred to generally herein as "data processing circuits. The data processing circuit may be embodied in whole or in part in software, hardware, firmware, or any other combination. Furthermore, the data processing circuitry may be a single stand-alone processing module, or incorporated, in whole or in part, into any of the other elements in the computer terminal 10 (or mobile device). As referred to in the embodiments of the present application, the data processing circuit acts as a processor control (e.g., selection of the path of the variable resistor termination to interface).

The memory 104 may be used to store software programs and modules of application software, such as a program instruction/data storage device corresponding to the () method in the embodiment of the present invention, and the processor 102 executes the software programs and modules stored in the memory 104, thereby executing various functional applications and data processing, that is, implementing the above-mentioned vulnerability detection method of application program. Memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include memory located remotely from the processor 102, which may be connected to the computer terminal 10 via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.

The transmission means 106 is arranged to receive or transmit data via a network. The specific examples of the network described above may include a wireless network provided by a communication provider of the computer terminal 10. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, NIC) that can connect to other network devices through a base station to communicate with the internet. In one example, the transmission device 106 may be a Radio Frequency (RF) module for communicating with the internet wirelessly.

The display may be, for example, a touch screen type Liquid Crystal Display (LCD) that may enable a user to interact with a user interface of the computer terminal 10 (or mobile device).

It should be noted here that, in some alternative embodiments, the computer device (or mobile device) shown in fig. 1 described above may include hardware elements (including circuitry), software elements (including computer code stored on a computer-readable medium), or a combination of both hardware and software elements. It should be noted that fig. 1 is only one example of a specific example, and is intended to illustrate the types of components that may be present in the computer device (or mobile device) described above.

In the above-described operating environment, the present application provides a flowchart of a method of processing a court trial record as shown in fig. 2. Fig. 2 is a flowchart of a method of processing a court trial record according to embodiment 1 of the present invention.

Step S21, obtaining a court trial record recorded in the court trial process, wherein the court trial record comprises at least one theme module, and the theme module at least records legal information generated in different court trial stages in the court trial process.

Specifically, the court trial record refers to the text record of all trial interaction real conditions in the court trial process.

In an alternative embodiment, taking an online internet court as an example, the internet court needs a court officer to reserve the court by using a system and input information of each party's litigation participant, when a person logs in, the system can apply a face recognition technology to analyze and compare a face picture of a remote party with a face image database of a public security part, so that identity confirmation of the party is performed to ensure court trial fairness. In the court trial process, the court trial process is archived in a manner of audio recording or video recording.

After the audio or video of the court trial process is obtained, the audio or video is passed through ASR (Automatic Speech Recognition, speech to text technology), the audio information in the audio or video is converted into corresponding text (raw text, i.e. corpus not processed by a computer), and the text is used as the court trial record.

The court trial process may generally be divided into the following sections: checking the identity condition of a principal, declaring forensic notes, claiming original notes, being told to acknowledge or deny original notes, disputing focus by forensic organizations, proving quality certificates by focus by forensic organizations, questioning the fact parts of cases by two parties, questioning the original notes by forensic, forensic dialogs, mediating and declaring by forensic organizations. In an alternative embodiment, the court trial record includes the 10 topic modules described above.

Step S23, determining an information extraction model corresponding to each topic module based on the identification information of different topic modules and corresponding preset parameters.

The identification information and the topic module have a preset corresponding relation, and the court trial stage to which the topic module belongs is determined by identifying the corresponding identification information for the topic module.

The information extraction model is used for extracting useful legal information from the court trial records, so that the abstract of the court trial records is formed, and the intelligent trial system is convenient for trial on legal cases.

The preset parameters are used for indicating the types of the contents which need to be extracted from the theme module, extracting the contents corresponding to the preset parameters from the theme module, inputting the contents into the legal knowledge graph, and performing intelligent judgment on legal cases.

Since each topic module belongs to different stages in the court trial process, legal information included in each topic module may be different, and thus preset parameters of each topic module are different. For example: in the "forensic organization disputed focus" phase, the preset parameters, namely the information to be extracted, include: disputed focus information; in the "proof of quality" phase, its preset parameters, i.e. the information to be extracted, include: the original notice, the evidence submitted by the notice and the three-nature identification result among the parties, if the newly submitted evidence exists, the newly submitted evidence also needs to be used as a supplement for the content of the vestibular forepart, such as prosecution, answer and debate; in the forensic phase, the preset parameters, namely the information to be extracted, include: the core questions and answers of each person and the conclusions of the judges.

In an alternative embodiment, the preset parameters come from legal knowledge graph, and different information extraction methods are selected according to the identification of the topic module and different preset parameters. The information extraction model may be obtained through training, for example, training data for training the information extraction module may be obtained, where the training data includes: the method comprises the steps of training a preset neural network model by using training data to obtain an information extraction model by using a theme module and legal information required to be extracted from the theme module.

In the court trial process, topic modules in different stages are provided with information extraction models which are trained in advance, after a court trial record comprising a plurality of topic modules is obtained, the content of each topic module in which stage in the court trial process belongs is determined according to the identification of each topic module, and the corresponding information extraction model is determined according to the stage to which the topic module belongs and the preset parameters corresponding to the topic module.

And S25, extracting information elements of each topic module from the court trial record based on the information extraction model, wherein the information elements are used for initializing legal knowledge maps.

Specifically, the legal knowledge graph is used for intelligently judging legal cases.

FIG. 3 is a schematic diagram of a legal knowledge graph according to embodiment 1 of the present application, and in combination with FIG. 3, "number of litigation of original notice in Chinese referee' and" number of litigation of original notice in Internet court "are elements in the legal knowledge graph; five nodes, namely whether to report self-confirming complaints, whether to be reported and evidence the original self-confirming complaints, whether to be above 3 times of Chinese judge document net litigation, whether to be above 3 times of Internet court litigation and whether to report the original complaints, are the distinguishing key points. In the legal knowledge graph, each layer of nodes is connected with at least one node of the next layer of nodes through a directed line, so that the trend of the legal knowledge graph is represented.

It should be noted that, the composition of the legal knowledge graph mainly includes:

(1) An entity. In the present application, the entities of the legal knowledge graph mainly include objective facts and legal elements. As shown in fig. 3, the objective fact element is "number of litigation of original notice in the national referee's net" and "number of litigation of original notice in the internet court", and the legal element is "whether to notice abused complaints".

(2) Relationship. In this application, the relationship of legal knowledge graph may refer to attribute relationship and logic relationship. For example, in FIG. 3, the relationship between various legal elements is an OR relationship.

(3) And (5) triad. The application mainly comprises the following three types: an "objective fact element-attribute relationship-objective fact element" (e.g., "original notice-attribute relationship-identification number"), "objective fact element-logical relationship-legal element" (e.g., "commodity description-logical relationship-whether a medical effect is advertised"), "legal element-logical relationship-legal element" (e.g., "whether an original price is fictitious-logical relationship-whether a discount price is wrong").

The initial legal indication map refers to that information elements are used as input values of legal knowledge maps, and assignment is carried out on corresponding nodes in the legal knowledge maps.

In the scheme, based on the information extraction model, the information elements in each topic module are extracted from the court trial record and serve as the content corresponding to the element nodes in the legal knowledge graph, so that the legal knowledge graph can intelligently judge legal cases according to the input required information elements.

The method comprises the steps that court trial records recorded in the court trial process are obtained, wherein the court trial records comprise at least one theme module, and the theme module at least records legal information generated in different court trial stages in the court trial process; determining an information extraction model corresponding to each topic module based on the identification information of different topic modules and corresponding preset parameters; and extracting information elements of each topic module from the court trial record based on the information extraction model, wherein the information elements are used for initializing legal knowledge maps.

Through the scheme, the online Internet court does not need a bookman to make a summary record of the court trial process, and the information elements can be automatically extracted from the court trial record according to the court trial record of the court trial process, so that the technical problem that the abstract of the pen record of the online Internet court is difficult to obtain in the prior art is solved, and the Internet court can be popularized to a traditional trial mode.

As an alternative embodiment, the information extraction model is used to characterize the information extraction method employed for legal information generated for different trial court phases.

Specifically, the information extraction method represented by the information extraction module may include any one or more of the following: classification, sequence tagging, central word extraction, syntactic analysis, part-of-speech tagging, and the like.

Taking a part-of-speech tagging method as an example, extracting information in a part-of-speech tagging mode mainly comprises the following two steps: (1) generating a keyword candidate list; (2) selecting keywords by adopting a preset algorithm. In step (1), a keyword candidate table may be screened out through a preset rule, for example: in step (2), TF-IDF (term frequency-inverse document frequency, a common weighting technique for information retrieval and data mining) algorithm, text rank (keyword extraction algorithm) and other algorithms may be selected to extract keywords from the keyword candidate table.

As an optional embodiment, after obtaining the court trial record recorded in the court trial process, the method further includes: obtaining a text abstract technical model corresponding to each topic module in a court trial record; based on the text abstract technical model, extracting information elements and conclusions of each topic module; at least one summary text is generated based on the information elements and the conclusions of each topic module, wherein the summary text comprises at least one information element written in the referee document.

Specifically, in the court trial process, there are generally newly proposed complaints, evidences, questions, etc. except for complaints and answers, and these information elements need to be represented in the judge document in addition to being used in judge reasoning. It is therefore an essential step to summarize the information elements newly proposed in the current world.

According to the information elements newly proposed in the court, the generated abstract text at least comprises the information elements written in the referee document, and the information elements are used in referee pushing and need to be written in the referee text.

In an alternative embodiment, different text summarization technical models may be employed for different topic modules to draw different information elements and conclusions. The core elements come from legal knowledge maps constructed in the first step, and the technology used is extraction type abstract, so that different abstract models need to be trained in advance for different topic modules.

As an optional embodiment, after extracting the information element of each topic module from the court trial record, the method further includes: analyzing and processing the extracted information elements of each topic module based on the legal knowledge graph to obtain a judge result, wherein the legal knowledge graph at least comprises: a plurality of nodes, association relation and direction information between each node, each node being for characterizing one of: element, discrimination point and logic gate.

Specifically, in fig. 3, "number of times of original report in chinese referee net litigation", "number of times of original report in internet court litigation", and "whether to original report abusing" are elements, "whether to original report self-certifying abusing", "whether to be reported to certify original report self-certifying abusing", "whether to be in chinese referee net litigation 3 times or more", "whether to be in internet court litigation 3 times or more" are discrimination points, and "or" is a logic gate.

The information elements of each extracted topic module are analyzed and processed based on the legal knowledge graph, namely the information elements are used as input information of the legal knowledge graph, and the legal knowledge graph is traversed, so that a final judge result is obtained.

It should be noted that, in an alternative embodiment, before the information elements of each extracted topic module are analyzed and processed based on the legal knowledge graph, the case information acquired before the court trial may be further processed by using the legal knowledge graph to obtain an initial judge result, and then the information elements of each extracted topic module are analyzed and processed based on the legal knowledge graph to obtain a further judge result. And finally, combining the initial referee result with the further referee result to obtain a final referee result.

As an optional embodiment, the analyzing and processing the extracted information element of each topic module based on the legal knowledge graph to obtain the judge result includes: traversing the information elements of each topic module in a legal knowledge graph to obtain legal fact entities as judge results; in the traversing process, mapping the information elements of each topic module to a legal knowledge graph, and acquiring a judgment fact for generating a judgment result if the information elements are successfully mapped to the corresponding nodes.

Specifically, the legal facts refer to objective conditions that legal relationships can be generated, changed and eliminated. The legal facts entity refers to nodes in the legal knowledge graph for representing legal facts.

In the above scheme, the information element of each topic module is traversed in the legal knowledge graph to obtain the judging result, the traversing can be used for representing searching or searching in the legal knowledge graph, the information element of each topic module needs to be mapped to a corresponding node of the legal knowledge graph in the traversing process, and after the information element of each topic module is mapped to the node on the legal knowledge graph, the mapped node can be assigned according to the information element of each topic module.

After determining the nodes mapped by the information elements of each topic module in the legal knowledge graph, the mapped nodes can be assigned according to the information elements of each topic module, and the assigned values are information for generating a referee result.

Still taking fig. 3 as an example, for the node "number of litigation of original notice in the chinese referee document net", the number of litigation of original notice in the chinese referee document net may be extracted from the information element of the topic module and mapped to the node, so that the node "number of litigation of original notice in the chinese referee document net" in the legal knowledge map obtains the information for generating the referee result.

As an optional embodiment, traversing the information element of each topic module in the legal knowledge graph, and taking the obtained legal fact entity as a judge result, including: matching the information elements of the topic modules with candidate nodes in the legal knowledge graph, wherein each topic module corresponds to different candidate nodes in the legal knowledge graph; under the condition that the matching is successful, mapping to obtain at least one selected node; a legal facts entity of the topic module is determined based on the at least one selected node.

The information elements of the topic module represent legal information at this stage in the court trial process, and the information elements of the topic module are matched with candidate nodes of the legal knowledge graph to determine the corresponding relationship between the information elements of the topic module and the nodes in the legal knowledge graph.

In an alternative embodiment, the nodes in the legal knowledge graph have corresponding descriptive information, such as: name, attribute, semantic information, etc., node attributes may be used to indicate whether the node is a legal facts entity or an objective facts entity, and semantic information is used to describe the node further. When matching is performed, matching can be performed based on the semantics of the information elements of the topic module and the description information of the nodes.

And under the condition that the matching is successful, mapping to obtain at least one selected node, and after determining the selected node to which the information element of the topic module is mapped, determining a legal fact entity of the topic module according to the selected node, wherein the legal fact entity is used for representing the assignment of the information element of the topic module to the selected node.

As an optional embodiment, in the case that the topic module is a evidence module, after extracting the information element of each topic module from the court trial record, the method further includes: based on the evidence text corresponding to the evidence module, extracting evidence information needing evidence; traversing the evidence information in the corresponding evidence knowledge graph to obtain the identification result of the evidence; in the traversing process, the evidence information is mapped to the evidence knowledge graph, and if the evidence information is successfully mapped to the corresponding node, the information for generating the identification result is acquired.

In an alternative embodiment, the evidence newly generated in the court trial process can be selected from the evidence text corresponding to the evidence quality module as evidence information needing to be verified. For example, if the evidence text corresponding to the evidence module in the court trial record is the same as the evidence information submitted before the court trial, the evidence is considered to be unnecessary to be recognized again, and if the evidence text corresponding to the evidence module in the court trial record is the newly proposed evidence information in the court trial process and is different from the evidence information submitted before the court trial, the evidence text is considered to be required to be verified again.

Specifically, the identification process of the evidence information is essentially the identification process of the three properties (namely the authenticity, the relativity and the legality) of the evidence information, and the evidence identification module determines to adopt the evidence information only when the evidence information accords with the three characteristics, otherwise, the evidence information is not adopted.

The evidence knowledge graph can comprise a plurality of nodes, wherein the nodes have a directional connection relationship, when the evidence information is subjected to quality assurance, the evidence information is mapped in the evidence knowledge graph, and the evidence knowledge graph is traversed along with the directional connection relationship between the nodes until a final identification result is traversed.

In order to ensure the accuracy of the identification result of the evidence information, after the first identification result of the evidence information by the judging system is obtained, the feedback module can also receive feedback information of the first identification result, namely a second identification result, of the user, and generate a judging result according to the evidence information determined by the second identification result.

As an alternative embodiment, after the result of the affirmation of the obtained evidence, the method further comprises:

under the condition of successful mapping to the identification result of the evidence and the information elements of each topic module, determining an activation area of the court trial record in the legal knowledge graph, wherein the mapping is used for representing that the nodes in the legal knowledge graph are activated, and the activation area is formed by the area where the activated nodes in the legal knowledge graph are located;

and processing the court trial records in the activation area by using an uncertainty reasoning technology to obtain legal fact entities as judge results.

Specifically, the mapped node is the activated node, fig. 4 is a schematic diagram of an alternative legal indication map according to embodiment 1 of the present application, where white nodes are used to represent the nodes that are not mapped, i.e. the nodes that are not activated, and black nodes are used to represent the nodes that are successfully mapped, i.e. the nodes that are activated. The judge result obtained from the legal knowledge graph shown in fig. 4 is refund loan or triple reimbursement. The activated nodes constitute the activated regions in the legal knowledge graph.

Specifically, the uncertain reasoning technology refers to a process of reasoning a certain degree of uncertainty, but a reasonable or nearly reasonable conclusion by using uncertain indication or rule from the evidence of uncertainty. When the uncertain reasoning technology is applied to the scheme, the case information and the court trial record have certain uncertainty, and the legal indication map has certain uncertainty, so that the deduced legal fact entity also has certain uncertainty, but is reasonable.

As an alternative embodiment, in the activation area, the case information is processed by using an uncertainty reasoning technology to obtain a referee result, including: traversing the court trial record in the legal knowledge graph based on the association information and the direction information among the nodes in the legal knowledge graph to obtain nodes for pointing to the judge result corresponding to the case information; and taking the legal fact entity pointed by the obtained node as a judge result.

Specifically, because the association relationship (for example, the logic relationship) among the plurality of nodes in the legal knowledge graph has directionality, the court trial record can be based on the logic reasoning rule in the legal knowledge graph to traverse in the legal knowledge graph along the logic relationship direction until the legal fact entity is pointed, and the pointed legal fact entity is used as a final judge result.

As an alternative embodiment, in the activation area, the case information is processed by using an uncertainty reasoning technology to obtain a referee result, including: vectorizing each node in the legal knowledge graph to obtain a legal knowledge graph based on vectorization representation; on the legal knowledge graph based on vectorization representation, randomly walking on the legal knowledge graph based on court trial records, and determining a probability value corresponding to each walking node; and under the condition that the probability value of the node which is moved is larger than the preset probability, continuing to move the next node until the node for representing the judging result is obtained.

In an alternative embodiment, each node in the legal knowledge Graph may be vectorized by Graph embedding, so as to obtain the legal knowledge Graph based on vectorization representation.

In the above scheme, the random walk process in the legal knowledge graph may traverse the legal knowledge graph using a random walk (random walk) manner. And when walking randomly, the legal knowledge graph is regarded as a random field, and the random walk is carried out according to connectivity and transition probability among the nodes so as to transfer the association relationship among the nodes.

In the random walking process, the probability value corresponding to the traversed node can be determined by the probability value of the node at the upper layer pointing to the node, and under the condition that the node at the upper layer pointing to the node is multiple, the probability values of the multiple nodes are weighted according to the information of the preset weight of each side, so that the probability value corresponding to the node is obtained.

If the probability value of the node which is walked is larger than the preset probability, continuing to walk to the next node along the node at random, and if the probability value of the node is smaller than the preset probability, stopping the path and not walking on the downlink path where the node is located.

Optionally, according to a logic inference rule in the legal knowledge graph, traversing on the logic graph along the logic relationship direction, calculating a probability value corresponding to each node, and comparing the probability value of the node with a preset probability value, for example, the probability value corresponding to the node 1 is 80%, the preset probability value is 90%, and the probability value of the node 1 is smaller than the preset probability value, so that the judge inference module does not judge the case by adopting the content corresponding to the node 1, and does not traverse the next node of the node 1. If the probability value corresponding to the node 2 is 95%, and the preset probability value is 90%, and the probability value of the node 2 is greater than the preset probability value, the judge reasoning module judges the case by adopting the content corresponding to the node 2, and continues to traverse the next node of the node 2.

As an alternative embodiment, after obtaining the legal facts entity as the referee result, the method further comprises: receiving feedback information of a target object aiming at a judging result; and adjusting the judge result of the case according to the feedback information.

Specifically, the feedback information can be sent by a judge, the disputed focus is a risk point when the judge is judged, and after the algorithm feeds back the finally obtained judge result and the disputed focus to the judge through an interactive interface, the judge determines whether to determine or modify the disputed focus, and the information is the feedback information. After the judge modifies the assignment of the dispute focus, the judge result needs to be adjusted according to the dispute focus modified by the judge.

Fig. 5 is a schematic diagram of an interactive interface displayed to a judge by an algorithm according to embodiment 1 of the present application, and in combination with fig. 5, the problem points are the above-mentioned dispute points, and after each problem point, judging opinions given by the algorithm, that is, "whether the judging opinion accords with facts" and evidence corresponding to the dispute point are displayed, so that the judge can modify the judging result of the dispute point or whether the evidence accords with facts, so that the algorithm adjusts the judging result according to the modified evidence and the problem points.

As an alternative embodiment, the risk point is determined according to the credibility of each node, and the steps include: determining a risk degree corresponding to the credibility of each node based on the risk model, wherein the risk degree is used for representing the conditional probability of the referee result on the credibility of each node; and sequencing the risk degrees corresponding to each node, and setting the node with the largest risk degree as a risk point. Specifically, the reliability refers to the probability that the assignment of a node is true.

Specifically, the credibility of the node is used for representing the integrity degree and the accuracy degree of the information element of the topic module mapped to the node, and the credibility of each node in the legal knowledge graph can be determined according to the credibility of the root node in the legal knowledge graph.

In an alternative embodiment, the risk level of each node value to the decision result is determined according to the value status of the node in the current legal knowledge graph, where the risk level of each node value to the decision result may be represented by the decision result and the conditional probability of other nodes to the current node, for example, the risk level of the current node is referred to as Pimp, pimp=p (node|case, other nodes). The calculation mode of the conditional probability can be obtained by a statistical method through enumerating all values of the nodes.

After the Pimp value of each node is obtained, the node corresponding to the Pimp maximum value can be used as a risk point, and the node with Pimp higher than the threshold can be screened out through the Pimp threshold which is set empirically, and the screened out nodes are used as examination points.

In order to make the user know the reason of the judge result more clearly, the value result of each risk point can be displayed in the interactive page again to serve as the reason of the departure degree early warning.

As an optional embodiment, adjusting the referee result of the case according to the risk point to obtain a target referee result includes: receiving a modification result of the target object on the content corresponding to the risk point; adjusting the judge result of the case according to the modification result to obtain an adjusted judge result; and taking the adjusted referee result as a target referee result.

Specifically, the target object may be a judge, and in the above scheme, through interaction between the judge and the intelligent judging system, the risk points and the result calculated by the algorithm are displayed to the judge, and are recognized and modified by the judge. And transmitting the modified result to an intelligent judging system, and recalculating the judging result.

The target referee result is the final referee result. After the content corresponding to the risk points is modified by the judges, the law knowledge graph needs to be traversed again according to the modified risk points, so that an adjusted judge result is obtained.

In an alternative embodiment, fig. 6 is a schematic diagram of a forensic feedback risk point according to embodiment 1 of the present application, and in combination with fig. 6, the risk point one is used to represent "unsafe food", the evidence used to determine the risk point one includes original evidence and reported evidence, the risk point one has options "yes" and "no", currently shows a determination result given by the algorithm, that is, "unsafe food", the forensic can select "no" to manually change the determination result of the risk point, or modify the evidence corresponding to the risk point one (the "yes" and "no" options after the evidence is modified, determine whether the evidence is trustworthy), and the algorithm determines the risk point again and again according to the modified evidence. Similarly, the judge may also make modifications to the second risk point.

As an optional embodiment, obtaining a court trial record recorded in a court trial process includes: converting audio data in the multimedia information for recording the court trial process into text content; and carrying out segmentation processing on the text content to obtain a plurality of theme modules.

Specifically, the audio data recorded in the multimedia information of the court trial process may be an audio recording or a video recording in the court trial process, in the above scheme, after the audio recording or the video recording in the court trial process is obtained, the audio recording or the video recording is passed through ASR (Automatic Speech Recognition, voice-to-text technology), the audio information in the audio recording or the video recording is converted into a corresponding raw text (raw text, i.e. corpus not processed by a computer), and the raw text is used as the court trial record.

In an alternative embodiment, the text content is divided, that is, the text content corresponding to the court hearing role is obtained by recognizing the voice, that is, the content spoken by the court hearing aid in the court hearing process is obtained, then a classification model is constructed based on the content spoken by the court hearing aid, and the start mark and the end mark of each topic module in the text content are determined based on the classification model, so that the text content is divided into multiple parts of content, and each part of content is a topic module.

FIG. 7 is a flow chart of a legal case referee method according to embodiment 1 of the present application, combined with FIG. 7, comprising the following steps:

s71, legal knowledge graph construction.

The legal knowledge atlas is used for automatically judging and reasoning legal cases, different cases possibly have corresponding legal knowledge atlas, a seed knowledge base can be formed by means of logical combing and summarizing of expert knowledge, and the legal knowledge atlas can be expanded and updated by combining big data knowledge with machine learning.

In the construction and constitution of the legal knowledge graph, objective fact entities and legal fact entities are distinguished according to the judgment logic, so that a mapping relation is formed between the objective fact entities and the legal fact entities, and the final judgment is realized. Thus, legal knowledge graph includes attribute relationships among entities, for example: "original-attribute relationship-identification number", "original-attribute relationship (employment relationship) -original lawyer", a logical inference relationship is also defined for describing the conversion relationship of objective facts to legal facts, such as: "commodity description-inference relation-whether or not there is a medical effect advertised".

S72, court trial record analysis.

Any of the processing methods of the court trial records described in embodiment 1 can be used for analysis of the court trial records.

S73, information extraction/data mining.

The information extraction or information mining module extracts or mines the case information used as the input of legal knowledge graph. In an alternative embodiment, the input sources of the case information may include prosecution books, answer books, evidence information, court trial records, and the like, and may further include data network data that is kept, for example, for claim disputes of a counterfeit and inferior commodity, the original reported information may be extracted from the prosecution books; extracting whether refund and reserve the right to pursue from the evidence transaction log; extracting discount price, actual price and other information from the evidence commodity information; the number of complaints to be complained is mined from the user portraits of the shopping website.

S74, confirming the three evidences.

The essence of the process of recognizing evidence information is the process of recognizing the three properties (namely authenticity, relatedness and legitimacy) of the evidence information, and the evidence recognition module determines to adopt the evidence information only if the evidence information accords with the three characteristics, otherwise, the evidence information is not adopted.

S75, automatic judge reasoning.

In the above step S73, a legal knowledge graph may be obtained; based on the court trial records, determining a corresponding activation area of legal cases in the legal knowledge graph, and processing case information in the activation area by using an uncertainty reasoning technology to obtain a judge result.

S76, automatic risk point identification.

In the above steps, the risk point may be determined according to the reliability of each node, specifically, the risk degree corresponding to the reliability of each node may be determined based on the risk model, where the risk degree is used to represent the conditional probability of the judge result on the reliability of each node; and sequencing the risk degrees corresponding to each node, and setting the node with the largest risk degree as a risk point.

And S77, the judge confirms.

And the judge is that the judge feeds back according to the judge result and the evidence identification result fed back by the algorithm, whether the content fed back by the algorithm is identified is confirmed, if the content fed back by the algorithm is modified by the judge, the judge reasoning is carried out again by the algorithm according to the content modified by the judge, and the adjusted judge result is obtained.

S78, generating a judging result.

And obtaining a final judge result based on the identified evidence and the automatic judge reasoning.

It should be noted that, for simplicity of description, the foregoing method embodiments are all described as a series of acts, but it should be understood by those skilled in the art that the present invention is not limited by the order of acts described, as some steps may be performed in other orders or concurrently in accordance with the present invention. Further, those skilled in the art will also appreciate that the embodiments described in the specification are all preferred embodiments, and that the acts and modules referred to are not necessarily required for the present invention.

From the description of the above embodiments, it will be clear to a person skilled in the art that the method according to the above embodiments may be implemented by means of software plus the necessary general hardware platform, but of course also by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (e.g. ROM/RAM, magnetic disk, optical disk) comprising several instructions for causing a terminal device (which may be a mobile phone, a computer, a server, or a network device, etc.) to perform the method of the various embodiments of the present invention.

Example 2

According to an embodiment of the present invention, there is further provided an apparatus for processing a court trial record for implementing the method for processing a court trial record in embodiment 1, and fig. 8 is a schematic diagram of an apparatus for processing a court trial record according to embodiment 2 of the present application, as shown in fig. 8, and the apparatus 800 includes:

the first obtaining module 802 is configured to obtain a court trial record recorded in a court trial process, where the court trial record includes at least one topic module, and the topic module records at least legal information generated in different court trial stages in the court trial process.

The first determining module 804 is configured to determine an information extraction model corresponding to each topic module based on the identification information of different topic modules and corresponding preset parameters.

The first extraction module 806 is configured to extract, based on the information extraction model, an information element of each topic module from the court trial record, where the information element is used to initialize a legal knowledge graph.

It should be noted that, the first obtaining module 802, the first determining module 804, and the first extracting module 806 correspond to steps S21 to S25 in embodiment 1, and the three modules are the same as the examples and application scenarios implemented by the corresponding steps, but are not limited to those disclosed in the first embodiment. It should be noted that the above-described module may be operated as a part of the apparatus in the computer terminal 10 provided in the first embodiment.

As an alternative embodiment, the above device further comprises: the second acquisition module is used for acquiring a text abstract technical model corresponding to each topic module in the court trial record after acquiring the court trial record recorded in the court trial process; the second extraction module is used for extracting information elements and conclusions of each topic module based on the text abstract technical model; and the generation module is used for generating at least one abstract text based on the information elements and the conclusions of each topic module, wherein the abstract text comprises at least one information element written in the judge document.

As an alternative embodiment, the above device further comprises: the analysis module is used for analyzing and processing the extracted information elements of each topic module based on the legal knowledge graph after extracting the information elements of each topic module from the court trial record to obtain a judge result, wherein the legal knowledge graph at least comprises: a plurality of nodes, association relation and direction information between each node, each node being for characterizing one of: element, discrimination point and logic gate.

As an alternative embodiment, the analysis module comprises: the first traversing sub-module is used for traversing the information elements of each topic module in the legal knowledge graph, and the obtained legal fact entity is used as a judge result; in the traversing process, mapping the information elements of each topic module to a legal knowledge graph, and acquiring a judgment fact for generating a judgment result if the information elements are successfully mapped to the corresponding nodes.

As an alternative embodiment, the first traversal submodule includes: the matching unit is used for matching the information elements of the topic modules with candidate nodes in the legal knowledge graph, wherein each topic module corresponds to different candidate nodes in the legal knowledge graph; the mapping unit is used for mapping to obtain at least one selected node under the condition that the matching is successful; and the determining unit is used for determining legal fact entities of the theme module based on at least one selected node.

As an alternative embodiment, the above device further comprises: the third extraction module is used for extracting evidence information needing to be verified based on the evidence text corresponding to the evidence verification module after extracting the information element of each topic module from the court trial record under the condition that the topic module is the evidence verification module; the traversing module is used for traversing the evidence information in the corresponding evidence knowledge graph to obtain the identification result of the evidence; in the traversing process, the evidence information is mapped to the evidence knowledge graph, and if the evidence information is successfully mapped to the corresponding node, the information for generating the identification result is acquired.

As an alternative embodiment, the above device further comprises: the second determining module is used for determining an activation area of the court trial record in the legal knowledge graph under the condition of successfully mapping the identification result of the obtained evidence and the information elements of each topic module after the identification result of the obtained evidence, wherein the mapping success is used for representing that the nodes in the legal knowledge graph are activated, and the activation area is formed by the area where the activated nodes in the legal knowledge graph are located; and the reasoning module is used for processing the court trial records in the activation area by using an uncertainty reasoning technology to obtain legal fact entities serving as judge results.

As an optional embodiment, the reasoning module includes a second traversal sub-module, configured to traverse the court trial record in the legal knowledge graph based on association information and direction information among the plurality of nodes in the legal knowledge graph, and obtain a node for pointing to a referee result corresponding to the case information; the first determining submodule is used for taking the legal fact entity pointed by the acquired node as a judge result.

As an alternative embodiment, the reasoning module comprises: the processing submodule is used for carrying out vectorization processing on each node in the legal knowledge graph to obtain the legal knowledge graph based on vectorization representation; the random walking module is used for randomly walking on the legal knowledge graph based on the vectorization representation and on the legal knowledge graph based on the court trial record, and determining a probability value corresponding to each walking node; and the continuous walking module is used for continuously walking the next node until the node for representing the judging result is obtained under the condition that the probability value of the walking node is larger than the preset probability.

As an alternative embodiment, the above device further comprises: the receiving module is used for receiving feedback information of the target object aiming at the judge result after obtaining the legal fact entity serving as the judge result; the first adjusting module is used for adjusting the judge result of the case according to the feedback information.

As an alternative embodiment, the above device further comprises: the third determining module is configured to determine a risk point according to the reliability of each node, and the third determining module further includes: the second determining submodule is used for determining the risk degree corresponding to the credibility of each node based on the risk model, wherein the risk degree is used for representing the conditional probability of the referee result on the credibility of each node; and the setting sub-module is used for sequencing the risk degrees corresponding to each node and setting the node with the largest risk degree as a risk point.

As an alternative embodiment, the above device further comprises: the second adjusting module is used for adjusting the judge result of the case according to the risk points to obtain a target judge result; wherein the second adjustment module comprises: the receiving sub-module is used for receiving the modification result of the target object on the content corresponding to the risk point; the adjusting sub-module is used for adjusting the judge result of the case according to the modification result to obtain an adjusted judge result; and the third determination submodule is used for taking the adjusted referee result as a target referee result.

As an alternative embodiment, the first acquisition module includes: the conversion sub-module is used for converting audio data in the multimedia information for recording the court trial process into text content; and the segmentation sub-module is used for carrying out segmentation processing on the text content to obtain a plurality of theme modules.

Example 3

According to an embodiment of the present invention, there is further provided an embodiment of a method for obtaining a referee result of a case, and the present application provides a flowchart of the method for obtaining a referee result of a case as shown in fig. 9. Fig. 9 is a flowchart of a method of acquiring a referee result of a case according to embodiment 3 of the present application.

Step S91, displaying a court trial record recorded in the court trial process, wherein the court trial record comprises at least one theme module, and the theme module at least records legal information generated in different court trial stages in the court trial process.

Step S93, outputting an information extraction model corresponding to each topic module, and extracting information elements of each topic module from the court trial record, wherein the information elements are used for initializing legal knowledge maps, and the information extraction model is obtained based on identification information of different topic modules and corresponding preset parameters.

Specifically, the legal knowledge graph is used for intelligently judging legal cases. In the scheme, based on the information extraction model, the information elements in each topic module are extracted from the court trial record and serve as the content corresponding to the element nodes in the legal knowledge graph, so that the legal knowledge graph can intelligently judge legal cases according to the input required information elements.

Example 4

According to an embodiment of the present invention, there is further provided an apparatus for processing a court trial record for implementing the method for processing a court trial record in embodiment 3, and fig. 10 is a schematic diagram of an apparatus for processing a court trial record according to embodiment 4 of the present application, as shown in fig. 10, where the apparatus 100 includes:

the display module 1002 is configured to display a court trial record recorded in a court trial process, where the court trial record includes at least one theme module, and the theme module records at least legal information generated in different court trial stages in the court trial process.

The output module 1004 is configured to output an information element of each topic module extracted from the court trial record based on an information extraction model corresponding to each topic module, where the information element is used to initialize a legal knowledge graph, and the information extraction model is obtained based on identification information of different topic modules and corresponding preset parameters.

Here, it should be noted that the display module 1002 and the output module 1004 correspond to steps S91 to S93 in embodiment 3, and the two modules are the same as the examples and application scenarios implemented by the corresponding steps, but are not limited to the disclosure of the first embodiment. It should be noted that the above-described module may be operated as a part of the apparatus in the computer terminal 10 provided in the first embodiment.

Example 5

The embodiment of the invention can provide a judge result system for acquiring cases, which comprises the following steps:

a processor; and

a memory, coupled to the processor, for providing instructions to the processor for processing the steps of:

acquiring a court trial record recorded in a court trial process, wherein the court trial record comprises at least one theme module, and the theme module at least records legal information generated in different court trial stages in the court trial process;

determining an information extraction model corresponding to each topic module based on the identification information of different topic modules and corresponding preset parameters;

and extracting information elements of each topic module from the court trial record based on the information extraction model, wherein the information elements are used for initializing legal knowledge maps.

The above memory is further used to provide instructions for the processor to process and execute other steps in embodiment 1, which will not be described herein.

Example 6

Embodiments of the present invention may provide a computer terminal, which may be any one of a group of computer terminals. Alternatively, in the present embodiment, the above-described computer terminal may be replaced with a terminal device such as a mobile terminal.

Alternatively, in this embodiment, the above-mentioned computer terminal may be located in at least one network device among a plurality of network devices of the computer network.

In this embodiment, the computer terminal may execute the program code of the following steps in the vulnerability detection method of the application program: acquiring a court trial record recorded in a court trial process, wherein the court trial record comprises at least one theme module, and the theme module at least records legal information generated in different court trial stages in the court trial process; determining an information extraction model corresponding to each topic module based on the identification information of different topic modules and corresponding preset parameters; and extracting information elements of each topic module from the court trial record based on the information extraction model, wherein the information elements are used for initializing legal knowledge maps.

Alternatively, fig. 11 is a block diagram of a computer terminal according to an embodiment of the present invention. As shown in fig. 11, the computer terminal a may include: one or more (only one is shown) processors 1102, memory 1104, and peripheral devices 1106.

The memory may be used to store software programs and modules, such as program instructions/modules corresponding to the security vulnerability detection method and device in the embodiments of the present invention, and the processor executes various functional applications and data processing by running the software programs and modules stored in the memory, thereby implementing the above-mentioned method for detecting a system vulnerability attack. The memory may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory may further include memory remotely located with respect to the processor, which may be connected to terminal a through a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.

The processor may call the information and the application program stored in the memory through the transmission device to perform the following steps: acquiring a court trial record recorded in a court trial process, wherein the court trial record comprises at least one theme module, and the theme module at least records legal information generated in different court trial stages in the court trial process; determining an information extraction model corresponding to each topic module based on the identification information of different topic modules and corresponding preset parameters; and extracting information elements of each topic module from the court trial record based on the information extraction model, wherein the information elements are used for initializing legal knowledge maps.

Optionally, the above processor may further execute program code for: the information extraction model is used for representing information extraction methods adopted for legal information generated in different court trial stages.

Optionally, the above processor may further execute program code for: obtaining a text abstract technical model corresponding to each topic module in a court trial record; based on the text abstract technical model, extracting information elements and conclusions of each topic module; at least one summary text is generated based on the information elements and the conclusions of each topic module, wherein the summary text comprises at least one information element written in the referee document.

Optionally, the above processor may further execute program code for: after extracting information elements of each topic module from the court trial record, analyzing and processing the extracted information elements of each topic module based on a legal knowledge graph to obtain a judge result, wherein the legal knowledge graph at least comprises: a plurality of nodes, association relation and direction information between each node, each node being for characterizing one of: element, discrimination point and logic gate.

Optionally, the above processor may further execute program code for: traversing the information elements of each topic module in a legal knowledge graph to obtain legal fact entities as judge results; in the traversing process, mapping the information elements of each topic module to a legal knowledge graph, and acquiring a judgment fact for generating a judgment result if the information elements are successfully mapped to the corresponding nodes.

Optionally, the above processor may further execute program code for: matching the information elements of the topic modules with candidate nodes in the legal knowledge graph, wherein each topic module corresponds to different candidate nodes in the legal knowledge graph; under the condition that the matching is successful, mapping to obtain at least one selected node; a legal facts entity of the topic module is determined based on the at least one selected node.

Optionally, the above processor may further execute program code for: under the condition that the topic module is an evidence quality module, after information elements of each topic module are extracted from the court trial record, evidence information needing quality evidence is extracted based on evidence texts corresponding to the evidence quality evidence module; traversing the evidence information in the corresponding evidence knowledge graph to obtain the identification result of the evidence; in the traversing process, the evidence information is mapped to the evidence knowledge graph, and if the evidence information is successfully mapped to the corresponding node, the information for generating the identification result is acquired.

Optionally, the above processor may further execute program code for: after the obtained evidence identification result, under the condition of successfully mapping the evidence identification result and the information elements of each topic module, determining an activation area of the court trial record in the legal knowledge graph, wherein the mapping success is used for representing that the nodes in the legal knowledge graph are activated, and the activation area is formed by the area where the activated nodes in the legal knowledge graph are located; and processing the court trial records in the activation area by using an uncertainty reasoning technology to obtain legal fact entities as judge results.

Optionally, the above processor may further execute program code for: traversing the court trial record in the legal knowledge graph based on the association information and the direction information among the nodes in the legal knowledge graph to obtain nodes for pointing to the judge result corresponding to the case information; and taking the legal fact entity pointed by the obtained node as a judge result.

Optionally, the above processor may further execute program code for: vectorizing each node in the legal knowledge graph to obtain a legal knowledge graph based on vectorization representation; on the legal knowledge graph based on vectorization representation, randomly walking on the legal knowledge graph based on court trial records, and determining a probability value corresponding to each walking node; and under the condition that the probability value of the node which is moved is larger than the preset probability, continuing to move the next node until the node for representing the judging result is obtained.

Optionally, the above processor may further execute program code for: after obtaining legal fact entity as referee result, receiving feedback information of target object for referee result; and adjusting the judge result of the case according to the feedback information.

Optionally, the above processor may further execute program code for: determining a risk degree corresponding to the credibility of each node based on the risk model, wherein the risk degree is used for representing the conditional probability of the referee result on the credibility of each node; and sequencing the risk degrees corresponding to each node, and setting the node with the largest risk degree as a risk point.

Optionally, the above processor may further execute program code for: receiving a modification result of the target object on the content corresponding to the risk point; adjusting the judge result of the case according to the modification result to obtain an adjusted judge result; and taking the adjusted referee result as a target referee result.

Optionally, the above processor may further execute program code for: converting audio data in the multimedia information for recording the court trial process into text content; and carrying out segmentation processing on the text content to obtain a plurality of theme modules.

By adopting the embodiment of the invention, a scheme for acquiring the judge result of the case is provided. The method comprises the steps of obtaining a court trial record recorded in a court trial process, wherein the court trial record comprises at least one theme module, and the theme module at least records legal information generated in different court trial stages in the court trial process; determining an information extraction model corresponding to each topic module based on the identification information of different topic modules and corresponding preset parameters; and extracting information elements of each topic module from the court trial record based on the information extraction model, wherein the information elements are used for initializing legal knowledge maps. Through the scheme, the online Internet court does not need a bookman to make a summary record of the court trial process, and the information elements can be automatically extracted from the court trial record according to the court trial record of the court trial process, so that the technical problem that the abstract of the pen record of the online Internet court is difficult to obtain in the prior art is solved, and the Internet court can be popularized to a traditional trial mode.

It will be appreciated by those skilled in the art that the structure shown in the figure is merely illustrative, and the computer terminal may be a smart phone (such as an Android phone, an iOS phone, etc.), a tablet computer, a palm computer, a mobile internet device (Mobile Internet Devices, MID), a PAD, etc. Fig. 10 is not limited to the structure of the electronic device. For example, the computer terminal 10 may also include more or fewer components (e.g., network interfaces, display devices, etc.) than shown in FIG. 10, or have a different configuration than shown in FIG. 10.

Those of ordinary skill in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be implemented by a program for instructing a terminal device to execute in association with hardware, the program may be stored in a computer readable storage medium, and the storage medium may include: flash disk, read-Only Memory (ROM), random-access Memory (Random Access Memory, RAM), magnetic or optical disk, and the like.

Example 4

The embodiment of the invention also provides a storage medium. Alternatively, in this embodiment, the storage medium may be used to store the program code executed by the judge result method for acquiring a case provided in the first embodiment.

Alternatively, in this embodiment, the storage medium may be located in any one of the computer terminals in the computer terminal group in the computer network, or in any one of the mobile terminals in the mobile terminal group.

Alternatively, in the present embodiment, the storage medium is configured to store program code for performing the steps of: acquiring a court trial record recorded in a court trial process, wherein the court trial record comprises at least one theme module, and the theme module at least records legal information generated in different court trial stages in the court trial process; determining an information extraction model corresponding to each topic module based on the identification information of different topic modules and corresponding preset parameters; and extracting information elements of each topic module from the court trial record based on the information extraction model, wherein the information elements are used for initializing legal knowledge maps.

The foregoing embodiment numbers of the present invention are merely for the purpose of description, and do not represent the advantages or disadvantages of the embodiments.

In the foregoing embodiments of the present invention, the descriptions of the embodiments are emphasized, and for a portion of this disclosure that is not described in detail in this embodiment, reference is made to the related descriptions of other embodiments.

In the several embodiments provided in the present application, it should be understood that the disclosed technology content may be implemented in other manners. The above-described embodiments of the apparatus are merely exemplary, and the division of the units, such as the division of the units, is merely a logical function division, and may be implemented in another manner, for example, multiple units or components may be combined or may be integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be through some interfaces, units or modules, or may be in electrical or other forms.

The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.

The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied essentially or in part or all of the technical solution or in part in the form of a software product stored in a storage medium, including instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a removable hard disk, a magnetic disk, or an optical disk, or other various media capable of storing program codes.

The foregoing is merely a preferred embodiment of the present invention and it should be noted that modifications and adaptations to those skilled in the art may be made without departing from the principles of the present invention, which are intended to be comprehended within the scope of the present invention.

Claims

1. A method of processing a court trial record, comprising:

obtaining a court trial record recorded in a court trial process, wherein the court trial record comprises at least one theme module, the theme module at least records legal information generated in different court trial stages in the court trial process, and after obtaining the court trial record recorded in the court trial process, the method further comprises the steps of: obtaining a text abstract technical model corresponding to each topic module in the court trial record; based on the text abstract technical model, extracting information elements and conclusions of each topic module; generating at least one abstract text based on the information elements and the conclusions of each topic module, wherein the abstract text comprises at least one information element written in a referee document;

2. The method of claim 1, wherein the information extraction model is used to characterize information extraction methods employed for legal information generated for different trial phases.

3. The method of any of claims 1-2, wherein after extracting information elements for each topic module from the trial log, the method further comprises:

analyzing and processing the extracted information elements of each topic module based on the legal knowledge graph to obtain a judge result, wherein the legal knowledge graph at least comprises: a plurality of nodes, association relation and direction information between each node, each node being for characterizing one of: element, discrimination point and logic gate.

4. The method of claim 3, wherein analyzing the extracted information element of each topic module based on the legal knowledge graph to obtain the judge result comprises:

traversing the information elements of each topic module in the legal knowledge graph to obtain legal fact entities as the judge result;

And mapping the information elements of each topic module to the legal knowledge graph in the traversing process, and acquiring a judgment fact for generating the judge result if the information elements are successfully mapped to the corresponding nodes.

5. The method of claim 4, wherein traversing the information element of each topic module in the legal knowledge graph, the legal fact entity obtained as the referee result comprises:

matching the information elements of the topic modules with candidate nodes in the legal knowledge graph, wherein each topic module corresponds to different candidate nodes in the legal knowledge graph;

under the condition that the matching is successful, mapping to obtain at least one selected node;

determining legal facts entity of the topic module based on the at least one selected node.

6. The method of claim 1, wherein, in the case where the topic module is a evidence module, after extracting information elements of each topic module from the court trial record, the method further comprises:

based on the evidence text corresponding to the evidence module, extracting evidence information needing evidence;

Traversing the evidence information in the corresponding evidence knowledge graph to obtain the identification result of the evidence;

and mapping the evidence information to the evidence knowledge graph in the traversing process, and acquiring information for generating the identification result if the evidence information is successfully mapped to the corresponding node.

7. The method of claim 6, wherein after the affirmation of the evidence obtained, the method further comprises:

under the condition of successful mapping to the identification result of the evidence and the information elements of each topic module, determining an activation area of the court trial record in the legal knowledge graph, wherein the mapping success is used for representing that the nodes in the legal knowledge graph are activated, and the activation area is formed by the area where the activated nodes in the legal knowledge graph are located;

and processing the court trial records in the activation area by using an uncertainty reasoning technology to obtain legal fact entities serving as judge results.

8. The method of claim 7, wherein processing the court trial records within the activation region using uncertainty reasoning techniques to arrive at a legal facts entity as a referee result comprises:

Traversing the court trial record in the legal knowledge graph based on the association information and the direction information among a plurality of nodes in the legal knowledge graph to obtain nodes for pointing to judge results corresponding to case information;

and taking the legal fact entity pointed by the acquired node as the judge result.

9. The method of claim 7, wherein processing the court trial records within the activation region using uncertainty reasoning techniques to arrive at a legal facts entity as a referee result comprises:

vectorizing each node in the legal knowledge graph to obtain a legal knowledge graph based on vectorization representation;

on the legal knowledge graph based on the vectorization representation, randomly walking on the legal knowledge graph based on the court trial record, and determining a probability value corresponding to each walking node;

and under the condition that the probability value of the node which is moved is larger than the preset probability, continuing to move the next node until the node for representing the judging result is obtained.

10. The method according to claim 8 or 9, wherein after obtaining a legal facts entity as a result of the referee, the method further comprises:

Receiving feedback information of a target object aiming at the judge result;

and adjusting the judge result of the case according to the feedback information.

11. The method of claim 8, wherein the risk points are determined according to the trustworthiness of each node, the step comprising:

determining a risk degree corresponding to the credibility of each node based on a risk model, wherein the risk degree is used for representing the conditional probability of the referee result on the credibility of each node;

and sequencing the risk degrees corresponding to each node, and setting the node with the largest risk degree as the risk point.

12. The method of claim 11, wherein adjusting the referee result for the case based on the risk point to obtain a target referee result comprises:

receiving a modification result of the target object on the content corresponding to the risk point;

adjusting the judge result of the case according to the modification result to obtain an adjusted judge result;

and taking the adjusted referee result as the target referee result.

13. The method of claim 1, wherein obtaining a court trial record of a recording during a court trial comprises:

Converting audio data in the multimedia information for recording the court trial process into text content;

and carrying out segmentation processing on the text content to obtain a plurality of theme modules.

14. A method for obtaining judge results of cases comprises the following steps:

displaying a court trial record recorded in a court trial process, wherein the court trial record comprises at least one theme module, the theme module at least records legal information generated in different court trial stages in the court trial process, and after the court trial record recorded in the court trial process is acquired, the method further comprises the steps of: obtaining a text abstract technical model corresponding to each topic module in the court trial record; based on the text abstract technical model, extracting information elements and conclusions of each topic module; generating at least one abstract text based on the information elements and the conclusions of each topic module, wherein the abstract text comprises at least one information element written in a referee document;

outputting an information extraction model corresponding to each topic module, and extracting information elements of each topic module from the court trial record, wherein the information elements are used for initializing legal knowledge maps, the information extraction model is obtained based on identification information of different topic modules and corresponding preset parameters, and after extracting the information elements of each topic module from the court trial record, the method further comprises the steps of: and analyzing and processing the extracted information elements of each topic module based on the legal knowledge graph to obtain a judge result.

15. A system for obtaining referee results for a case, comprising:

a processor; and

a memory, coupled to the processor, for providing instructions to the processor to process the following processing steps:

the method comprises the steps of obtaining a court trial record recorded in a court trial process, wherein the court trial record comprises at least one theme module, the theme module at least records legal information generated in different court trial stages in the court trial process, and after obtaining the court trial record recorded in the court trial process, the memory is further used for providing instructions for the processor to process the following processing steps:

obtaining a text abstract technical model corresponding to each topic module in the court trial record; based on the text abstract technical model, extracting information elements and conclusions of each topic module; generating at least one abstract text based on the information elements and the conclusions of each topic module, wherein the abstract text comprises at least one information element written in a referee document;

based on the information extraction model, extracting information elements of each topic module from the court trial record, wherein the information elements are used for initializing legal knowledge maps, and after the information elements of each topic module are extracted from the court trial record, the memory is further used for providing instructions for the processor to process the following processing steps: and analyzing and processing the extracted information elements of each topic module based on the legal knowledge graph to obtain a judge result.