CN113407727B - Qualitative measure and era recommendation method based on legal knowledge graph and related equipment - Google Patents
Qualitative measure and era recommendation method based on legal knowledge graph and related equipment Download PDFInfo
- Publication number
- CN113407727B CN113407727B CN202110304173.2A CN202110304173A CN113407727B CN 113407727 B CN113407727 B CN 113407727B CN 202110304173 A CN202110304173 A CN 202110304173A CN 113407727 B CN113407727 B CN 113407727B
- Authority
- CN
- China
- Prior art keywords
- violation
- label
- penalty
- case
- labels
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 37
- 230000006399 behavior Effects 0.000 claims abstract description 41
- 230000007246 mechanism Effects 0.000 claims abstract description 21
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 15
- 230000004044 response Effects 0.000 claims abstract description 14
- 238000004364 calculation method Methods 0.000 claims abstract description 13
- 238000013135 deep learning Methods 0.000 claims abstract description 13
- 238000004590 computer program Methods 0.000 claims description 6
- 238000004891 communication Methods 0.000 description 10
- 230000009471 action Effects 0.000 description 8
- 238000013528 artificial neural network Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 6
- 239000011159 matrix material Substances 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 238000013473 artificial intelligence Methods 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000007635 classification algorithm Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000001737 promoting effect Effects 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 230000004931 aggregating effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013145 classification model Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/18—Legal services
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Life Sciences & Earth Sciences (AREA)
- Databases & Information Systems (AREA)
- Tourism & Hospitality (AREA)
- Evolutionary Biology (AREA)
- Human Resources & Organizations (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Technology Law (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Animal Behavior & Ethology (AREA)
- Computational Linguistics (AREA)
- Artificial Intelligence (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The disclosure provides a qualitative measure and discipline recommendation method based on a legal knowledge graph and related equipment. The method comprises the following steps: in response to receiving text description of the illegal case, based on the illegal behavior label tree, performing multi-label classification on the text description by using a multi-label attention mechanism of a deep learning algorithm to predict an illegal behavior category label corresponding to the illegal case; matching and reasoning calculation are carried out on the basis of a punishment rule knowledge graph according to the violation behavior category label so as to determine punishment clause information and violation property label corresponding to the violation case; and outputting the punishment clause information and the violation property label as a recommendation qualitative result of the violation case. According to the method, legal information is automatically extracted from existing related documents and cases, a set of knowledge maps in related fields is established, reasonable judicial judgment is given based on the maps, and assistance is provided for judicial authorities to judge cases.
Description
Technical Field
The disclosure relates to the technical field of knowledge graphs, in particular to a qualitative quantum discipline recommendation method based on legal knowledge graphs and related equipment.
Background
The construction of legal knowledge maps is an indispensable basic project for realizing wisdom judicial law. In recent years, the judicial field actively applies advanced technologies such as big data, cloud computing, artificial intelligence and the like, effectively improves the case handling efficiency, assists judicial management, serves litigation of people, and has a great promoting effect on accelerating the modernization of the judicial. The legal knowledge map is used for extracting the laws in the related fields, the entities in the laws and the ordinances, and the relations and attributes between the entities according to the application scenes, constructing the logical relations and forming the structured knowledge expression. Based on the legal knowledge map, the intelligent application of the judicial service scene is finally realized, and effective reference and basis are provided for case handling and case judgment of judicial personnel.
The qualitative measure belongs to the category of legal decision-making, and the rationale thereof is legal demonstration. Traditional legal demonstration is mainly completed manually, the quality degree of the demonstration completely depends on the demonstration capability and knowledge reserve of a legal verifier, the verifier serving as a natural person has limitations of value orientation, knowledge structure and the like, and the obtained decision is often influenced by subjective factors. In legal decision-making, especially in the stage of insufficient experience of legal argumentation person, the decision is often misbiased due to common sense or logical error, and the related intelligent assistance system for legal decision-making can reduce the occurrence of such error by ensuring partial computability of legal problem.
Disclosure of Invention
In view of this, the present disclosure aims to provide a qualitative quantum discipline recommendation method based on legal knowledge graphs and related devices.
Based on the above purpose, the present disclosure provides a qualitative quantum discipline recommendation method based on a legal knowledge graph, including:
in response to receiving text description of the violation case, performing multi-label classification on the text description by using a multi-label attention mechanism of a deep learning algorithm based on a violation label tree to predict a violation category label corresponding to the violation case;
matching and reasoning calculation are carried out on the basis of a punishment rule knowledge graph according to the violation behavior category label so as to determine punishment clause information and violation property label corresponding to the violation case;
outputting the penalty term information and the violation property tag as a qualitative recommendation for the violation case,
wherein the penalty specification knowledge-graph is constructed in advance based on a penalty specification related file for the violation, and the violation label tree is constructed in advance based on the penalty specification related file for the violation.
Based on the same inventive concept, the disclosure also provides a qualitative quantum discipline recommendation method based on the legal knowledge graph, which comprises the following steps:
in response to receiving text description of an illegal case and an episode severity label of the illegal case, performing multi-label classification on the text description by using a multi-label attention mechanism of a deep learning algorithm based on an illegal action label tree to predict an illegal action category label corresponding to the illegal case;
according to the violation behavior category label and the plot severity label, performing matching and inference calculation based on a penalty rule knowledge graph to determine penalty clause information and violation property label corresponding to the violation case and penalty measure information for the violation case;
outputting the penalty term information, the violation property label, and the penalty measure information as a result of the recommended determination for the violation case,
wherein the penalty specification knowledge-graph is constructed in advance based on a penalty specification related file for the violation, and the violation label tree is constructed in advance based on the penalty specification related file for the violation.
Based on the same inventive concept, the present disclosure also provides a qualitative quantum discipline recommendation device based on legal knowledge graph, including:
a classification module configured to: in response to receiving text description of the violation case, performing multi-label classification on the text description by using a multi-label attention mechanism of a deep learning algorithm based on a violation label tree to predict a violation category label corresponding to the violation case;
a matching and reasoning module configured to: matching and reasoning calculation are carried out based on a penalty rule knowledge graph according to the violation behavior category label so as to determine penalty clause information and violation property label corresponding to the violation case;
an output module configured to output the penalty term information and the violation property tag as a qualitative recommendation for the violation case,
wherein the penalty rule knowledge-graph is constructed in advance based on a penalty rule related file for the violation, and the violation label tree is constructed in advance based on the penalty rule knowledge-graph.
Based on the same inventive concept, the present disclosure also provides a qualitative quantum discipline recommendation device based on legal knowledge graph, comprising: a classification module configured to: in response to receiving text description of an illegal case and an episode severity label of the illegal case, performing multi-label classification on the text description by using a multi-label attention mechanism of a deep learning algorithm based on an illegal action label tree to predict an illegal action category label corresponding to the illegal case;
a matching and reasoning module configured to: matching and reasoning calculation are carried out on the basis of a punishment regulation knowledge graph according to the violation behavior category label and the plot severity label so as to determine punishment clause information and violation property label corresponding to the violation case and punishment measure information aiming at the violation case;
an output module configured to output the penalty term information, the violation property label, and the penalty measure information as a result of a recommended determination for the violation case,
wherein the penalty rule knowledge-graph is constructed in advance based on a penalty rule related file for the violation, and the violation label tree is constructed in advance based on the penalty rule knowledge-graph.
Based on the same inventive concept, the present disclosure also provides an electronic device comprising a memory, a processor and a computer program stored on the memory and executable by the processor, the processor implementing the qualitative quantum computation recommendation method based on a legal knowledge graph as described above when executing the computer program.
From the above, according to the qualitative quantum discipline recommendation method based on legal knowledge graphs and the related equipment provided by the disclosure, the legal information is automatically extracted from the existing legal related files and cases, a set of knowledge graphs in related fields are established, for a given case, matching reasoning is performed based on the graphs, and finally reasonable judicial judgment is given, so that the defects that the traditional legal recommendation data is insufficient in structuralization and the intelligent technology recommendation is not accurate enough are overcome, and assistance is provided for judicial authorities to judge cases.
Drawings
In order to more clearly illustrate the technical solutions in the present disclosure or related technologies, the drawings needed to be used in the description of the embodiments or related technologies are briefly introduced below, and it is obvious that the drawings in the following description are only embodiments of the present disclosure, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.
FIG. 1 is a schematic flow chart diagram of a qualitative quantum recommendation method based on legal knowledge graphs in accordance with an embodiment of the present disclosure;
FIG. 2 is a schematic illustration of a hierarchical relationship between various types of penalty term nodes in a penalty specification knowledge-graph according to an embodiment of the disclosure;
FIG. 3 is a schematic diagram of a qualitative quantum discipline recommendation device based on legal knowledge maps in accordance with an embodiment of the present disclosure;
fig. 4 is a schematic structural diagram of an electronic device according to an embodiment of the disclosure.
Detailed Description
For the purpose of promoting a better understanding of the objects, aspects and advantages of the present disclosure, reference is made to the following detailed description taken in conjunction with the accompanying drawings.
It is to be noted that technical terms or scientific terms used in the embodiments of the present disclosure should have a general meaning as understood by one having ordinary skill in the art to which the present disclosure belongs, unless otherwise defined. The use of "first," "second," and similar terms in the embodiments of the disclosure is not intended to indicate any order, quantity, or importance, but rather is used to distinguish one element from another. The word "comprising" or "comprises", and the like, means that the element or item listed before the word covers the element or item listed after the word and its equivalents, but does not exclude other elements or items. The terms "connected" or "coupled" and the like are not restricted to physical or mechanical connections, but may include electrical connections, whether direct or indirect.
As described in the background, unlike the conventional legal information system, the penalty rule knowledge base is a structured semantic knowledge base processed by analyzing and then processing violation-rule-related documents, and is used for describing concepts and relationships thereof in violation-rule rules. The method and the device realize quick response and reasoning of knowledge by decomposing a large amount of data from coarse granularity at a text level into fine granularity at a data level and aggregating a large amount of knowledge. Knowledge matching and reasoning are carried out based on the constructed penalty rule knowledge graph, so that the current situation of the traditional low-efficiency artificial intelligence algorithm with insufficient data structuralization and disordered algorithm can be changed.
The construction of the penalty rule knowledge graph needs to extract key information from penalty rule related files aiming at illegal behaviors, analyze and sort the key information, clarify the logic relationship among information points and finally structure the information points. Classifying the illegal and discipline behaviors of a given case needs to be classified by applying a deep learning classification algorithm based on the class of the illegal and discipline behaviors in a related file. The classification algorithm integrates the hierarchy inheritance of the classes into an algorithm structure so as to improve the accuracy of final classification.
The technical solution of the present disclosure is described in detail below with reference to the accompanying drawings.
Referring to fig. 1, an embodiment of the present disclosure provides a qualitative quantum discipline recommendation method based on a legal knowledge graph, including the following steps:
step S101, in response to receiving text description of the violation case, based on the violation label tree, performing multi-label classification on the text description by using a multi-label attention mechanism of a deep learning algorithm to predict a violation category label corresponding to the violation case.
In some embodiments, case descriptions are classified based on a constructed violation label tree, each node of which is a violation category label. The violation label tree is a tree-structured knowledge graph with inheritance relationships between each level. In this embodiment, the penalty rule knowledge graph is constructed according to "Communist Party of China discipline division" (hereinafter, referred to as "regulation"), the law in the "regulation" has natural class inheritance and conforms to the attribute of the violation label tree, and the violation class label having an inheritance relationship is extracted from the "regulation" as a node for constructing the violation label tree. And classifying the illegal case description through the constructed illegal behavior tag tree. After the user inputs the description of the violation cases, the deep learning algorithm is applied to classify the violation cases, the hierarchical structure of the violation label tree is integrated into the classification model, and the final classification accuracy can be improved.
And S102, according to the violation behavior category label, performing matching and inference calculation based on a penalty rule knowledge graph to determine penalty clause information and a violation property label corresponding to the violation case.
In some embodiments, the violation category label is a part of nodes in the penalty rule knowledge graph, the same node of the violation category label in the penalty rule knowledge graph is found through regular matching, and other data including specific laws, money, terms, quantity and discipline information and the like are associated through the node, so that penalty term information and violation property label corresponding to the violation case are determined.
And step S103, outputting the punishment clause information and the violation property label as a recommendation qualitative result of the violation case.
The recommendation qualitative result comprises violation regulations related to the case, final qualitative behaviors and reference quantitative criteria. And (4) taking the fixed recommendation qualitative result as a reference by related judicial personnel, and making qualitative quantitative determination on the behavior described by the illegal case according to the specific situation of the case and the past experience. Wherein the penalty specification knowledge-graph is constructed in advance based on a penalty specification related file for the violation, and the violation label tree is constructed in advance based on the penalty specification related file for the violation.
In some embodiments, their corresponding penalty rules knowledge-graph is constructed based on the contents of the "regulations", with structural levels from high to low as: bars, money, items. The hierarchy that may occur includes: a single strip; bars and money; bars, money, items; bars, items. The nodes that build the knowledge-graph according to the hierarchy may contain containment relationships as shown in fig. 2. Wherein the hierarchical label of "style" is not explicitly given in the "rules" but is presented in a segmented form.
In some embodiments, after outputting the penalty term information and the violation property label, in response to receiving a user-entered episode severity label for the violation case, matching the penalty term information and the episode severity label to the penalty specification knowledge-graph to determine penalty action information for the violation case; and outputting the penalty measure information as a recommended penalty measure result for the violation case. The plot severity labels correspond to different punishment measures, and the labels can be specifically divided into 'light plot', 'heavy plot' and 'severe plot'.
The established penalty rule knowledge graph provides a foundation for the qualitative and quantitative era of subsequent cases.
In some embodiments, the violation label tree has an inheritance hierarchy in which a first-level label, a plurality of second-level labels, and a plurality of third-level labels are arranged in a high-to-low order. The node labels in the violation label tree can be associated with parent labels or child labels, so that the qualitative behaviors of the violation case description are preliminarily determined.
In some embodiments, multi-label classifying the textual description using the multi-label attention mechanism to predict the violation category label comprises: classifying the text description according to the second-layer labels by using the multi-label attention mechanism to obtain respective first scores of the second-layer labels; sorting the plurality of second-layer labels from high to low according to the first score, and selecting the first N second-layer labels in the sorted plurality of second-layer labels, wherein N represents a preset number; for each of the first N second-level tags, determining a third-level tag of the plurality of third-level tags that is associated with the second-level tag as a candidate tag based on the violation behavior tag tree; classifying the text description according to the candidate tags by using the multi-tag attention mechanism to obtain respective second scores of the candidate tags; determining the candidate label with the highest second score among the candidate labels as the violation behavior category label.
Specifically, the multi-label classification mainly comprises the following steps:
1. inputting the illegal case description into a Bert-Chinese pre-training model for coding to obtain the vector representation of each word of the text description
Wherein
Is the number of words of the text description.
2. Computing a weighted average of each word vector using an attention layer in a neural network, resulting in a vector representation of the textual description for different violation category labels
Wherein alpha is ij The weight parameters can be updated through training and continuous optimization, j =1, \ 8230, and L is the number of labels of the violation categories.
In the present embodiment, the word vector based representation
Calculating a vector representation of the textual description corresponding to each second-layer label by an attention mechanism
The second layer label is used as the first vector representation, and the total number of the second layer label is L, namely L vector representations of the text description are obtained
3. And performing calculation conversion on the first vector representation through a full connection layer and an output layer of the neural network to obtain first scores of the first vector representation, wherein each first score corresponds to each second layer label, and the higher the value of the first score is, the greater the relevance between the corresponding second layer label and the text description is.
4. And selecting the next-level labels associated with the first N second-level labels with higher first score values, namely the third-level labels as candidate labels.
5. Word vector representation based on the textual description
Referring to step 2, a vector representation of the textual description corresponding to each candidate tag is derived by the attention layer of the neural network
As a second vector representation, where P is the number of sub-tags, i.e. a vector representation of P said textual descriptions is obtained
6. And performing calculation conversion on the second vector representation through a full connection layer and an output layer of the neural network to obtain second scores of the second vector representation, wherein each second score corresponds to each candidate label, and the higher the value of the second score is, the greater the relevance between the corresponding sub-label and the text description is.
7. And selecting the candidate label with the highest second score as the violation behavior category label, that is, selecting the candidate label most related to the text description as the violation behavior category label of the text description, or selecting several labels with higher scores as the violation behavior category labels of the text description according to specific situations.
In some embodiments, the first vector representation is associated with a first weight matrix by a fully-connected layer of a neural network
Multiplying and outputting a first intermediate hidden layer vector; combining the first intermediate hidden layer vector with a second weight matrix through an output layer of a neural network
Multiplying and outputting a first score represented by the first vector.
Specifically, in the above steps 3 and 6, the first vector and the second vector are respectively associated with the first weight matrix by using the fully-connected layer of the neural network
And multiplying, and performing further implicit semantic expression on the first vector and the second vector to respectively obtain a first intermediate hidden layer vector and a second intermediate hidden layer vector. And respectively combining the first intermediate hidden layer vector and the second intermediate hidden layer vector with a second weight matrix through an output layer
And multiplying to obtain the corresponding score of each violation behavior category label, namely obtaining the first score of each second-layer label and the second score of each candidate label.
Based on the same inventive concept, the disclosure also provides a qualitative quantum discipline recommendation method based on the legal knowledge graph, which comprises the following steps:
in response to receiving text description of an illegal case and an episode severity label of the illegal case, performing multi-label classification on the text description by using a multi-label attention mechanism of a deep learning algorithm based on an illegal action label tree to predict an illegal action category label corresponding to the illegal case;
according to the violation behavior category label and the plot severity label, performing matching and inference calculation based on a penalty rule knowledge graph to determine penalty clause information and violation property label corresponding to the violation case and penalty measure information for the violation case;
outputting the penalty term information, the violation property label, and the penalty measure information as a result of the recommended determination for the violation case,
wherein the penalty specification knowledge-graph is constructed in advance based on a penalty specification related file for the violation, and the violation label tree is constructed in advance based on the penalty specification related file for the violation.
Further, the violation label tree has an inheritance hierarchy in which a first-layer label, a plurality of second-layer labels, and a plurality of third-layer labels are arranged in order from top to bottom.
Further, multi-label classifying the textual description using the multi-label attention mechanism to predict the violation category label comprises:
classifying the text description according to the second-layer labels by using the multi-label attention mechanism to obtain respective first scores of the second-layer labels;
sorting the plurality of second-layer labels from high to low according to the first score, and selecting the first N second-layer labels in the sorted plurality of second-layer labels, wherein N represents a preset number;
for each of the first N second-level tags, determining a third-level tag of the plurality of third-level tags that is associated with the second-level tag as a candidate tag based on the violation behavior tag tree;
classifying the text description according to the candidate tags by using the multi-tag attention mechanism to obtain respective second scores of the candidate tags;
determining the candidate label with the highest second score among the candidate labels as the violation behavior category label.
It should be noted that the method of the embodiments of the present disclosure may be executed by a single device, such as a computer or a server. The method of the embodiment can also be applied to a distributed scene and completed by the mutual cooperation of a plurality of devices. In such a distributed scenario, one of the multiple devices may only perform one or more steps of the method of the embodiments of the present disclosure, and the multiple devices interact with each other to complete the method.
It should be noted that the above describes some embodiments of the disclosure. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments described above and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.
Based on the same inventive concept, corresponding to the method of any embodiment, the disclosure also provides a qualitative quantitative era recommendation device based on the legal knowledge graph.
Referring to fig. 3, the qualitative quantum recommendation device based on legal knowledge graph comprises:
a classification module 301 configured to: in response to receiving text description of the illegal case, based on the illegal behavior label tree, performing multi-label classification on the text description by using a multi-label attention mechanism of a deep learning algorithm to predict an illegal behavior category label corresponding to the illegal case;
a matching and reasoning module 302 configured to: matching and reasoning calculation are carried out based on a penalty rule knowledge graph according to the violation behavior category label so as to determine penalty clause information and violation property label corresponding to the violation case;
an output module 303 configured to output the penalty term information and the violation property label as a recommendation qualitative result for the violation case,
wherein the penalty specification knowledge-graph is constructed in advance based on penalty specification-related documents for the violation, and the violation label tree is constructed in advance based on the penalty specification knowledge-graph.
Based on the same inventive concept, corresponding to the method of any embodiment, the disclosure also provides a qualitative quantitative era recommendation device based on the legal knowledge graph.
The qualitative quantum discipline recommendation device based on the legal knowledge graph comprises:
a classification module configured to: in response to receiving text description of an illegal case and an episode severity label of the illegal case, performing multi-label classification on the text description by using a multi-label attention mechanism of a deep learning algorithm based on an illegal behavior label tree to predict an illegal behavior category label corresponding to the illegal case;
a matching and reasoning module configured to: according to the violation behavior category label and the plot severity label, performing matching and inference calculation based on a penalty rule knowledge graph to determine penalty clause information and violation property label corresponding to the violation case and penalty measure information for the violation case;
an output module configured to output the penalty term information, the violation property label, and the penalty measure information as a result of a recommended determination for the violation case,
wherein the penalty rule knowledge-graph is constructed in advance based on a penalty rule related file for the violation, and the violation label tree is constructed in advance based on the penalty rule knowledge-graph.
For convenience of description, the above devices are described as being divided into various modules by functions, and are described separately. Of course, the functionality of the various modules may be implemented in the same one or more software and/or hardware implementations of the present disclosure.
The device of the above embodiment is used for implementing the corresponding qualitative quantum intelligence recommendation method based on the penalty rule knowledge graph in any of the foregoing embodiments, and has the beneficial effects of the corresponding method embodiment, which are not described herein again.
Based on the same inventive concept, corresponding to any of the above embodiments, the present disclosure further provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the qualitative quantum computation recommendation method based on the legal knowledge graph according to any of the above embodiments.
Fig. 4 is a schematic diagram illustrating a more specific hardware structure of an electronic device according to this embodiment, where the device may include: a processor 1010, a memory 1020, an input/output interface 1030, a communication interface 1040, and a bus 1050. Wherein the processor 1010, memory 1020, input/output interface 1030, and communication interface 1040 are communicatively coupled to each other within the device via a bus 1050.
The processor 1010 may be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an Application Specific Integrated Circuit (ASIC), or one or more Integrated circuits, and is configured to execute related programs to implement the technical solutions provided in the embodiments of the present disclosure.
The Memory 1020 may be implemented in the form of a ROM (Read Only Memory), a RAM (Random Access Memory), a static storage device, a dynamic storage device, or the like. The memory 1020 may store an operating system and other application programs, and when the technical solutions provided by the embodiments of the present specification are implemented by software or firmware, the relevant program codes are stored in the memory 1020 and called by the processor 1010 for execution.
The input/output interface 1030 is used for connecting an input/output module to input and output information. The i/o module may be configured as a component in a device (not shown) or may be external to the device to provide a corresponding function. The input devices may include a keyboard, a mouse, a touch screen, a microphone, various sensors, etc., and the output devices may include a display, a speaker, a vibrator, an indicator light, etc.
The communication interface 1040 is used for connecting a communication module (not shown in the drawings) to implement communication interaction between the present apparatus and other apparatuses. The communication module can realize communication in a wired mode (for example, USB, network cable, etc.), and can also realize communication in a wireless mode (for example, mobile network, WIFI, bluetooth, etc.).
It should be noted that although the above-mentioned device only shows the processor 1010, the memory 1020, the input/output interface 1030, the communication interface 1040 and the bus 1050, in a specific implementation, the device may also include other components necessary for normal operation. In addition, those skilled in the art will appreciate that the above-described apparatus may also include only those components necessary to implement the embodiments of the present description, and not necessarily all of the components shown in the figures.
The electronic device of the above embodiment is used to implement the qualitative quantitative record recommendation method based on the legal knowledge graph in any of the foregoing embodiments, and has the beneficial effects of the corresponding method embodiment, which are not described herein again.
Based on the same inventive concept, corresponding to any of the above-described embodiment methods, the present disclosure also provides a non-transitory computer-readable storage medium storing computer instructions for causing the computer to perform the qualitative quantum ensemble recommendation method based on a legal knowledge graph as described in any of the above embodiments.
Computer-readable media, including both permanent and non-permanent, removable and non-removable media, for storing information may be implemented in any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device.
The computer instructions stored in the storage medium of the above embodiment are used to enable the computer to execute the qualitative quantum computation recommendation method based on the legal knowledge graph according to any of the above embodiments, and have the beneficial effects of corresponding method embodiments, which are not described herein again.
Those of ordinary skill in the art will understand that: the discussion of any embodiment above is meant to be exemplary only, and is not intended to intimate that the scope of the disclosure, including the claims, is limited to these examples; within the idea of the present disclosure, also technical features in the above embodiments or in different embodiments may be combined, steps may be implemented in any order, and there are many other variations of the different aspects of the embodiments of the present disclosure as described above, which are not provided in detail for the sake of brevity.
In addition, well-known power/ground connections to Integrated Circuit (IC) chips and other components may or may not be shown in the provided figures for simplicity of illustration and discussion, and so as not to obscure the embodiments of the disclosure. Furthermore, devices may be shown in block diagram form in order to avoid obscuring embodiments of the present disclosure, and this also takes into account the fact that specifics with respect to implementation of such block diagram devices are highly dependent upon the platform within which the embodiments of the present disclosure are to be implemented (i.e., specifics should be well within purview of one skilled in the art). Where specific details (e.g., circuits) are set forth in order to describe example embodiments of the disclosure, it should be apparent to one skilled in the art that the embodiments of the disclosure can be practiced without, or with variation of, these specific details. Accordingly, the description is to be regarded as illustrative instead of restrictive.
While the present disclosure has been described in conjunction with specific embodiments thereof, many alternatives, modifications, and variations of these embodiments will be apparent to those of ordinary skill in the art in light of the foregoing description. For example, other memory architectures, such as Dynamic RAM (DRAM), may use the discussed embodiments.
The disclosed embodiments are intended to embrace all such alternatives, modifications and variances which fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalents, improvements, and the like that may be made without departing from the spirit or scope of the embodiments of the present disclosure are intended to be included within the scope of the disclosure.
Claims (4)
1. A qualitative quantum discipline recommendation method based on a legal knowledge graph comprises the following steps:
in response to receiving a text description of an illegal case, based on an illegal behavior label tree, performing multi-label classification on the text description by using a multi-label attention mechanism of a deep learning algorithm to predict an illegal behavior category label corresponding to the illegal case, wherein the method comprises the following steps of:
the violation behavior label tree is provided with an inheritance hierarchical structure, and a first-layer label, a plurality of second-layer labels and a plurality of third-layer labels are arranged in sequence from high to low in the inheritance hierarchical structure;
classifying the text description according to the second-layer labels by using the multi-label attention mechanism to obtain respective first scores of the second-layer labels;
sorting the plurality of second-layer labels from high to low according to the first score, and selecting the first N second-layer labels in the sorted plurality of second-layer labels, wherein N represents a preset number;
for each of the first N second-level tags, determining a third-level tag of the plurality of third-level tags that is associated with the second-level tag as a candidate tag based on the violation behavior tag tree;
classifying the text description according to the candidate labels by using the multi-label attention mechanism to obtain respective second scores of the candidate labels;
determining the candidate label with the highest second score of the candidate labels as the violation behavior category label; matching and reasoning calculation are carried out based on a penalty rule knowledge graph according to the violation behavior category label so as to determine penalty clause information and violation property label corresponding to the violation case;
outputting the penalty term information and the violation property label as a recommendation qualitative result for the violation case,
wherein the penalty specification knowledge-graph is constructed in advance based on the penalty specification related files for the violation, and the violation label tree is constructed in advance based on the penalty specification related files for the violation.
2. The legal knowledge graph-based qualitative quantum recommendation method of claim 1, further comprising:
after the penalty term information and the violation property label are output, in response to receiving a scenario severity label of the violation case input by a user, matching the penalty term information and the scenario severity label with the penalty rule knowledge graph to determine penalty measure information for the violation case;
and outputting the penalty measure information as a recommended penalty measure result for the violation case.
3. A qualitative quantum discipline recommendation device based on legal knowledge graphs, comprising:
a classification module configured to: in response to receiving a text description of an illegal case, based on an illegal behavior tag tree, performing multi-tag classification on the text description by using a multi-tag attention mechanism of a deep learning algorithm to predict an illegal behavior category tag corresponding to the illegal case, wherein the method comprises the following steps:
the violation behavior label tree is provided with an inheritance hierarchical structure, and a first-layer label, a plurality of second-layer labels and a plurality of third-layer labels are arranged in sequence from high to low in the inheritance hierarchical structure;
classifying the text description according to the second-layer labels by using the multi-label attention mechanism to obtain respective first scores of the second-layer labels;
sorting the second-layer labels from high to low according to the first score, and selecting the first N second-layer labels in the sorted second-layer labels, wherein N represents a preset number;
for each of the first N second-tier tags, determining a third-tier tag of the plurality of third-tier tags associated with the second-tier tag as a candidate tag based on the violation tag tree;
classifying the text description according to the candidate labels by using the multi-label attention mechanism to obtain respective second scores of the candidate labels;
determining the candidate label with the highest second score of the candidate labels as the violation behavior category label;
a matching and reasoning module configured to: matching and reasoning calculation are carried out based on a penalty rule knowledge graph according to the violation behavior category label so as to determine penalty clause information and violation property label corresponding to the violation case;
an output module configured to output the penalty term information and the violation property label as a recommendation qualitative result for the violation case,
wherein the penalty rule knowledge-graph is constructed in advance based on a penalty rule related file for the violation, and the violation label tree is constructed in advance based on the penalty rule knowledge-graph.
4. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable by the processor, the processor implementing the method of any one of claims 1 to 2 when executing the computer program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110304173.2A CN113407727B (en) | 2021-03-22 | 2021-03-22 | Qualitative measure and era recommendation method based on legal knowledge graph and related equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110304173.2A CN113407727B (en) | 2021-03-22 | 2021-03-22 | Qualitative measure and era recommendation method based on legal knowledge graph and related equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113407727A CN113407727A (en) | 2021-09-17 |
| CN113407727B true CN113407727B (en) | 2023-01-13 |
Family
ID=77677880
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110304173.2A Active CN113407727B (en) | 2021-03-22 | 2021-03-22 | Qualitative measure and era recommendation method based on legal knowledge graph and related equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113407727B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107577785A (en) * | 2017-09-15 | 2018-01-12 | 南京大学 | A kind of level multi-tag sorting technique suitable for law identification |
| CN110825879A (en) * | 2019-09-18 | 2020-02-21 | 平安科技（深圳）有限公司 | Case decision result determination method, device and equipment and computer readable storage medium |
| CN111611397A (en) * | 2019-02-26 | 2020-09-01 | 广州慧睿思通信息科技有限公司 | Information matching method and device, computer equipment and storage medium |
| US10769766B1 (en) * | 2018-05-31 | 2020-09-08 | Amazon Technologies, Inc. | Regularized multi-label classification from partially labeled training data |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109886554B (en) * | 2019-01-24 | 2024-05-03 | 平安科技（深圳）有限公司 | Illegal behavior discrimination method, device, computer equipment and storage medium |
| CN110533912B (en) * | 2019-09-16 | 2022-05-20 | 腾讯科技（深圳）有限公司 | Driving behavior detection method and device based on block chain |
-
2021
- 2021-03-22 CN CN202110304173.2A patent/CN113407727B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107577785A (en) * | 2017-09-15 | 2018-01-12 | 南京大学 | A kind of level multi-tag sorting technique suitable for law identification |
| US10769766B1 (en) * | 2018-05-31 | 2020-09-08 | Amazon Technologies, Inc. | Regularized multi-label classification from partially labeled training data |
| CN111611397A (en) * | 2019-02-26 | 2020-09-01 | 广州慧睿思通信息科技有限公司 | Information matching method and device, computer equipment and storage medium |
| CN110825879A (en) * | 2019-09-18 | 2020-02-21 | 平安科技（深圳）有限公司 | Case decision result determination method, device and equipment and computer readable storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 弱监督条件下的实体关系抽取探究;黄恺瑜;《中国优秀博硕士学位论文全文数据库(硕士)信息科技辑（月刊）》;20181015(第10期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113407727A (en) | 2021-09-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113822494B (en) | Risk prediction method, device, equipment and storage medium | |
| RU2628436C1 (en) | Classification of texts on natural language based on semantic signs | |
| CN103345645B (en) | Commodity image class prediction method towards net purchase platform | |
| WO2021208685A1 (en) | Method and apparatus for executing automatic machine learning process, and device | |
| CN110458324B (en) | Method and device for calculating risk probability and computer equipment | |
| CN112163099A (en) | Text recognition method and device based on knowledge graph, storage medium and server | |
| KR20190114166A (en) | Industrial classifying system and method using autoencoder | |
| CN112036659B (en) | Social network media information popularity prediction method based on combination strategy | |
| CN111680506A (en) | External key mapping method and device of database table, electronic equipment and storage medium | |
| CN110046251A (en) | Community content methods of risk assessment and device | |
| CN115408525A (en) | Petition text classification method, device, equipment and medium based on multi-level label | |
| US20220222924A1 (en) | Scalable pipeline for machine learning-based base-variant grouping | |
| CN111259975B (en) | Method and device for generating classifier and method and device for classifying text | |
| CN113407727B (en) | Qualitative measure and era recommendation method based on legal knowledge graph and related equipment | |
| CN113590846B (en) | Legal knowledge map construction method and related equipment | |
| CN113705201B (en) | Text-based event probability prediction evaluation algorithm, electronic device and storage medium | |
| CN115544212A (en) | Document-level event element extraction method, apparatus and medium | |
| CN116842936A (en) | Keyword recognition method, keyword recognition device, electronic equipment and computer readable storage medium | |
| CN115168609A (en) | Text matching method and device, computer equipment and storage medium | |
| Jayawickrama et al. | Seeking sinhala sentiment: Predicting facebook reactions of sinhala posts | |
| CN115017894A (en) | Public opinion risk identification method and device | |
| CN115204179A (en) | Entity relationship prediction method and device based on power grid public data model | |
| CN114428806A (en) | Case implicit knowledge mining method and device and electronic equipment | |
| CN114880635A (en) | User security level identification method, system, electronic device and medium of model integrated with lifting tree construction | |
| CN113255857A (en) | Risk detection method, device and equipment for graphic code |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |