A case diversion method and system based on situation concept analysis

Abstract

The present invention provides a case diversion method and system based on situation concept analysis. The method includes: Step 1: Using natural language processing technology to process existing cases and their corresponding diversion results, and extracting criminal acts and diversion results in the existing cases; Step 2: Processing the criminal acts and diversion results in the existing cases into a formal context, and using the situation concept analysis method to process the formal context to generate object concepts, where the extension of the object concept is the case and the intension is the criminal act; Step 3: Using natural language processing technology to process the case to be diverted, obtaining the criminal act of the case to be diverted, and matching the criminal act in the case to be diverted with the object concept to obtain the diversion result of the case to be diverted.

Description

Technical Field

[0001] The present invention relates to the technical field of data mining, and in particular to a case diversion method and system based on formal concept analysis. Background Art

[0002] Nowadays, the number of court case judgment documents has increased significantly, while the number of relevant practitioners is not sufficient to handle a large number of cases in a timely manner. Therefore, how to quickly screen out key information from cases and classify cases in combination with relevant laws and regulations so that professional practitioners can handle them better and faster is one of the current research directions.

[0003] Most of the existing case diversion technologies are based on manual diversion. However, manual diversion has the disadvantages of low efficiency and high cost. At the same time, due to the complexity of this work, only professional personnel who have been trained for a long time can perform accurate diversion relatively quickly. However, with the development of society, the incidence of civil cases is also rising continuously. With the existing manual methods, cases often cannot be diverted in a timely manner. Therefore, a fast and accurate case diversion technology is needed. Summary of the Invention

[0004] Aiming at the problems of low efficiency and high cost existing in the existing method of using manual case diversion, the method of the present invention introduces formal concept analysis into case diversion, and provides a case diversion method and system based on formal concept analysis, which can classify unprocessed cases quickly and accurately, reduce costs and improve efficiency.

[0005] On the one hand, the present invention provides a case diversion method based on formal concept analysis, including:

[0006] Step 1: Using natural language processing technology to process existing cases and their corresponding diversion results, and extracting criminal acts and diversion results in the existing cases;

[0007] Step 2: Processing the criminal acts and diversion results in the existing cases into a formal context, and using the formal concept analysis method to process the formal context to generate object concepts, where the extension of the object concept is the case and the intension is the criminal act;

[0008] Step 3: Using natural language processing technology to process the cases to be diverted, obtaining the criminal acts of the cases to be diverted, and matching the criminal acts in the cases to be diverted with the object concepts to obtain the diversion results of the cases to be diverted.

[0009] Further, in step 2, the processing of the criminal acts and diversion results in the existing cases into a formal context specifically includes:

[0010] Step A1: Take the criminal acts and diversion results as attributes in the formal context, and take each case as an object in the formal context.

[0011] Step A2: For each object, if the object contains the corresponding attribute, mark the intersection cell of the object and the attribute as 1, otherwise mark it as 0. Do this until all objects are marked, obtaining a table that only contains 0, 1, objects, and attributes. This table is the formal context.

[0012] Furthermore, in Step 2, the method of formal concept analysis is used to process the formal context to generate object concepts, which specifically includes:

[0013] Step B1: For each crime name, calculate its object concept (g(f(A)), f(A)); where A represents the set of objects corresponding to the diversion of the criminal act, f(A) represents the set of common attributes contained among all objects in the object set A, and g(f(A)) represents the set of common objects corresponding to all attributes in the attribute set f(A).

[0014] Step B2: According to Step B1, calculate the object concepts of all crime names, and record and save all the object concepts.

[0015] On the other hand, the present invention provides a case diversion system based on formal concept analysis, including:

[0016] A text processing module, which is used to process existing cases and their corresponding diversion results by using natural language processing technology, and extract criminal acts and diversion results in the existing cases.

[0017] A model construction module, which is used to process criminal acts and diversion results in existing cases into a formal context, and use the method of formal concept analysis to process the formal context to generate object concepts. The extension of the object concept is the case, and the intension is the criminal act.

[0018] A case diversion module, which is used to process the case to be diverted by using natural language processing technology to obtain the criminal act of the case to be diverted, and match the criminal act in the case to be diverted with the object concept to obtain the diversion result of the case to be diverted.

[0019] Furthermore, the model construction module includes an attribute and object extraction unit and a formal context generation unit.

[0020] The attribute and object extraction unit is used to take criminal acts and diversion results as attributes in the formal context, and take each case as an object in the formal context.

[0021] The formal context generation unit is used to, for each object, when the object contains the corresponding attribute, mark the cross-cell of the object and the attribute as 1, otherwise mark it as 0. This is done until all objects are marked, resulting in a table that only contains 0, 1, objects, and attributes. This table is used as the formal context.

[0022] Further, the model construction module further includes an object concept calculation unit and an object concept storage unit;

[0023] The object concept calculation unit is used to calculate the object concept (g(f(A)), f(A)) for each crime name; where A represents the set of objects corresponding to the diversion of the criminal act, f(A) represents the set of common attributes contained among all objects in the object set A, and g(f(A)) represents the set of common objects corresponding to all attributes in the attribute set f(A);

[0024] The object concept storage unit is used to record and store the object concepts of all crime names.

[0025] Advantages of the present invention:

[0026] The present invention uses formal concept analysis to process existing classification results to generate object concepts, and uses the generated object concepts to divert unprocessed cases, improving the accuracy and processing speed, greatly reducing the manpower required for previous case diversion, and improving work efficiency. Description of the Drawings

[0027] Figure 1 It is a schematic flowchart of a case diversion method based on formal concept analysis provided by an embodiment of the present invention;

[0028] Figure 2 It is a schematic structural diagram of a case diversion system based on formal concept analysis provided by an embodiment of the present invention. Detailed Embodiments

[0029] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0030] Formal Concept Analysis (FCA) is a tool for data description and knowledge discovery. The present invention creatively applies Formal Concept Analysis to the case diversion technology, which can better find the common points of similar cases from the existing case information, so as to classify the unclassified cases better, faster and more accurately. At the same time, the reasoning rules based on Formal Concept Analysis can also better improve the accuracy of its classification.

[0031] Embodiment 1

[0032] As Figure 1 shown, an embodiment of the present invention provides a case diversion method based on Formal Concept Analysis, including the following steps:

[0033] S101: Use natural language processing technology to process the existing cases and their corresponding diversion results, and extract the criminal acts and diversion results in the existing cases;

[0034] Specifically, based on the machine learning method, a supervised learning method can be used to extract the criminal acts and diversion results in the existing cases. It should be noted that after extraction, the texts of the obtained criminal acts and diversion results should be screened to remove duplicate criminal acts and diversion results;

[0035] S102: Process the criminal acts and diversion results in the existing cases into a formal context, and use the Formal Concept Analysis method to process the formal context to generate object concepts. The extension of the object concept is the case, and the intension is the criminal act;

[0036] As an implementable manner, the processing of the criminal acts and diversion results in the existing cases into a formal context specifically includes:

[0037] Step A1: Take the criminal acts and diversion results as attributes in the formal context, and take each case as an object in the formal context;

[0038] Step A2: For each object, if the object contains the corresponding attribute, mark the cross cell of the object and the attribute as 1, otherwise mark it as 0 until all objects are marked, and a table containing only 0, 1, objects, and attributes is obtained. This table is the formal context.

[0039] As an implementable manner, the use of the Formal Concept Analysis method to process the formal context to generate object concepts specifically includes:

[0040] Step B1: For each charge, calculate its object concept (g(f(A)), f(A)); where A represents the set of objects corresponding to the diversion of the criminal act, f(A) represents the set of common attributes among all objects in the object set A, and g(f(A)) represents the set of common objects corresponding to all attributes in the attribute set f(A).

[0041] Specifically, f() and g() are two operators in formal concept analysis. f() represents the common attribute operator. For example, if C represents any object set, f(C) represents the set of common attributes among all objects in the object set C; g() represents the common object operator. For example, if D represents any attribute set, g(D) represents the set of common objects corresponding to all attributes in the attribute set D. For example, if the formal context is as shown in Table 1, and A = {1, 2}, then f(A) = {a, b}, and g(f(A)) = {1, 2, 3}.

[0042] Table 1

[0043]

[0044] Step B2: Calculate the object concepts of all charges according to Step B1, and record and save all the object concepts.

[0045] S103: Use natural language processing technology to process the case to be diverted, obtain the criminal act of the case to be diverted, and match the criminal act in the case to be diverted with the object concept to obtain the diversion result of the case to be diverted.

[0046] Specifically, match the criminal act in the case to be diverted with all the object concepts, and consider that the cases included in the successfully matched object concepts belong to the same type of cases as the case to be diverted, and the diversion result corresponding to the successfully matched object concept is the diversion result of the case to be diverted.

[0047] It should be noted that the same case may correspond to multiple diversion results.

[0048] The present invention proposes a case diversion method based on formal concept analysis. Based on natural language processing and formal concept analysis (FCA) technology, it can quickly and accurately divert cases, thereby assisting relevant practitioners to quickly and accurately divert cases according to the case situation.

[0049] Embodiment 2

[0050] Corresponding to the above case diversion method based on formal concept analysis, the embodiment of the present invention also provides a case diversion system based on formal concept analysis, including: a text processing module, a model construction module, and a case diversion module.

[0051] Among them, the text processing module is used to process the existing cases and their corresponding diversion results by using natural language processing technology, and extract the criminal acts and diversion results in the existing cases. The model construction module is used to process the criminal acts and diversion results in the existing cases into a formal context, and use the formal concept analysis method to process the formal context to generate object concepts. The extension of the object concept is the case, and the intension is the criminal act. The case diversion module is used to process the case to be diverted by using natural language processing technology, obtain the criminal act of the case to be diverted, match the criminal act in the case to be diverted with the object concept, and obtain the diversion result of the case to be diverted.

[0052] As an implementable manner, the model construction module includes an attribute and object extraction unit and a formal context generation unit; among them, the attribute and object extraction unit is used to take the criminal act and diversion result as attributes in the formal context, and each case as an object in the formal context. The formal context generation unit is used to, for each object, when the object contains the corresponding attribute, mark the cross cell of the object and the attribute as 1, otherwise mark it as 0, until all objects are marked, to obtain a table containing only 0, 1, objects, and attributes, and take this table as the formal context.

[0053] As an implementable manner, the model construction module further includes an object concept calculation unit and an object concept storage unit; among them, the object concept calculation unit is used to calculate its object concept (g(f(A)), f(A)) for each crime name; where A represents the set of objects corresponding to the diversion of the criminal act, f(A) represents the set of common attributes contained among all objects in the object set A, and g(f(A)) represents the set of common objects corresponding to all attributes in the attribute set f(A). The object concept storage unit is used to record and store the object concepts of all crime names.

[0054] The present invention uses formal concept analysis to process the existing classification results to generate object concepts, and uses the generated object concepts to divert and process the unprocessed cases, improving the accuracy and processing speed, greatly reducing the manpower required for the previous case diversion, and improving the work efficiency.

[0055] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A case diversion method based on situation concept analysis, characterized in that Including: Step 1: Use natural language processing technology to process existing cases and their corresponding diversion results, and extract the criminal acts and diversion results in the existing cases; Step 2: Process the criminal acts and diversion results in the existing cases into a formal context, and use the formal concept analysis method to process the formal context to generate object concepts. The extension of the object concept is the case, and the intension is the criminal act; The process of processing the criminal acts and diversion results in the existing cases into a formal context specifically includes: Step A1: Take the criminal acts and diversion results as attributes in the formal context, and take each case as an object in the formal context; Step A2: For each object, if the object contains the corresponding attribute, mark the intersection cell of the object and the attribute as 1, otherwise mark it as 0. Until all objects are marked, a table containing only 0, 1, objects, and attributes is obtained, and this table is the formal context; The process of using the formal concept analysis method to process the formal context to generate object concepts specifically includes: Step B1: For each crime name, calculate its object concept (g(f(A)), f(A)); where A represents the set of objects corresponding to the diversion of the criminal act, f(A) represents the set of common attributes contained among all objects in the object set A, and g(f(A)) represents the set of common objects corresponding to all attributes in the attribute set f(A); Step B2: According to Step B1, calculate the object concepts of all crime names, record and save all object concepts; Step 3: Use natural language processing technology to process the case to be diverted, obtain the criminal acts of the case to be diverted, match the criminal acts in the case to be diverted with the object concepts, and obtain the diversion result of the case to be diverted.

2. A case diversion system based on situation concept analysis, characterized in that Including: A text processing module for using natural language processing technology to process existing cases and their corresponding diversion results, and extract the criminal acts and diversion results in the existing cases; A model construction module for processing the criminal acts and diversion results in the existing cases into a formal context, and using the formal concept analysis method to process the formal context to generate object concepts. The extension of the object concept is the case, and the intension is the criminal act; A case diversion module for using natural language processing technology to process the case to be diverted, obtain the criminal acts of the case to be diverted, match the criminal acts in the case to be diverted with the object concepts, and obtain the diversion result of the case to be diverted; The model construction module includes an attribute and object extraction unit, a formal context generation unit, an object concept calculation unit, and an object concept storage unit; The attribute and object extraction unit is used to take the criminal acts and diversion results as attributes in the formal context, and take each case as an object in the formal context; The formal context generation unit is configured to, for each object, when the object contains the corresponding attribute, mark the cross-cell of the object and the attribute as 1, otherwise mark it as 0. After marking all objects, a table containing only 0, 1, objects, and attributes is obtained, and this table is used as the formal context; The object concept calculation unit is configured to calculate the object concept (g(f(A)), f(A)) for each crime name; where A represents the set of objects corresponding to the diversion of the criminal act, f(A) represents the set of common attributes contained among all objects in the object set A, and g(f(A)) represents the set of common objects corresponding to all attributes in the attribute set f(A); The object concept storage unit is configured to record and store the object concepts of all crime names.

Images