Contradiction dispute event risk prediction method and device, equipment and storage medium
By using a conflicting dispute event risk prediction model pre-trained by a large amount of data, combined with text similarity characteristics and risk characteristics, the problem of low accuracy of conflicting dispute event risk prediction in the existing technology is solved, and higher risk prediction accuracy and timeliness are achieved.
Patent Information
- Application Number
- CN202411752071.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-02
- Publication Date
- 2025-03-21
AI Technical Summary
The risk prediction of conflict and dispute events in the prior art is low, and it is difficult to timely detect and prevent potential high-risk individuals, resulting in the occurrence of conflict and dispute events.
A target conflict and dispute incident risk prediction model that has been pre-trained with a large amount of data is used to predict risks of conflict and dispute incidents. This model calculates the similarity characteristics of the target text and the derived risk characteristics, and conducts risk prediction, which improves the accuracy of risk prediction of conflicting and dispute events.
It improves the accuracy of risk prediction of conflict and dispute incidents, and can deeply explore and analyze a large amount of data without manual participation, and promptly discover and prevent potential high-risk individuals.
Smart Images
Figure CN119671275A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of computer technology, and in particular to a method, device, equipment and storage medium for predicting the risks of conflict and dispute events. Background Art
[0002] With the rapid development of society and the increasing complexity of interpersonal relationships, various conflicts and disputes emerge in an endless stream, posing a serious threat to social stability and public security.
[0003] In related technologies, most conflict and dispute events mainly rely on simple rule engines or manual judgment to conduct risk assessment. These methods are often based on limited data and experience for analysis, and lack the ability to deeply mine and analyze large amounts of data, resulting in low risk prediction accuracy and difficulty in timely detection and prevention of potential high-risk individuals, leading to many avoidable conflict and dispute events. How to improve the accuracy of risk prediction for conflict and dispute events has become a problem that needs to be solved. Summary of the invention
[0004] In view of this, the present invention provides a method, device, equipment and storage medium for predicting the risks of conflict and dispute events to solve the problem of low risk prediction accuracy in related technologies.
[0005] In a first aspect, the present invention provides a method for predicting the risk of conflict and dispute events, the method comprising:
[0006] Obtain data on target conflict and dispute events;
[0007] The acquired data of the target conflict and dispute event is input into the pre-trained target conflict and dispute event risk prediction model to obtain the risk prediction results corresponding to the data of the target conflict and dispute event, including:
[0008] According to the target text similarity feature and the target risk feature, a feature for risk prediction is obtained; wherein the target text similarity feature indicates a text similarity feature of semantic similarity between texts in the data of the target conflict and dispute event, and the target risk feature indicates a relevant risk feature derived from the data of the target conflict and dispute event;
[0009] According to the features used for risk prediction, the risk prediction results corresponding to the data of the target conflict and dispute event are obtained.
[0010] In an optional implementation, the data of the conflict and dispute event includes the content of the conflict and dispute, information of the parties involved, the time of the dispute and the handling status;
[0011] The features used for risk prediction are obtained according to the target text similarity features and the target risk features, including:
[0012] Convert each word in the content of the conflict into a corresponding word vector;
[0013] Calculate the similarity between each word vector to obtain the target text similarity feature;
[0014] Risk derivation is performed based on the party information, dispute time and handling status to obtain the target risk characteristics.
[0015] In an optional implementation, obtaining the risk prediction result corresponding to the data of the target conflict and dispute event according to the features used for risk prediction includes:
[0016] According to the target text similarity characteristics and target risk characteristics, the probability value of the target conflict and dispute event belonging to the risk event is obtained;
[0017] The probability value is compared with a set threshold to determine whether the target conflict and dispute event belongs to the risk prediction result of a risk event.
[0018] In an optional implementation, the risk derivation based on the party information, dispute time and handling status to obtain the target risk characteristics includes:
[0019] The party information is associated with other conflict and dispute events that occurred to the party, and based on the party information and other conflict and dispute events that occurred to the party, the frequency risk characteristics of the conflict and dispute events related to the party are derived;
[0020] Based on the time of dispute, the risk characteristics of dispute urgency are derived;
[0021] Based on the handling situation, the risk characteristics of whether the overdue payment is settled are derived.
[0022] In an optional implementation, obtaining the probability value of the target conflict and dispute event being a risk event based on the target text similarity feature and the target risk feature is achieved through an XGBoost classification model.
[0023] In an optional implementation, before obtaining the data of the target conflict and dispute event, the method further includes:
[0024] Obtain data on original conflict and dispute events;
[0025] Clean and organize the conflict and dispute contents in the original conflict and dispute event data;
[0026] Generate data of target conflict and dispute events including cleaned and organized conflict and dispute contents.
[0027] In a second aspect, the present invention provides a device for predicting the risk of conflict and dispute events, the device comprising:
[0028] A data acquisition module is used to acquire data of target conflict and dispute events;
[0029] The conflict and dispute event risk prediction module is used to input the acquired target conflict and dispute event data into the pre-trained target conflict and dispute event risk prediction model to obtain the risk prediction result corresponding to the target conflict and dispute event data;
[0030] The conflict and dispute event risk prediction module includes:
[0031] A risk prediction feature acquisition unit is used to obtain features for risk prediction based on target text similarity features and target risk features; wherein the target text similarity features indicate text similarity features of semantic similarity between texts in the data of the target conflict and dispute event, and the target risk features indicate relevant risk features derived from the data of the target conflict and dispute event;
[0032] The risk prediction result determination unit is used to obtain the risk prediction result corresponding to the data of the target conflict and dispute event based on the characteristics used for risk prediction.
[0033] In an optional implementation, the risk prediction feature acquisition unit includes:
[0034] A word vector determination subunit is used to convert each word in the content of the conflict into a corresponding word vector;
[0035] The text similarity feature determination subunit is used to calculate the similarity between each word vector and obtain the target text similarity feature;
[0036] The risk characteristic determination subunit is used to derive risks based on the party information, dispute time and handling status to obtain the target risk characteristics.
[0037] In a third aspect, the present invention provides a computer device comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the method for predicting the risk of conflict and dispute events according to the first aspect or any corresponding embodiment thereof by executing the computer instructions.
[0038] In a fourth aspect, the present invention provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to enable a computer to execute the method for predicting the risk of conflict and dispute events according to the first aspect or any corresponding embodiment thereof.
[0039] The conflict and dispute event risk prediction method provided by the embodiment of the present invention uses a target conflict and dispute event risk prediction model pre-trained with a large amount of data to predict the risk of conflict and dispute events, which can deeply mine and analyze a large amount of data without manual participation, thereby improving the accuracy of risk prediction. Among them, risk prediction is performed based on the text similarity feature of the semantic similarity between texts in the data representing the target conflict and dispute event risk, and other relevant risk features derived from the target conflict and dispute event risk data, further improving the risk prediction accuracy of conflict and dispute events. BRIEF DESCRIPTION OF THE DRAWINGS
[0040] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0041] Figure 1 is a schematic flow chart of a method for predicting the risk of conflict and dispute events according to an embodiment of the present invention;
[0042] Figure 2 is a structural block diagram of a device for predicting the risk of conflict and dispute events according to an embodiment of the present invention;
[0043] Figure 3 It is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION
[0044] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.
[0045] In related technologies, most conflicts and disputes mainly rely on simple rule engines or manual judgment to conduct risk assessment. These methods are often based on limited data and experience for analysis and lack the ability to conduct in-depth mining and analysis of large amounts of data, resulting in low accuracy in risk prediction and difficulty in timely detection and prevention of potential high-risk individuals, leading to many avoidable conflicts and disputes.
[0046] In order to solve the problem that conflict and dispute events rely on simple rule engines or manual judgment to perform risk assessment, which leads to low accuracy in risk prediction of conflict and dispute events, an embodiment of the present invention provides a conflict and dispute event risk prediction method, which predicts the risk of conflict and dispute events through a conflict and dispute event risk prediction model without manual participation, and can calculate and analyze large amounts of data, thereby improving the accuracy of risk prediction of conflict and dispute events.
[0047] According to an embodiment of the present invention, an embodiment of a method for predicting the risk of conflict and dispute events is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed 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 can be executed in an order different from that shown here.
[0048] In this embodiment, a method for predicting the risk of conflict and dispute events is provided. Figure 1 FIG. 1 is a flow chart of a method for predicting the risk of conflict and dispute events according to an embodiment of the present invention. Figure 1 As shown, the process includes the following steps:
[0049] Step S101, obtaining data of target conflict and dispute events.
[0050] The data of the target conflict and dispute event may be text data corresponding to any conflict and dispute event occurring in daily life, such as the dispute content, time of occurrence of the dispute, and information of the parties involved in a civil dispute.
[0051] Step S102: input the acquired data of the target conflict and dispute event into a pre-trained risk prediction model for the target conflict and dispute event to obtain a risk prediction result corresponding to the data of the target conflict and dispute event.
[0052] The pre-training may mean that the pre-training of the target conflict and dispute event risk prediction model is completed before the first execution of step S102.
[0053] In addition, the target conflict and dispute event risk prediction model may also be pre-trained and fine-tuned. Fine-tuning may mean that the target conflict and dispute event risk prediction model is fine-tuned before the first execution of step S102.
[0054] Specifically, the above step S102 includes:
[0055] Step S1021, obtaining features for risk prediction based on target text similarity features and target risk features; wherein the target text similarity features indicate text similarity features of semantic similarity between texts in the data of target conflict and dispute events, and the target risk features indicate relevant risk features derived from the data of target conflict and dispute events.
[0056] The text similarity feature is used to measure the semantic similarity between texts in the data of the target conflict and dispute event. The text similarity feature can be calculated by cosine similarity, Euclidean distance, Manhattan distance, etc., which are not specifically limited here.
[0057] By calculating the similarity between texts in the data of the target conflict and dispute events, the embodiment of the present invention can better understand the relevance and background information of the conflict and dispute events, and better explain the risk prediction results, thereby improving the risk prediction accuracy of the target conflict and dispute event risk prediction model.
[0058] Risk features are other risk features related to conflict and dispute events derived from the data of the target conflict and dispute events, that is, new features created from the data of the target conflict and dispute events. These risk features are generated by the evolution, transformation or chain reaction of the risks, that is, the main risks, of the target conflict and dispute events. The risk feature derivation method can adopt proportional aggregation feature derivation, feature derivation based on business experience, Cartesian product derivation, decision tree and boosting tree derivation, time window statistical features, etc., which are not specifically limited here.
[0059] The embodiment of the present invention can reduce the noise in the data of target conflict and dispute events by creating other risk features related to conflict and dispute events, making the target conflict and dispute event risk prediction model more robust and improving the robustness of the model; at the same time, the risk features can capture the nonlinear relationship in the data, thereby improving the performance of the model.
[0060] Step S1022, obtaining risk prediction results corresponding to the data of the target conflict and dispute event based on the features used for risk prediction.
[0061] Among them, the risk prediction result represents whether the target conflict and dispute event is a risk event.
[0062] In summary, the conflict and dispute event risk prediction method provided by the embodiment of the present invention uses a target conflict and dispute event risk prediction model that has been pre-trained with a large amount of data to predict the risk of conflict and dispute events, which can deeply mine and analyze a large amount of data without manual participation, thereby improving the accuracy of risk prediction. Among them, risk prediction is performed based on the text similarity feature of the semantic similarity between texts in the data representing the target conflict and dispute event risk, and other relevant risk features derived from the target conflict and dispute event risk data, further improving the risk prediction accuracy of conflict and dispute events.
[0063] In some optional implementations, the data of the conflict and dispute event includes the content of the conflict and dispute, information of the parties involved, the time of the dispute, and the handling status;
[0064] Based on the target text similarity features and target risk features, the features used for risk prediction are obtained, including:
[0065] Convert each word in the content of the conflict into a corresponding word vector;
[0066] Calculate the similarity between each word vector to obtain the target text similarity feature;
[0067] Risk derivation is performed based on the party information, dispute time and handling status to obtain the target risk characteristics.
[0068] Among them, the embodiment of the present invention can convert each word in the conflict and dispute content into a corresponding word vector through a word vector model. The word vector maps each word in the conflict and dispute content to the vector space, so that semantically similar words are closer in the vector space. At this time, the distance or similarity between each word vector is calculated, and the obtained target text similarity feature can characterize the semantic similarity between the texts in the data of the target conflict and dispute event risk.
[0069] It should be noted that the word vector model can be a word2vec model, a Doc2vec model, a bag-of-words model, One-hot encoding, etc., and is not specifically limited here.
[0070] In an optional implementation, obtaining risk prediction results corresponding to data of target conflict and dispute events based on the features used for risk prediction includes:
[0071] According to the target text similarity characteristics and target risk characteristics, the probability value of the target conflict and dispute event belonging to the risk event is obtained;
[0072] The probability value is compared with a set threshold to determine whether the target conflict and dispute event belongs to the risk prediction result of a risk event.
[0073] Among them, the embodiment of the present invention can input the similarity feature and the target risk feature into the classification model to obtain the probability value of the target conflict and dispute event belonging to the risk event, and the probability value is a value between [0,1], and then compare the probability value with the set threshold to determine the risk prediction result of 0 or 1. Among them, 0 indicates that the target conflict and dispute event is not a risk event, that is, the probability value is less than the set threshold; 1 indicates that the target conflict and dispute event is a risk event, that is, the probability value is greater than the set threshold.
[0074] It should be noted that the classification model can be an XGBoost classification model, a classification decision tree model, a random forest classification model, an AdaBoost classification model, etc., and is not specifically limited here.
[0075] In some optional implementations, risk derivation is performed on the party information, dispute time and handling status to obtain target risk characteristics, including:
[0076] The party information is associated with other conflict and dispute events that occurred to the party, and based on the party information and other conflict and dispute events that occurred to the party, the frequency risk characteristics of the conflict and dispute events related to the party are derived;
[0077] Based on the time of dispute, the risk characteristics of dispute urgency are derived;
[0078] Based on the handling situation, the risk characteristics of whether the overdue payment is settled are derived.
[0079] Among them, associating the party information with other conflict and dispute events that occurred with the party can show that the party is involved in other conflict and dispute events while being involved in the current conflict and dispute event. At this time, it can be determined that the party involved in other conflict and dispute events has a higher conflict and dispute risk than the party not involved in other conflict and dispute events.
[0080] In addition, the urgency of the conflict or dispute incident can be known based on the time sequence of the dispute. At this time, the more urgent conflict or dispute incident can be identified as a high-risk incident.
[0081] According to the processing situation, we can know the processing progress of the current conflict and dispute event, whether the current conflict and dispute event can be completed within the processing period, whether the current conflict and dispute event is in an overdue processing state, etc. At this time, the conflict and dispute event that is overdue for processing can be identified as a high-risk event.
[0082] The embodiment of the present invention uses the frequency of other conflicts and disputes involving the parties, the urgency of the disputes, and whether they are resolved overdue as risk features, so that the risk prediction model of the target conflict and dispute events is more robust, and can extract information that is helpful for risk prediction from the risk features, thereby improving the performance of the model.
[0083] In some optional implementations, obtaining the probability value of the target conflict and dispute event being a risk event is achieved through an XGBoost classification model based on the target text similarity feature and the target risk feature.
[0084] In some optional implementations, before obtaining the data of the target conflict and dispute event, the conflict and dispute event risk prediction method further includes:
[0085] Obtain data on original conflict and dispute events;
[0086] Clean and organize the conflict and dispute contents in the original conflict and dispute event data;
[0087] Generate data of target conflict and dispute events including cleaned and organized conflict and dispute contents.
[0088] Among them, the method for cleaning and arranging text can be to remove repeated and useless words and sentences in the dispute content. The embodiment of the present invention does not specifically limit the cleaning and arranging method, and technical personnel in this field can choose according to actual conditions.
[0089] Exemplarily, the pre-training process of the target conflict and dispute event risk prediction model is described below:
[0090] Step a, obtaining a sample data set; the sample data set includes text data of multiple conflict and dispute events and preset label information; the preset label information includes: a label value representing whether the conflict and dispute event is a risk event; the text data of each conflict and dispute event includes: the content of the conflict and dispute, information of the parties involved, the time of the dispute and the handling status, etc.
[0091] Among them, the method for obtaining the sample data set can be obtained from the client platform docking, or it can be obtained from the search engine through a crawler. The embodiment of the present invention does not make specific settings for the sample data set and the method for obtaining the sample data set, and technical personnel in this field can set them according to actual conditions.
[0092] Step b: Clean and organize the conflict and dispute content in the sample data set to improve the risk prediction accuracy of the target conflict and dispute event risk prediction model.
[0093] Step c, inputting the cleaned and sorted conflict and dispute content into the word2vec model for training, and obtaining the word vector corresponding to each word in the conflict and dispute content;
[0094] In step d, the text cosine similarity is calculated between the word vectors corresponding to each word in the content of the conflict and dispute to obtain the text similarity feature; then, based on the text similarity feature, the optimal parameters of the word2vec model are evaluated. When the text similarity feature reaches the expected threshold, the word2vec model training is completed, and then the next step is entered.
[0095] Step e, deriving risk characteristics from the party information, dispute time and handling status in the sample data set to obtain risk characteristics;
[0096] Step f, putting the text similarity features, the derived risk features and the preset label information into the XGBoost classification model to obtain the risk prediction probability value of each sample;
[0097] Step g, calculate the precision and recall rate according to the probability value, and continuously tune the parameters of the XGBoost classification model. When the precision and recall rate reach the expected level, determine the optimal parameters of the XGBoost classification model, and then obtain the final trained target conflict and dispute event risk prediction model.
[0098] In this embodiment, a device for predicting the risk of conflict and dispute events is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and will not be repeated hereafter. As used below, the term "module" can implement a combination of software and / or hardware of a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.
[0099] The embodiment of the present invention provides a device for predicting the risk of conflict and dispute events, such as Figure 2 As shown, the conflict and dispute event risk prediction device includes:
[0100] The data acquisition module 201 is used to acquire data of the target conflict and dispute event;
[0101] The conflict and dispute event risk prediction module 202 is used to input the acquired target conflict and dispute event data into a pre-trained target conflict and dispute event risk prediction model to obtain a risk prediction result corresponding to the target conflict and dispute event data;
[0102] The conflict and dispute event risk prediction module 202 includes:
[0103] The risk prediction feature acquisition unit 221 is used to obtain features for risk prediction based on the target text similarity feature and the target risk feature; wherein the target text similarity feature indicates the text similarity feature of the semantic similarity between the texts in the data of the target conflict and dispute event, and the target risk feature indicates the relevant risk feature derived from the data of the target conflict and dispute event;
[0104] The risk prediction result determination unit 2022 is used to obtain the risk prediction result corresponding to the data of the target conflict and dispute event according to the features used for risk prediction.
[0105] In some optional implementations, the risk prediction feature acquisition unit 2021 includes:
[0106] A word vector determination subunit is used to convert each word in the content of the conflict into a corresponding word vector;
[0107] The text similarity feature determination subunit is used to calculate the similarity between each word vector and obtain the target text similarity feature;
[0108] The risk characteristic determination subunit is used to derive risks based on the party information, dispute time and handling status to obtain the target risk characteristics.
[0109] The further functional description of each of the above modules and units is the same as that of the above corresponding embodiments and will not be repeated here.
[0110] The conflict and dispute event risk prediction device in this embodiment is presented in the form of a functional unit, where the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.
[0111] The embodiment of the present invention also provides a computer device having the above Figure 2 The risk prediction device for conflict and dispute events shown.
[0112] like Figure 3 As shown, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Various components are connected to each other using different buses for communication, and can be installed on a common mainboard or installed in other ways as needed. The processor can process the instructions executed in the computer device, including instructions stored in or on the memory to display the graphical information of the GUI on an external input / output device (such as a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system).
[0113] The processor 10 may be a central processing unit, a network processor or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.
[0114] The memory 20 stores instructions executable by at least one processor 10, so that the at least one processor 10 executes the method shown in the above embodiment.
[0115] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely arranged relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.
[0116] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid state drive; the memory 20 may also include a combination of the above types of memory.
[0117] The computer device also includes an input device 30 and an output device 40. The processor 10, the memory 20, the input device 30 and the output device 40 may be connected via a bus or other means. Figure 3 The example of connecting through bus is taken in the following.
[0118] The input device 30 can receive input digital or character information, and generate key signal input related to the user settings and function control of the computer device, such as a touch screen, a keypad, a mouse, a track pad, a touch pad, an indicator bar, one or more mouse buttons, a trackball, a joystick, etc. The output device 40 may include a display device, an auxiliary lighting device (e.g., an LED) and a tactile feedback device (e.g., a vibration motor), etc. The above-mentioned display device includes but is not limited to a liquid crystal display, a light emitting diode, a display and a plasma display. In some optional embodiments, the display device can be a touch screen.
[0119] The embodiment of the present invention also provides a computer-readable storage medium. The method according to the embodiment of the present invention can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium through a network download, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state hard disk, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor, or hardware, the method shown in the above embodiment is implemented.
[0120] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations are all within the scope defined by the appended claims.
Claims
1. A method for predicting the risk of conflict and dispute events, characterized in that: The method comprises: Obtain data on target conflict and dispute events; The acquired data of the target conflict and dispute event is input into the pre-trained target conflict and dispute event risk prediction model to obtain the risk prediction results corresponding to the data of the target conflict and dispute event, including: According to the target text similarity feature and the target risk feature, a feature for risk prediction is obtained; wherein the target text similarity feature indicates a text similarity feature of semantic similarity between texts in the data of the target conflict and dispute event, and the target risk feature indicates a relevant risk feature derived from the data of the target conflict and dispute event; According to the features used for risk prediction, the risk prediction results corresponding to the data of the target conflict and dispute event are obtained.
2. The method according to claim 1, characterized in that The data of the conflict and dispute incidents include the content of the conflict and dispute, information of the parties involved, the time of the dispute and the situation of its handling; The features used for risk prediction are obtained according to the target text similarity features and the target risk features, including: Convert each word in the content of the conflict into a corresponding word vector; Calculate the similarity between each word vector to obtain the target text similarity feature; Risk derivation is performed based on the party information, dispute time and handling status to obtain the target risk characteristics.
3. The method according to claim 2, characterized in that The risk prediction result corresponding to the data of the target conflict and dispute event is obtained according to the features used for risk prediction, including: According to the target text similarity characteristics and target risk characteristics, the probability value of the target conflict and dispute event belonging to the risk event is obtained; The probability value is compared with a set threshold to determine whether the target conflict and dispute event belongs to the risk prediction result of a risk event.
4. The method according to claim 2, characterized in that: The risk derivation based on the information of the parties, the time of the dispute and the handling situation is used to obtain the target risk characteristics, including: The party information is associated with other conflict and dispute events that occurred to the party, and based on the party information and other conflict and dispute events that occurred to the party, the frequency risk characteristics of the conflict and dispute events related to the party are derived; Based on the time of dispute, the risk characteristics of dispute urgency are derived; Based on the handling situation, the risk characteristics of whether the overdue payment is settled are derived.
5. The method according to claim 3, characterized in that: The method of obtaining the probability value of the target conflict and dispute event being a risk event based on the target text similarity feature and the target risk feature is implemented through the XGBoost classification model.
6. The method according to any one of claims 1 to 5, characterized in that Before acquiring the data of the target conflict and dispute event, the method further includes: Obtain data on original conflict and dispute events; Clean and organize the conflict and dispute contents in the original conflict and dispute event data; Generate data of target conflict and dispute events including cleaned and organized conflict and dispute contents.
7. A conflict and dispute event risk prediction device, characterized in that: The device comprises: A data acquisition module is used to acquire data of target conflict and dispute events; The conflict and dispute event risk prediction module is used to input the acquired target conflict and dispute event data into the pre-trained target conflict and dispute event risk prediction model to obtain the risk prediction result corresponding to the target conflict and dispute event data; The conflict and dispute event risk prediction module includes: A risk prediction feature acquisition unit is used to obtain features for risk prediction based on target text similarity features and target risk features; wherein the target text similarity features indicate text similarity features of semantic similarity between texts in the data of the target conflict and dispute event, and the target risk features indicate relevant risk features derived from the data of the target conflict and dispute event; The risk prediction result determination unit is used to obtain the risk prediction result corresponding to the data of the target conflict and dispute event based on the characteristics used for risk prediction.
8. The device according to claim 7, characterized in that The risk prediction feature acquisition unit comprises: A word vector determination subunit is used to convert each word in the content of the conflict into a corresponding word vector; The text similarity feature determination subunit is used to calculate the similarity between each word vector and obtain the target text similarity feature; The risk characteristic determination subunit is used to derive risks based on the party information, dispute time and handling status to obtain the target risk characteristics.
9. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the method according to any one of claims 1 to 6 by executing the computer instructions.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the method according to any one of claims 1 to 6.