A case allocation and disposal method, device, equipment and medium

Abstract

The application provides a case allocation handling method, device, equipment and medium, the method comprises the following steps: using a LangChain framework and a large language model to construct an intelligent acceptance system, and training the large language model of the intelligent acceptance system by using collected case data; collecting case data in real time through a data acquisition device, the case data includes case descriptions, and uploading the collected case data to the intelligent acceptance system; the intelligent acceptance system automatically accepts the received case data, uses the trained large language model to automatically classify the accepted cases, and automatically allocates the accepted cases to the handling terminal used by the handler in combination with a rule engine. The application has the advantages that the processing efficiency and accuracy of urban governance cases can be greatly improved, the complexity of manual operation is reduced, and more convenient and personalized urban governance service experiences are provided for users.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, in particular to a case allocation and disposal method, device, equipment and medium. BACKGROUND

[0002] In the process of urban governance, the collection, acceptance, allocation, assessment and evaluation of cases are often involved in the disposal process. However, in the existing urban governance process, the disposal process of cases has the following defects: the existing case management system often needs manual classification and allocation of cases, which is tedious and prone to errors, has low case processing efficiency, and lacks personalized services, and cannot provide customized assistance content according to the characteristics and requirements of different cases. SUMMARY

[0003] The technical problem to be solved by the present application is to provide a case allocation and disposal method, device, equipment and medium, which solves the problem that the existing case management system needs manual classification and allocation of cases, which is tedious and prone to errors, has low case processing efficiency, and lacks personalized services, and cannot provide customized assistance content according to the characteristics and requirements of different cases.

[0004] In a first aspect, the present application provides a case allocation and disposal method, which comprises:

[0005] An intelligent acceptance system is constructed using a LangChain framework and a large language model, and the large language model of the intelligent acceptance system is trained using collected case data;

[0006] Case data is collected in real time by a data collection device, and the case data includes case descriptions, and the collected case data is uploaded to the intelligent acceptance system;

[0007] The intelligent acceptance system automatically accepts the received case data, uses the trained large language model to automatically classify the accepted cases, and automatically allocates the accepted cases to the disposal terminal used by the disposer in combination with a rule engine.

[0008] Further, the method further comprises: connecting the intelligent acceptance system with other information systems, so as to realize collaborative disposal and information sharing between the intelligent acceptance system and the other information systems.

[0009] Further, the training of the large language model of the intelligent acceptance system using the collected case data is specifically:

[0010] The case data is collected in advance, and the collected case data is labeled, specifically including: assigning a correct classification and allocation result to each case according to the case description in the case data, and after labeling all cases, dividing the labeled case data into a training data set and a test data set;

[0011] The large language model is used for text processing and feature extraction on the case description in the case data to generate a feature vector of the case; a suitable machine learning algorithm or deep learning algorithm is selected, and the large language model is trained using a training data set; after the training is completed, the trained large language model is evaluated using a test data set.

[0012] Further, the intelligent receiving system automatically receives the case data, specifically:

[0013] After the intelligent receiving system receives the case data, it determines whether the case is valid according to the case description in the case data. If the case is invalid, it is not accepted, and the received case data is returned to the corresponding data collection device. If the case is valid, it is accepted, and Embedding is used for text similarity matching of the accepted case, and cases with a text similarity exceeding a preset similarity value are merged and processed.

[0014] Further, the large language model includes a ChatGLM model and a Bert model.

[0015] Further, the accepted case is automatically classified using the trained large language model, and the accepted case is automatically distributed to the disposal terminal used by the disposer in combination with a rule engine, specifically including:

[0016] After the case is accepted, the Bert model is used for text processing on the case description in the case data to obtain text information; the trained ChatGLM model is then used for semantic analysis and feature extraction on the text information to convert the text information into a feature vector, and a machine learning algorithm or a deep learning algorithm is used to automatically classify the accepted case according to the feature vector, while the accepted case is automatically distributed to the disposal terminal used by the disposer in combination with the case distribution rules set in the rule engine, so that the disposer can dispose of the distributed case;

[0017] At the same time, the intelligent receiving system automatically tracks the distributed case and unifies the examination and evaluation of the distribution and disposal results of the case.

[0018] In a second aspect, the present application provides a case distribution and disposal device, which includes a system construction module, a data collection module and a distribution and disposal module.

[0019] The system construction module is used to construct an intelligent receiving system using a LangChain framework and a large language model, and to train the large language model of the intelligent receiving system using collected case data.

[0020] The data collection module is configured to collect case data in real time through a data collection device, wherein the case data includes a case description, and upload the collected case data to the intelligent acceptance system.

[0021] The distribution processing module is configured to automatically accept the received case data by the intelligent acceptance system, automatically classify the accepted cases by using a trained large language model, and automatically distribute the accepted cases to a processing terminal used by a processor in combination with a rule engine.

[0022] Further, the device further comprises a system interfacing module configured to interface the intelligent acceptance system with other information systems, so as to realize collaborative processing and information sharing between the intelligent acceptance system and the other information systems.

[0023] Further, the training of the large language model of the intelligent acceptance system by using the collected case data specifically comprises:

[0024] Pre-collecting case data, and labeling the collected case data, specifically including: assigning a correct classification and distribution result to each case according to the case description in the case data, and dividing the labeled case data into a training data set and a test data set after labeling all the cases.

[0025] Text processing and feature extraction are performed on the case description in the case data by using the large language model to generate a feature vector of the case, a suitable machine learning algorithm or deep learning algorithm is selected, the training data set is used to train the large language model, and after the training is completed, the test data set is used to evaluate the trained large language model.

[0026] Further, the automatic acceptance of the received case data by the intelligent acceptance system specifically comprises:

[0027] After the intelligent acceptance system receives the case data, it determines whether the case is valid according to the case description in the case data, if the case is invalid, it is not accepted, and the received case data is returned to the corresponding data collection device; if the case is valid, the case is accepted, and Embedding is used to perform text similarity matching on the accepted case, and cases with a text similarity exceeding a preset similarity value are merged.

[0028] Further, the large language model comprises a ChatGLM model and a Bert model.

[0029] Further, the automatic classification of the accepted cases by using the trained large language model, and the automatic distribution of the accepted cases to the processing terminal used by the processor in combination with the rule engine specifically comprises:

[0030] After the case is accepted, the case description in the case data is processed by using the Bert model to obtain text information; then the ChatGLM model trained is used to perform semantic analysis and feature extraction on the text information, so as to convert the text information into a feature vector, and use a machine learning algorithm or a deep learning algorithm to automatically classify the accepted case according to the feature vector, and at the same time, the accepted case is automatically distributed to the handling terminal used by the handler according to the case distribution rule set in the rule engine, so as to be handled by the handler.

[0031] At the same time, the intelligent acceptance system automatically tracks the distributed cases and unifies the examination and evaluation of the distribution and handling results of the cases.

[0032] In a third aspect, the present application provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to realize the method of the first aspect.

[0033] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, wherein the program is executed by a processor to realize the method of the first aspect.

[0034] The one or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0035] 1. By adopting the LangChain framework and the ChatGLM model to construct the intelligent acceptance system, the constructed intelligent acceptance system is applied to the urban governance cases, and the ChatGLM model of the intelligent acceptance system is trained by using the collected case data, so that the intelligent acceptance system obtained after training can automatically accept, classify and distribute the urban governance cases, thereby greatly improving the processing efficiency and accuracy of the urban governance cases, reducing the tediousness of manual operation, and providing users with more convenient and personalized urban governance service experience.

[0036] 2. Since the LangChain framework is a framework that allows developers to connect language models to other data sources and interact with their environment; by using the LangChain framework and the ChatGLM model to construct the intelligent acceptance system, the intelligent acceptance system can be well connected with other information systems, so that the intelligent acceptance system can interact with other information systems to realize collaborative handling and information sharing, thereby improving the processing efficiency of urban governance cases, and providing customized assistance content according to the characteristics and requirements of different cases.

[0037] 3. After the intelligent acceptance system receives the case data, it first judges the effectiveness of the case according to the case description, then performs text similarity matching on the accepted cases, and merges the cases with a text similarity exceeding a preset similarity value, thereby realizing the early filtering of invalid cases and repeated cases, reducing the process time occupied by invalid cases and repeated cases, and helping to improve the processing efficiency of cases.

[0038] The above description is only a summary of the technical solutions of the present application. In order to better understand the technical means of the present application, the content of the specification can be implemented, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the following specific embodiments of the present application are described. BRIEF DESCRIPTION OF DRAWINGS

[0039] The present application will be further described below with reference to the accompanying drawings and embodiments.

[0040] Figure 1 An execution flowchart of a case allocation and handling method in the first embodiment of the present application;

[0041] Figure 2 A structural schematic diagram of a case allocation and handling device in the second embodiment of the present application;

[0042] Figure 3 A structural schematic diagram of an electronic device in the third embodiment of the present application;

[0043] Figure 4 A structural schematic diagram of a medium in the fourth embodiment of the present application. DETAILED DESCRIPTION

[0044] In order to better understand the technical solutions of the present application, the technical solutions of the present application will be described in detail below in combination with the drawings and specific embodiments of the specification.

[0045] Before that, some terms involved in the technical solutions of the present application will be described:

[0046] LangChain framework: LangChain is a framework for developing applications driven by language models, allowing developers to connect language models to other data sources and interact with their environment; LangChain as an LLM application framework supports calling multiple different models and provides a relatively unified and convenient operation interface, allowing models to be plug-and-play.

[0047] ChatGLM model: ChatGLM is a dialogue robot based on the hundred-billion base model GLM-130B. It continuously pre-trains text and code based on GLM-130B and aligns human intentions through supervised fine-tuning and other technologies, and has the ability of copywriting, information extraction, role-playing, question and answer, and dialogue.

[0048] Bert model: BERT stands for Bidirectional Encoder Representation from Transformers. It is a pre-trained language representation model. It emphasizes that it is no longer pre-trained in the traditional way of using a single-directional language model or shallowly splicing two single-directional language models. Instead, it uses a new masked language model (MLM) to generate deep bidirectional language representations.

[0049] NLP technology: NLP technology is a branch of language information processing and one of the core topics of artificial intelligence (AI).

[0050] Embodiment one

[0051] The embodiment provides a case allocation and disposal method, which is specifically used for intelligent acceptance, classification and allocation and disposal of urban governance cases, such as Figure 1 As shown in the figure, the method comprises the following steps:

[0052] Step 1: Use the LangChain framework and large language model to build an intelligent acceptance system, train the large language model of the intelligent acceptance system with collected case data, so that the intelligent acceptance system can use the trained large language model to automatically accept, classify and allocate cases; in the present application, the large language model comprises a ChatGLM model and a Bert model;

[0053] Step 2: Collect case data in real time through a data acquisition device, which can be a mobile phone, various sensors, a computer, etc. The case data includes case descriptions, and of course, the case data also includes case types, times, places and other key information. Upload the collected case data to the intelligent acceptance system to process the case data using the intelligent acceptance system;

[0054] Step 3: The intelligent acceptance system automatically accepts the received case data, uses the trained large language model to automatically classify the accepted cases, and automatically allocates the accepted cases to the disposal terminal used by the disposer in combination with the rule engine, so that the disposer can dispose the allocated cases.

[0055] The application adopts the combination of the LangChain framework and the ChatGLM model to construct an intelligent receiving system, applies the constructed intelligent receiving system to city governance cases, and trains the ChatGLM model of the intelligent receiving system by using collected case data, so that the intelligent receiving system obtained after training can automatically receive, classify and allocate and dispose city governance cases, thereby greatly improving the processing efficiency and accuracy of city governance cases, reducing the tediousness of manual operation, and providing users with more convenient and personalized city governance service experience.

[0056] In some embodiments of the application, the method further comprises:

[0057] Step 10, the intelligent receiving system is connected with other information systems, so that the intelligent receiving system and other information systems realize collaborative disposal and information sharing; step 10 needs to be executed after the intelligent receiving system is constructed in step 1. Since the LangChain framework is a framework that allows developers to connect language models to other data sources and interact with their environment, the intelligent receiving system constructed by the application can be well connected with other information systems, so that the intelligent receiving system can interact with other information systems to realize collaborative disposal and information sharing.

[0058] When constructing the intelligent receiving system, the ChatGLM model and the LangChain framework can be integrated into a unified city governance platform, and serve as an intelligent allocation and disposal module of the unified city governance platform. At the same time, the intelligent allocation and disposal module is connected with other information systems, so as to realize collaborative disposal and information sharing with other information systems. At the same time, the ChatGLM model and the LangChain framework can also be deployed by using existing computing resources and cloud platforms, and the API interface provided by the same is used to realize real-time calling of city governance cases in receiving, classification and allocation and disposal.

[0059] The collaborative disposal and information sharing between the intelligent receiving system and other information systems of the application include but are not limited to the following contents:

[0060] A, the intelligent receiving system is connected with the legal aid information system, the case data reported by the user is matched with the corresponding legal matters in the legal aid information system, so as to determine the applicable legal provisions and assistance content, and the matched legal provisions and assistance content are automatically sent to the user, so as to provide legal assistance for the user;

[0061] B, the intelligent receiving system is connected with other city management systems, the resources of each other city management system are coordinated and allocated according to the nature and emergency degree of the received case, the sharing, cooperation and command among different systems are realized, so that the processing efficiency of city management cases is improved, and meanwhile, security measures such as data encryption, permission management and the like are taken for data transmission between the intelligent receiving system and other city management systems, so that the safety and privacy of data are ensured.

[0062] In the application, the training of the large language model of the intelligent receiving system by the collected case data is specifically:

[0063] The case data is collected in advance, and the collected case data includes key information such as case description, case type, time and place, the collected case data is labeled for subsequent supervised learning of the model, specifically including: according to the case description in the case data, the correct classification and distribution result is specified for each case, which can be classified and distributed according to the case type, the case emergency degree, the geographical position and the like, and after all the cases are labeled, the labeled case data is divided into a training data set and a test data set; in the specific implementation of the application, after the case data is collected, the case data needs to be cleaned and pretreated, which can be processed by filtering, filling missing values, standardization and the like to ensure the accuracy and consistency of the data;

[0064] The case description in the case data is processed and feature extracted by the large language model to generate a feature vector of the case, a suitable machine learning algorithm or deep learning algorithm is selected, and the large language model is trained by using the training data set; after the training is completed, the trained large language model is evaluated by using the test data set, including evaluating the accuracy of the model in case classification and distribution and other aspects of performance. In the specific implementation of the application, the machine learning algorithm or deep learning algorithm used for model training can be selected according to actual needs, such as CNN convolutional neural network, RNN recurrent neural network and the like for deep learning algorithm, and decision tree, random forest, support vector machine (SVM) and the like for machine learning algorithm.

[0065] In the application, the automatic receiving of the intelligent receiving system for the received case data is specifically:

[0066] The intelligent acceptance system can adopt NLP technology to judge whether the case is valid according to the case description in the case data after receiving the case data. If the case is invalid, the case data received is not accepted and is returned to the corresponding data collection device, so that the collector can know that the uploaded case data is an invalid case. If the case is valid, the case is accepted, and Embedding is adopted to perform text similarity matching on the accepted case. The cases with a text similarity exceeding a preset similarity value (the preset similarity value can be set according to actual needs) are processed by merging, for example, when the text similarity of two cases reaches 95%, it can be judged that the two cases belong to the same case, and the two cases can be processed by merging at this time. According to the present application, after receiving the case data, the validity of the case is first judged according to the case description, and then the accepted case is subjected to text similarity matching, and the cases with a text similarity exceeding a preset similarity value are processed by merging, so that invalid cases and repeated cases can be filtered out in advance, which can reduce the process time of invalid cases and repeated cases in subsequent processes, and help to improve the processing efficiency of cases.

[0067] More specifically, the use of the trained large language model to automatically classify the accepted case, and the automatic allocation of the accepted case to the handling terminal used by the handler by combining the rule engine specifically includes:

[0068] After the case is accepted, the Bert model is used to perform text processing on the case description in the case data, such as removing stop words, word segmentation, etc., so as to obtain text information; then the trained ChatGLM model is used to perform semantic analysis and feature extraction on the text information, so as to convert the text information into a feature vector, and use a machine learning algorithm or a deep learning algorithm to automatically classify the accepted case according to the feature vector, and at the same time, combine the case allocation rules set in the rule engine to automatically allocate the accepted case to the handling terminal used by the handler. The handling terminal can be a computer, a mobile phone, etc., so that the handler can handle the allocated case. In the specific implementation of the present application, the case allocation rules can be set in the rule engine in advance, and the specific case allocation rules can be set as needed, for example, according to the priority level of the set case allocation, the case with high level is preferentially allocated to the handler, for example, according to the fixed allocation time set, the accepted case is automatically allocated to the handler after the time arrives, etc. At the same time, after receiving the allocated case, the handler can take appropriate handling measures according to the characteristics and requirements of the case, such as dispatching police forces, collecting evidence, coordinating relevant departments, etc.

[0069] Meanwhile, the intelligent receiving system automatically tracks the distributed cases to master the disposal process and disposal result of the cases in real time, and can transmit the classification, distribution and disposal result to relevant departments or personnel; the distribution and disposal result of the cases are uniformly examined and evaluated, and specific examination and evaluation standards can be formulated according to actual use needs, and supervision and handling are carried out.

[0070] In the present application, when the case data is collected in real time by the data collection device, the collection of case data can be realized by daily inspection, operation monitoring and the like.

[0071] In summary, by adopting the technical scheme of the present application, at least the following beneficial effects are achieved:

[0072] 1. By adopting the combination of LangChain framework and ChatGLM model to construct an intelligent receiving system, applying the constructed intelligent receiving system to urban governance cases, and training the ChatGLM model of the intelligent receiving system using collected case data, the trained intelligent receiving system can automatically receive, classify and distribute and dispose urban governance cases, thereby greatly improving the processing efficiency and accuracy of urban governance cases, reducing the tediousness of manual operation, and providing users with more convenient and personalized urban governance service experience.

[0073] 2. Since the LangChain framework is a framework that allows developers to connect language models to other data sources and interact with their environment; by using the LangChain framework and ChatGLM model to construct an intelligent receiving system, the intelligent receiving system can be well connected with other information systems, so that the intelligent receiving system can interact with other information systems to realize collaborative disposal and information sharing, thereby improving the processing efficiency of urban governance cases, and being able to provide customized assistance content according to the characteristics and requirements of different cases.

[0074] 3. After the intelligent receiving system receives case data, it first judges the effectiveness of the case according to the case description, then performs text similarity matching on the received cases, and merges and processes the cases with a text similarity exceeding a preset similarity value, thereby realizing the early filtering of invalid cases and duplicate cases, which can reduce the occupation of subsequent process processing time by invalid cases and duplicate cases, and help to improve the processing efficiency of cases.

[0075] Based on the same inventive concept, the present application also provides a device corresponding to the method in embodiment one, which is described in detail in embodiment two.

[0076] Embodiment two

[0077] In the present embodiment, a case distribution and disposal device is provided, which is specifically used for intelligent receiving, classifying and distributing and disposing urban governance cases, such asFigure 2 As shown, the device comprises a system construction module, a data collection module and a distribution processing module;

[0078] The system construction module is used for constructing an intelligent receiving system using a LangChain framework and a large language model, training the large language model of the intelligent receiving system using collected case data, so that the intelligent receiving system can automatically receive, classify and distribute cases using the trained large language model; in the present application, the large language model comprises a ChatGLM model and a Bert model;

[0079] The data collection module is used for collecting case data in real time through a data collection device, which can be a mobile phone, various sensors, a computer, etc.; the case data includes case descriptions, and also includes case types, times, places and other key information; the collected case data is uploaded to the intelligent receiving system for processing by the intelligent receiving system;

[0080] The distribution processing module is used for automatically receiving case data by the intelligent receiving system, automatically classifying the received cases using the trained large language model, and automatically distributing the received cases to the handling terminal used by the handler in combination with a rule engine, so that the handler can handle the distributed cases.

[0081] The present application constructs an intelligent receiving system by combining a LangChain framework and a ChatGLM model, applies the constructed intelligent receiving system to city governance cases, and trains the ChatGLM model of the intelligent receiving system using collected case data, so that the trained intelligent receiving system can automatically receive, classify and distribute city governance cases, thereby greatly improving the processing efficiency and accuracy of city governance cases, reducing the tediousness of manual operation, and providing users with more convenient and personalized city governance service experience.

[0082] In some embodiments of the present application, the device further comprises a system interfacing module;

[0083] The system docking module is used for docking the intelligent receiving system with other information systems, so that the intelligent receiving system and the other information systems realize collaborative treatment and information sharing; the system docking module needs to be executed after the system construction module constructs the intelligent receiving system. Since the intelligent receiving system is constructed by using the LangChain framework and the large language model, and the LangChain framework is a framework that allows developers to connect language models to other data sources and interact with their environment; therefore, the intelligent receiving system constructed by the application can be well docked with other information systems, so that the intelligent receiving system can interact with other information systems to realize collaborative treatment and information sharing.

[0084] In the specific construction of the intelligent receiving system, the ChatGLM model and the LangChain framework can be integrated into a unified urban governance platform, and serve as an intelligent distribution and treatment module of the unified urban governance platform, and the intelligent distribution and treatment module is used to dock with other information systems, so as to realize collaborative treatment and information sharing with other information systems. Meanwhile, the ChatGLM model and the LangChain framework can be deployed by using existing computing resources and cloud platforms, and the API interface provided by the ChatGLM model and the LangChain framework is used to realize real-time calling of urban governance cases in the process of receiving, classifying and distributing and treating.

[0085] The collaborative treatment and information sharing between the intelligent receiving system and other information systems include but are not limited to the following contents:

[0086] A. The intelligent receiving system is docked with the legal aid information system, the case data reported by the user is matched with the corresponding legal matters in the legal aid information system, so as to determine the applicable legal provisions and assistance content, and the matched legal provisions and assistance content are automatically sent to the user, so as to provide legal assistance for the user;

[0087] B. The intelligent receiving system is docked with other urban governance systems, the resources of each other urban governance system are allocated according to the nature and emergency degree of the received cases, so as to realize the sharing, cooperation and command among different systems, thereby improving the processing efficiency of urban governance cases; meanwhile, security measures such as data encryption and permission management need to be taken for data transmission between the intelligent receiving system and other urban governance systems, so as to ensure the security and privacy of data.

[0088] In the application, the training of the large language model of the intelligent receiving system by using the collected case data is specifically:

[0089] The case data is collected in advance, and the collected case data includes key information such as case description, case type, time, and location. The collected case data is labeled to facilitate subsequent supervised learning of the model, specifically including: according to the case description in the case data, assigning a correct classification and distribution result to each case, which can be classified and distributed according to the case type, case urgency, geographical location, etc., and after labeling all cases, the labeled case data is divided into a training data set and a test data set; in the specific implementation of the present application, after collecting the case data, the case data needs to be cleaned and preprocessed, which can be processed by filtering, filling missing values, standardization, etc. to ensure the accuracy and consistency of the data;

[0090] The case description in the case data is processed and feature extracted by using a large language model to generate a feature vector of the case; a suitable machine learning algorithm or deep learning algorithm is selected, and the large language model is trained using the training data set; after training, the trained large language model is evaluated using the test data set, including evaluating the accuracy of the model in case classification and distribution, and other aspects of performance. In the specific implementation of the present application, the machine learning algorithm or deep learning algorithm used for model training can be selected according to actual needs, such as CNN convolutional neural network, RNN recurrent neural network, etc. for deep learning algorithm, and decision tree, random forest, support vector machine (SVM), etc. for machine learning algorithm.

[0091] In the present application, the intelligent receiving system automatically receives the received case data, specifically:

[0092] After receiving the case data, the intelligent receiving system can use NLP technology to determine whether the case is valid according to the case description in the case data. If the case is invalid, it will not be accepted, and the received case data will be returned to the corresponding data collection device, so that the collector can know that the uploaded case data is an invalid case. If the case is valid, the case is accepted, and Embedding is used to perform text similarity matching on the accepted case. Cases with a text similarity exceeding a preset similarity value (which can be set according to actual needs) are merged and processed. For example, when the text similarity of two cases reaches 95%, it can be determined that the two cases belong to the same case, and the two cases can be merged and processed at this time. The present application determines the validity of the case according to the case description after receiving the case data, then performs text similarity matching on the accepted case, and merges and processes the cases with a text similarity exceeding a preset similarity value, thereby filtering out invalid cases and duplicate cases in advance, which can reduce the processing time of invalid cases and duplicate cases in subsequent processes, and helps to improve the processing efficiency of cases.

[0093] More specifically, the use of the trained large language model to automatically classify the accepted case, and combined with the rule engine, the accepted case is automatically allocated to the handling terminal used by the handler, specifically comprising:

[0094] After the case is accepted, the Bert model is used to process the case description in the case data, such as removing stop words, word segmentation, etc., to obtain text information; then the trained ChatGLM model is used to perform semantic analysis and feature extraction on the text information, so as to convert the text information into a feature vector, and use a machine learning algorithm or a deep learning algorithm to automatically classify the accepted case according to the feature vector, and at the same time, combined with the case allocation rules set in the rule engine, the accepted case is automatically allocated to the handling terminal used by the handler, which can be a computer, a mobile phone, etc. for the handler to handle the allocated case; in the specific implementation of the present application, the case allocation rules can be set in the rule engine in advance, and the specific case allocation rules can be set as needed, for example, according to the priority level of the case allocation, the case with high level is preferentially allocated to the handler, for example, according to the fixed allocation time set, after the arrival time, the accepted case is automatically allocated to the handler, etc.; at the same time, after receiving the allocated case, the handler can take appropriate handling measures according to the characteristics and requirements of the case, such as dispatching police forces, collecting evidence, coordinating relevant departments, etc.

[0095] At the same time, the intelligent acceptance system automatically tracks the allocated case to real-time master the handling process and handling result of the case, and can pass the classification, allocation and handling result to the relevant department or personnel; the allocation and handling result of the case are unified evaluated and assessed, and the specific evaluation and assessment standard can be formulated according to the actual use needs, and supervision and management are carried out.

[0096] In the present application, when the case data is collected in real time by the data collection device, the case data can be collected by daily inspection, operation monitoring, etc.

[0097] Based on the same inventive concept, the present application provides an electronic device embodiment corresponding to embodiment one, which is described in detail in embodiment three.

[0098] Embodiment three

[0099] The present embodiment provides an electronic device, as shown in Figure 3 The present embodiment provides an electronic device, as shown in

[0100] Since the electronic device described in this embodiment is the device used to implement the method in Embodiment 1 of this application, those skilled in the art can understand the specific implementation method and various variations of the electronic device in this embodiment based on the method described in Embodiment 1 of this application. Therefore, how the electronic device implements the method in the embodiment of this application will not be described in detail here. Any device used by those skilled in the art to implement the method in the embodiment of this application falls within the scope of protection of this application.

[0101] Based on the same inventive concept, this application provides a storage medium corresponding to Embodiment 1, as detailed in Embodiment 4.

[0102] Example 4

[0103] This embodiment provides a computer-readable storage medium, such as... Figure 4 As shown, a computer program is stored thereon, which, when executed by a processor, can implement any of the embodiments in Example 1.

[0104] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, apparatus, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0105] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus, and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0106] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0107] These computer program instructions can also be loaded into computer or other programmable data processing devices, so that a series of operation steps are performed on the computer or other programmable data processing devices to generate computer-implemented processes, thus the instructions executed on the computer or other programmable data processing devices provide processes for implementing the flow Figure 1 one or more flows and / or blocks Figure 1 one or more blocks or steps of functions specified in the flow

[0108] Although the specific embodiments of the present application are described above, it should be understood by those skilled in the art that the specific embodiments described are merely illustrative, and not intended to limit the scope of the present application, and equivalent modifications and variations made in accordance with the spirit of the present application by those skilled in the art should be covered within the scope of the claims of the present application.

Claims

1. A method for case allocation and handling, characterized in that: The method comprises: An intelligent receiving system is constructed using a LangChain framework and a large language model, and the large language model of the intelligent receiving system is trained using collected case data; Case data is collected in real time by a data collection device, and the case data includes case descriptions; and the collected case data is uploaded to the intelligent receiving system; The intelligent receiving system automatically receives the received case data, specifically, after receiving the case data, the intelligent receiving system determines whether the case is valid according to the case description in the case data; if the case is invalid, the received case data is not received and is returned to the corresponding data collection device; if the case is valid, the case is received, and the received case is subjected to text similarity matching using Embedding; the received case is automatically classified using a trained large language model, and the received case is automatically distributed to a disposal terminal used by a disposer in combination with a rule engine; Further comprising: connecting the intelligent receiving system with other information systems to realize collaborative disposal and information sharing between the intelligent receiving system and the other information systems, specifically comprising: connecting the intelligent receiving system with a legal aid information system, matching the case data reported by a user with corresponding legal matters in the legal aid information system, and automatically sending the matched legal provisions and assistance content to the user; connecting the intelligent receiving system with other city governance systems, and coordinating and allocating resources of the other city governance systems according to the nature and urgency of the received case.

2. The case distribution and handling method according to claim 1, wherein: The training of the large language model of the intelligent receiving system using the collected case data specifically comprises: Pre-collecting case data, and labeling the collected case data, specifically comprising: assigning a correct classification and distribution result to each case according to the case description in the case data, and dividing the labeled case data into a training data set and a test data set after labeling all the cases; Using a large language model to perform text processing and feature extraction on the case description in the case data to generate a feature vector of the case; selecting a suitable machine learning algorithm, and training the large language model using the training data set; after training, evaluating the trained large language model using the test data set.

3. The method of claim 1, wherein: The large language model comprises a ChatGLM model and a Bert model.

4. The method of claim 3, wherein: The automatic classification of the received case using the trained large language model, and the automatic distribution of the received case to a disposal terminal used by a disposer in combination with a rule engine specifically comprises: After the case is received, the case description in the case data is subjected to text processing using the Bert model to obtain text information; the text information is subjected to semantic analysis and feature extraction using the trained ChatGLM model to convert the text information into a feature vector, and the received case is automatically classified using a machine learning algorithm according to the feature vector, and the received case is automatically distributed to a disposal terminal used by a disposer in combination with a case distribution rule set in the rule engine, so that the disposer disposes the distributed case; Meanwhile, the intelligent receiving system automatically tracks the distributed cases and unifies the evaluation of the distribution and disposal results of the cases.

5. A case distribution handling apparatus characterized by comprising: The device comprises a system construction module, a data collection module and a distribution and disposal module. The system construction module is configured to use a LangChain framework and a large language model to construct an intelligent receiving system, and train the large language model of the intelligent receiving system using collected case data. The data collection module is configured to collect case data in real time through a data collection device, and upload the collected case data to the intelligent receiving system. The distribution and disposal module is configured to automatically receive the case data by the intelligent receiving system, specifically, after receiving the case data, the intelligent receiving system determines whether the case is valid according to the case description in the case data, if the case is invalid, the case data is not received and is returned to the corresponding data collection device. If the case is valid, the case is received and processed, and the received case is matched for text similarity using Embedding, and cases with a text similarity exceeding a preset similarity value are merged; the received case is automatically classified using a trained large language model, and the received case is automatically distributed to a disposal terminal used by a disposer in combination with a rule engine. The device further comprises a system interfacing module configured to interface the intelligent receiving system with other information systems, so as to realize collaborative disposal and information sharing between the intelligent receiving system and the other information systems, specifically, the intelligent receiving system is interfaced with a legal aid information system, the case data reported by a user is matched with corresponding legal matters in the legal aid information system, and the matched legal provisions and assistance contents are automatically sent to the user; the intelligent receiving system is interfaced with other city governance systems, and resources of the other city governance systems are coordinated and allocated according to the nature and urgency of the received cases.

6. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the program to realize the method of any one of claims 1 to 4.

7. A computer readable storage medium having stored thereon a computer program, characterized in that The program is executed by the processor to realize the method of any one of claims 1 to 4. The program is executed by the processor to realize the method of any one of claims 1 to 4.

Images