Intelligent identification system and method applied to legal instruments

Abstract

The invention discloses an intelligent recognition system and method applied to legal instruments, and relates to the technical field of intelligent recognition of instruments, and the method comprises the steps: obtaining two-element-form filling data filled by a customer, carrying out the preprocessing and normalization of the filling data, obtaining the filling features of two-element-form tasks, calling the successful processing of the two-element-form tasks, and carrying out the recognition of the two-element-form tasks. Establishing a service personnel database according to the characteristics of successfully processing the element type two-state tasks, selecting adaptive service personnel, selecting the selected service personnel of the new element type two-state tasks according to the busy case number of the service personnel, and comprehensively considering the adaptation degree of the blank filling characteristics; and selecting the service personnel with the highest adaptation degree as the selected service personnel of the new element type two-state task, and if the client is not satisfied with the feedback of the new element type two-state task, analyzing the selected service personnel again to serve the client. And the subsequent matching better meets the core requirements of customers.

Description

Technical Field

[0001] This invention relates to the field of intelligent document recognition technology, specifically to an intelligent recognition system and method for legal documents. Background Technology

[0002] In the field of legal document processing, there is a lack of unified standards for the completion of element-based complaints and answers, as well as subsequent service coordination. Under the traditional model, while document formats are highly flexible, standards vary, leading to issues such as omissions, vague expressions, or formatting problems when clients fill them out, resulting in significant difficulties in subsequent information processing. Simultaneously, the raw data submitted by clients lacks a systematic preprocessing workflow; the information is fragmented and varied in format, requiring substantial manpower for screening, proofreading, and integration. This not only prolongs the task initiation cycle but also increases the risk of errors due to manual operation, impacting overall service efficiency. Furthermore, the significant differences in filling habits among different clients further increase the difficulty of data standardization, hindering the efficient advancement of legal document services. Currently, the allocation of service personnel for element-based legal document tasks relies heavily on manual judgment or simple experience matching, lacking a scientific and comprehensive personnel competency assessment system. The existing model does not fully integrate the characteristics of service personnel's past successful cases, failing to accurately identify their areas of expertise and service types, resulting in a low degree of mismatch between personnel allocation and task requirements. In some cases, service personnel with the corresponding experience are not prioritized, while personnel with incompatible experience are prone to service deviations, affecting the professionalism and accuracy of document processing. Meanwhile, the lack of sufficient consideration of workload differences during personnel allocation may lead to some staff being overwhelmed with tasks while others remain idle, resulting in a waste of human resources and making it difficult to guarantee the stability of service quality. The existing service system lacks a dynamic optimization mechanism and an effective customer feedback loop. When faced with new types of element-based legal document tasks, the traditional matching model struggles to quickly adapt to the new task characteristics, continuing to use fixed allocation logic, leading to delayed service response. Furthermore, customer feedback on service results has not been effectively transformed into optimization criteria, and problems encountered during the service process cannot be promptly reported to the personnel matching stage, causing similar problems to recur. Simultaneously, the lack of weighting and dynamic adjustment for different task characteristics makes it impossible to accurately focus on core customer needs. In the long run, the service model is prone to rigidity, making it difficult to adapt to diverse and complex legal document processing needs, and hindering the continuous improvement of customer satisfaction. Summary of the Invention

[0003] The purpose of this invention is to provide an intelligent recognition system and method for legal documents to solve the problems mentioned in the background art.

[0004] To address the aforementioned technical problems, this invention provides the following technical solution: an intelligent recognition method for legal documents, comprising the following steps: S1. Define the format for the two-dimensional element-based form, obtain the fill-in data of the two-dimensional element-based form entered by the customer, preprocess and normalize the fill-in data to obtain the fill-in characteristics of the two-dimensional element-based form task. S2. Analyze the service personnel, call the successful processing of element-based two-state tasks, and establish a service personnel database based on the characteristics of successfully processing element-based two-state tasks. S3. When a new element-based two-state condition appears, establish a preset task for the new element-based two-state condition, analyze the adaptability of service personnel to the preset task for the new element-based two-state condition, and select suitable service personnel. S4. When there are suitable service personnel for the new element-based two-state task, select the selected service personnel for the new element-based two-state task based on the number of cases the service personnel are busy with. S5. When there are no suitable service personnel for the new element-based two-state task, the service personnel with the highest degree of suitability shall be selected as the selected service personnel for the new element-based two-state task, taking into account the suitability of the fill-in-the-blank features. S6. After the service is completed, if the customer is not satisfied with the feedback on the new element-based two-state task, re-analyze and select service personnel to serve the customer.

[0005] Furthermore, in step S1, a format is defined for the element-based two-statement, which includes an element-based complaint and an element-based answer. The element-based two-statement is defined as a fixed table format with fixed fill-in-the-blank items, providing customers with fill-in-the-blank item options. These options are selected by the customer to fill in the blanks, thereby obtaining the fill-in-the-blank data within the element-based two-statement item options. The fill-in-the-blank data is preprocessed and normalized, resulting in N fill-in-the-blank features for the element-based two-statement task. By standardizing the fixed table format of the element-based two-statement, clarifying the fill-in-the-blank items and providing corresponding optional content, the difficulty of filling out the form for customers is greatly reduced, avoiding rework problems caused by inconsistent formats, ambiguous expressions, or missing content, making the filling process smoother and more efficient. Meanwhile, the collected fill-in-the-blank data is preprocessed and normalized to effectively remove invalid information, standardize data format, and achieve data standardization. This reduces the manpower input and error risk of subsequent information processing, and also extracts accurate fill-in-the-blank features, providing reliable and unified data support for the subsequent service personnel matching process, and helping to promote the efficient advancement of the overall service process.

[0006] Furthermore, in step S2, the service personnel handling the X element-based two-state tasks are analyzed. For the x-th service personnel (x=1,2,…,X), the number of element-based two-state tasks successfully handled by the x-th service personnel in historical data is Y. The successful handling of an element-based two-state task by the x-th service personnel indicates that after the x-th service personnel provided service for the element-based two-state task, the customer reported satisfactory results. Thus, the characteristics {A} of the y-th successfully handled element-based two-state task are obtained. y_1A y_2 ,…,A y_n ,…,A y_N}, where A y_n Let y represent the nth fill-in-the-blank feature of the y-th successfully processed element-based two-state task. Substituting each of these features into y=1,2,…,Y yields the features of Y successfully processed element-based two-state tasks. A service personnel database is then established based on these features. By focusing on customer satisfaction cases of service personnel, task features are extracted and used to construct the database, effectively integrating the historical success experiences of service personnel. This database clearly presents the task types and core competencies that each service personnel excels in, avoiding the drawbacks of traditional personnel competency assessments that rely on subjective judgment or fragmented experience. The success cases upon which the database is based are centered on customer feedback, ensuring the authenticity and effectiveness of the information. This provides solid data support for the accurate matching of subsequent tasks and personnel, making personnel competency assessments more objective and targeted, and contributing to improving the scientific rigor and reliability of overall service matching.

[0007] Furthermore, in step S3, when a new element-type two-state condition appears, a preset task for the new element-type two-state condition is established, and the adaptability of service personnel to the preset task for the new element-type two-state condition is analyzed. The fill-in-the-blank characteristics of the preset task for the new element-type two-state condition are {A1, A2, ..., A...}. n ,…,A N}, where A n Let n be the fill-in-the-blank feature of the new element-based two-state preset task. For the x-th service personnel, call the service personnel database and analyze the matching degree B between the y-th successfully processed element-based two-state task and the new element-based two-state preset task. y : ; Where k n This represents the preset weight of the nth fill-in-the-blank feature, which is then used to determine the matching relationship between the yth successfully processed element-based two-state task and the new element-based two-state preset task. When B y When >C, it is determined that the y-th successfully processed element-type two-state task matches the new element-type two-state preset task, and the y-th successfully processed element-type two-state task is a matching element-type two-state task of the new element-type two-state preset task; otherwise, it is determined that the y-th successfully processed element-type two-state task does not match the new element-type two-state preset task, and the y-th successfully processed element-type two-state task is not a matching element-type two-state task of the new element-type two-state preset task. Substituting y = 1, 2, ..., Y into each element, the matching relationship between Y successfully processed element-based two-state tasks and the new element-based two-state preset task is determined. This yields the Y0 matching element-based two-state task for the x-th service personnel. If Y0 is not 0, the x-th service personnel is listed as a suitable service personnel; otherwise, the x-th service personnel is listed as a candidate service personnel. Substituting x = 1, 2, ..., X into each element, Q suitable service personnel for the new element-based two-state preset task are obtained, and these suitable service personnel are added to the suitable service personnel directory for the new element-based two-state preset task. By analyzing the matching relationship between the new element-based two-state preset task and the service personnel's historical successful cases, suitable service personnel with corresponding service experience are accurately selected. This process relies on the characteristics of real successful cases for matching, eliminating the drawbacks of subjective experience judgment and ensuring that the selected suitable personnel can meet the fill-in-the-blank characteristics requirements of the new task. Clearly identifying highly suitable personnel as suitable candidates not only provides a precise personnel reserve for new task allocation but also maximizes the professional advantages of service personnel, effectively reducing service deviations caused by mismatched capabilities. This improves the accuracy of task-person matching from the source and lays a solid foundation for the stable guarantee of subsequent service quality.

[0008] Furthermore, in step S4, when a user issues a service request, the preset task for the new element-based two-state condition is transformed into a new element-based two-state task. The number of suitable service personnel is analyzed. When Q is not 0, it is determined that there are suitable service personnel for the new element-based two-state task. The Q-th suitable service personnel are analyzed, and the creation time of the new element-based two-state task is retrieved. The number of busy cases for the q-th suitable service personnel is D. y The number of busy cases refers to the number of element-based two-state tasks that the assigned service personnel have not completed within a preset period starting from the establishment time. Uncompleted element-based two-state tasks represent element-based two-state tasks within a preset period starting from the establishment time, where q = 1, 2, ..., Q. Substituting q = 1, 2, ..., Q into each of these values ​​yields the number of busy cases {E1, E2, ..., E} for Q assigned service personnel. q ,…,E Q}, where E qThis represents the number of busy cases for the q-th assigned service personnel. The service personnel with the fewest busy cases are then selected as the designated service personnel for the new element-based two-state task. By combining the busy case count of assigned service personnel for task allocation, a dynamic balance of human resources is effectively achieved. This mechanism avoids a decline in service quality due to task backlog, ensuring that each person undertaking a task can provide service under a reasonable workload, guaranteeing both service efficiency and maintaining stable service quality. Simultaneously, prioritizing personnel with the fewest busy cases ensures timely response and processing of new tasks, reducing customer wait times and improving service timeliness and customer satisfaction. This allocation method also fully leverages the professional value of assigned personnel, avoiding resource idleness or oversaturation, optimizing the overall operational efficiency of the service system, and achieving efficient task and personnel matching.

[0009] Furthermore, in step S5, when Q is 0, it is determined that there are no suitable service personnel for the new element-based two-state task. The X service personnel are analyzed. For the x-th service personnel, in the Y successfully processed element-based two-state tasks, the nth fill-in-the-blank feature of the Y successfully processed element-based two-state tasks is {A}. 1_n A 2_n ,…,A y_n ,…,A Y_n For the y-th successfully processed element-based two-state task, the degree of difference between the nth fill-in-the-blank feature of the y-th successfully processed element-based two-state task and the nth fill-in-the-blank feature of the new element-based two-state task is F. n_y F n_y =|(A y_n -A n ) / A y_n |, when F n_y <F y When the nth blank feature of the yth successfully processed element-based two-state task is similar to the nth blank feature of the new element-based two-state task, it is determined that the nth blank feature of the yth successfully processed element-based two-state task is not similar to the nth blank feature of the new element-based two-state task. This leads to the number of matching features G for the nth blank feature of the xth service personnel. n_x The number of adaptable features represents the number of fill-in-the-blank features in the nth fill-in-the-blank feature of the new element-based two-state task that are similar to the nth fill-in-the-blank feature of the new element-based two-state task. Substituting these into x=1,2,…,X, we obtain the number of adaptable features {G} for the nth fill-in-the-blank feature of X service personnel. n_1 G n_2 ,…,G n_x ,…,G n_X}, and thus obtain the nth fill-in-the-blank feature fit score H for the xth service personnel. n_x : ; Substitute each of the following into the given values ​​n=1,2,…,N to obtain the N fill-in-the-blank feature matching scores for the x-th service personnel, and then obtain the comprehensive score J for the x-th service personnel. x J x The sum of the fitting scores for the N fill-in-the-blank features of the x-th service personnel is substituted into x=1,2,…,X to obtain the comprehensive score of the X service personnel. The service personnel with the highest comprehensive score are selected as the designated service personnel for the new element-based two-state task. When there are no suitable service personnel, this mechanism quantifies the degree of personnel suitability by analyzing the similarity between the service personnel's historical success cases and the fill-in-the-blank features of the new task, avoiding the bias of subjective judgment. It compares and calculates the fitting score and comprehensive score one by one from the fill-in-the-blank feature dimension, providing an objective and precise basis for personnel selection, ensuring that even in the absence of direct experience matching, the service personnel most suitable for the new task can be selected. This method effectively fills the matching gap when there are vacancies in suitable personnel, ensures the scientific nature of task allocation and the stability of service quality, maximizes the satisfaction of customer needs, and improves the overall suitability of services and customer satisfaction.

[0010] Furthermore, in step S6, after the service of the new element-based two-state task is completed, if the customer is dissatisfied with the feedback of the new element-based two-state task, the fill-in-the-blank feature that the customer is dissatisfied with is invoked. If the nth fill-in-the-blank feature is the fill-in-the-blank feature that the customer is dissatisfied with, the weight of the nth fill-in-the-blank feature is replaced with α*k. n α is a preset weight amplification coefficient. If α > 1, the selected service personnel will be analyzed again to serve the customer. This constructs a dynamic optimization closed loop for service feedback, accurately transforming customer dissatisfaction feedback into the basis for adjusting the weights of fill-in-the-blank features. By amplifying the weights of corresponding features, subsequent service personnel analysis focuses more on the core concerns of customers, effectively avoiding the recurrence of similar service deviations and improving the targeting and repair efficiency of services. Simultaneously, this iterative mechanism based on real feedback allows the service system to continuously adapt to changes in customer needs, constantly improving the matching logic, ensuring the stability of service quality and the improvement of customer satisfaction in the long term, forming a virtuous cycle of "service-feedback-optimization-reservice".

[0011] An intelligent recognition system for legal documents includes: a fill-in-the-blank data feature extraction module, a service personnel database construction module, a new task adaptation module, a busyness filtering and selection module, a feature adaptation highest selection module, and a feedback secondary analysis module. The fill-in data feature extraction module is used to obtain the fill-in data of the element-type two-form according to the specified format, and to preprocess and normalize the fill-in data to obtain N fill-in features of the element-type two-form task. The service personnel database construction module is used to analyze service personnel, call successfully processed element-based two-state tasks, and build a service personnel database based on the characteristics of successfully processed element-based two-state tasks. The new task adaptation module is used to create a new element-based two-state preset task when a new element-based two-state appears, analyze the degree of adaptation of service personnel to the new element-based two-state preset task, and select the appropriate service personnel. The busyness filtering and selection module is used to select the appropriate service personnel for the new element-based two-state task based on the number of busy cases of the service personnel when there are suitable service personnel for the new element-based two-state task. The feature adaptation highest selection module is used to select the service personnel with the highest degree of adaptation as the selected service personnel for the new element-type two-state task when there are no adapted service personnel for the new element-type two-state task. The feedback secondary analysis module is used to re-analyze and select service personnel to serve customers if they are not satisfied with the feedback on the new element-based two-state task after the service is completed.

[0012] Compared with existing technologies, the beneficial effects achieved by this invention are as follows: On the one hand, by standardizing the fixed format of element-based legal documents and clarifying the blank items and optional range, the customer's filling process is clearer and smoother, reducing rework caused by inconsistent formats or unclear filling. The data preprocessing and normalization stages simplify the information organization steps, avoiding the tedious operation of manual screening and proofreading, and significantly shortening the task preparation time. At the same time, the construction of the service personnel database integrates historical service experience, eliminating the need to reassess personnel capabilities each time, making the task docking process more efficient, effectively reducing time and labor costs, and improving the overall service operation speed.

[0013] On the one hand, a service personnel database built upon the characteristics of successful cases provides a reliable basis for task matching. By analyzing the compatibility between task completion characteristics and the service personnel's historical successful cases, personnel with corresponding service capabilities are accurately selected, ensuring that the service personnel undertaking the tasks meet the client's needs. Whether prioritizing suitable personnel with relevant experience or selecting the most suitable candidates through comprehensive scoring, the professional advantages of service personnel can be maximized, reducing service deviations caused by mismatched capabilities, effectively guaranteeing the professionalism and accuracy of legal document processing, and enhancing client trust.

[0014] On the other hand, integration enables the service system to continuously optimize, adjusting the weights of corresponding fill-in-the-blank features for unsatisfactory service scenarios to ensure subsequent matching better aligns with core customer needs. This dynamic adjustment model can continuously adapt to various new types of element-based legal document tasks, avoiding the limitations of a rigid service model. Simultaneously, personnel selection criteria balance experience and workload, guaranteeing service quality while achieving a rational allocation of human resources, forming a virtuous cycle of "matching-service-feedback-optimization." In the long run, this continuously improves the stability and adaptability of the service system, better meeting diverse legal document processing needs. Attached Figure Description

[0015] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a structural diagram of an intelligent recognition system for legal documents according to the present invention; Figure 2 This is a flowchart of an intelligent recognition method for legal documents according to the present invention. Detailed Implementation

[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] Please see Figure 1 and Figure 2 The present invention provides a technical solution: an intelligent recognition method for legal documents, comprising the following steps: S1. Define the format for the two-dimensional element-based form, obtain the fill-in data of the two-dimensional element-based form entered by the customer, preprocess and normalize the fill-in data to obtain the fill-in characteristics of the two-dimensional element-based form task. S2. Analyze the service personnel, call the successful processing of element-based two-state tasks, and establish a service personnel database based on the characteristics of successfully processing element-based two-state tasks. S3. When a new element-based two-state condition appears, establish a preset task for the new element-based two-state condition, analyze the adaptability of service personnel to the preset task for the new element-based two-state condition, and select suitable service personnel. S4. When there are suitable service personnel for the new element-based two-state task, select the selected service personnel for the new element-based two-state task based on the number of cases the service personnel are busy with. S5. When there are no suitable service personnel for the new element-based two-state task, the service personnel with the highest degree of suitability shall be selected as the selected service personnel for the new element-based two-state task, taking into account the suitability of the fill-in-the-blank features. S6. After the service is completed, if the customer is not satisfied with the feedback on the new element-based two-state task, re-analyze and select service personnel to serve the customer.

[0018] In step S1, a format is defined for the element-based complaint and response documents. These documents are formatted as a fixed table with fixed fill-in-the-blank items. Customers are provided with fill-in-the-blank options, which they can then select to fill in the blanks. This yields the fill-in-the-blank data within the options. The fill-in-the-blank data is preprocessed and normalized, resulting in N fill-in-the-blank features for the element-based complaint task. By standardizing the fixed table format for the element-based complaint, clearly defining the fill-in-the-blank items and providing corresponding optional content, the difficulty for customers to complete the form is significantly reduced. This avoids rework caused by inconsistent formats, ambiguous descriptions, or missing content, making the completion process smoother and more efficient. Simultaneously, the collected fill-in-the-blank data undergoes preprocessing and normalization, effectively eliminating invalid information and standardizing the data format. This standardization reduces manpower and error risks in subsequent information processing and extracts accurate fill-in-the-blank features, providing reliable and unified data support for subsequent service personnel matching, thus facilitating the efficient advancement of the overall service process.

[0019] In step S2, the service personnel handling the X element-based two-state tasks are analyzed. For the x-th service personnel (x=1,2,…,X), the number of element-based two-state tasks successfully handled by the x-th service personnel in historical data is Y. The successful handling of an element-based two-state task by the x-th service personnel indicates that the customer feedback on the element-based two-state task was satisfactory after the x-th service personnel provided service. Therefore, the characteristics {A} of the y-th successfully handled element-based two-state task are obtained. y_1 A y_2 ,…,A y_n ,…,A y_N}, where A y_nLet y represent the nth fill-in-the-blank feature of the y-th successfully processed element-based two-state task. Substituting each of these features into y=1,2,…,Y yields the features of Y successfully processed element-based two-state tasks. A service personnel database is then established based on these features. By focusing on customer satisfaction cases of service personnel, task features are extracted and used to construct the database, effectively integrating the historical success experiences of service personnel. This database clearly presents the task types and core competencies that each service personnel excels in, avoiding the drawbacks of traditional personnel competency assessments that rely on subjective judgment or fragmented experience. The success cases upon which the database is based are centered on customer feedback, ensuring the authenticity and effectiveness of the information. This provides solid data support for the accurate matching of subsequent tasks and personnel, making personnel competency assessments more objective and targeted, and contributing to improving the scientific rigor and reliability of overall service matching.

[0020] In step S3, when a new element-type two-state condition appears, a preset task for the new element-type two-state condition is established. The adaptability of service personnel to the preset task for the new element-type two-state condition is analyzed. The fill-in-the-blank characteristics of the preset task for the new element-type two-state condition are {A1, A2, ..., A...}. n ,…,A N}, where A n Let n be the fill-in-the-blank feature of the new element-based two-state preset task. For the x-th service personnel, call the service personnel database and analyze the matching degree B between the y-th successfully processed element-based two-state task and the new element-based two-state preset task. y : ; Where k n This represents the preset weight of the nth fill-in-the-blank feature, which is then used to determine the matching relationship between the yth successfully processed element-based two-state task and the new element-based two-state preset task. When B y When >C, it is determined that the y-th successfully processed element-type two-state task matches the new element-type two-state preset task, and the y-th successfully processed element-type two-state task is a matching element-type two-state task of the new element-type two-state preset task; otherwise, it is determined that the y-th successfully processed element-type two-state task does not match the new element-type two-state preset task, and the y-th successfully processed element-type two-state task is not a matching element-type two-state task of the new element-type two-state preset task. Substituting y = 1, 2, ..., Y into each element, the matching relationship between Y successfully processed element-based two-state tasks and the new element-based two-state preset task is determined. This yields the Y0 matching element-based two-state task for the x-th service personnel. If Y0 is not 0, the x-th service personnel is listed as a suitable service personnel; otherwise, the x-th service personnel is listed as a candidate service personnel. Substituting x = 1, 2, ..., X into each element, Q suitable service personnel for the new element-based two-state preset task are obtained, and these suitable service personnel are added to the suitable service personnel directory for the new element-based two-state preset task. By analyzing the matching relationship between the new element-based two-state preset task and the service personnel's historical successful cases, suitable service personnel with corresponding service experience are accurately selected. This process relies on the characteristics of real successful cases for matching, eliminating the drawbacks of subjective experience judgment and ensuring that the selected suitable personnel can meet the fill-in-the-blank characteristics requirements of the new task. Clearly identifying highly suitable personnel as suitable candidates not only provides a precise personnel reserve for new task allocation but also maximizes the professional advantages of service personnel, effectively reducing service deviations caused by mismatched capabilities. This improves the accuracy of task-person matching from the source and lays a solid foundation for the stable guarantee of subsequent service quality.

[0021] In step S4, when a user issues a service request, the preset task for the new element-based two-state task is transformed into a new element-based two-state task. The number of suitable service personnel is analyzed. If Q is not 0, it is determined that there are suitable service personnel for the new element-based two-state task. The Q-th suitable service personnel are analyzed, and the creation time of the new element-based two-state task is retrieved. The number of busy cases for the q-th suitable service personnel is D. y The number of busy cases refers to the number of element-based two-state tasks that the assigned service personnel have not completed within a preset period starting from the establishment time. Uncompleted element-based two-state tasks represent element-based two-state tasks within a preset period starting from the establishment time, where q = 1, 2, ..., Q. Substituting q = 1, 2, ..., Q into each of these values ​​yields the number of busy cases {E1, E2, ..., E} for Q assigned service personnel. q ,…,E Q}, where E qThis represents the number of busy cases for the q-th assigned service personnel. The service personnel with the fewest busy cases are then selected as the designated service personnel for the new element-based two-state task. By combining the busy case count of assigned service personnel for task allocation, a dynamic balance of human resources is effectively achieved. This mechanism avoids a decline in service quality due to task backlog, ensuring that each person undertaking a task can provide service under a reasonable workload, guaranteeing both service efficiency and maintaining stable service quality. Simultaneously, prioritizing personnel with the fewest busy cases ensures timely response and processing of new tasks, reducing customer wait times and improving service timeliness and customer satisfaction. This allocation method also fully leverages the professional value of assigned personnel, avoiding resource idleness or oversaturation, optimizing the overall operational efficiency of the service system, and achieving efficient task and personnel matching.

[0022] In step S5, when Q is 0, it is determined that there are no suitable service personnel for the new element-based two-state task. The X service personnel are analyzed. For the x-th service personnel, in Y successfully processed element-based two-state tasks, the nth fill-in-the-blank feature of Y successfully processed element-based two-state tasks is {A}. 1_n A 2_n ,…,A y_n ,…,A Y_n For the y-th successfully processed element-based two-state task, the degree of difference between the nth fill-in-the-blank feature of the y-th successfully processed element-based two-state task and the nth fill-in-the-blank feature of the new element-based two-state task is F. n_y F n_y =|(A y_n -A n ) / A y_n |, when F n_y <F y When the nth blank feature of the yth successfully processed element-based two-state task is similar to the nth blank feature of the new element-based two-state task, it is determined that the nth blank feature of the yth successfully processed element-based two-state task is not similar to the nth blank feature of the new element-based two-state task. This leads to the number of matching features G for the nth blank feature of the xth service personnel. n_x The number of adaptable features represents the number of fill-in-the-blank features in the nth fill-in-the-blank feature of the new element-based two-state task that are similar to the nth fill-in-the-blank feature of the new element-based two-state task. Substituting these into x=1,2,…,X, we obtain the number of adaptable features {G} for the nth fill-in-the-blank feature of X service personnel. n_1 G n_2 ,…,G n_x ,…,G n_X}, and thus obtain the nth fill-in-the-blank feature fit score H for the xth service personnel. n_x : ; Substitute each of the following into the given values ​​n=1,2,…,N to obtain the N fill-in-the-blank feature matching scores for the x-th service personnel, and then obtain the comprehensive score J for the x-th service personnel. x J x The sum of the fitting scores for the N fill-in-the-blank features of the x-th service personnel is substituted into x=1,2,…,X to obtain the comprehensive score of the X service personnel. The service personnel with the highest comprehensive score are selected as the designated service personnel for the new element-based two-state task. When there are no suitable service personnel, this mechanism quantifies the degree of personnel suitability by analyzing the similarity between the service personnel's historical success cases and the fill-in-the-blank features of the new task, avoiding the bias of subjective judgment. It compares and calculates the fitting score and comprehensive score one by one from the fill-in-the-blank feature dimension, providing an objective and precise basis for personnel selection, ensuring that even in the absence of direct experience matching, the service personnel most suitable for the new task can be selected. This method effectively fills the matching gap when there are vacancies in suitable personnel, ensures the scientific nature of task allocation and the stability of service quality, maximizes the satisfaction of customer needs, and improves the overall suitability of services and customer satisfaction.

[0023] In step S6, after the service of the new element-based two-state task is completed, if the customer is dissatisfied with the feedback of the new element-based two-state task, the fill-in-the-blank feature that the customer is dissatisfied with is called. If the nth fill-in-the-blank feature is the fill-in-the-blank feature that the customer is dissatisfied with, the weight of the nth fill-in-the-blank feature is replaced with α*k. n α is a preset weight amplification coefficient. If α > 1, the selected service personnel are re-analyzed to serve the customer. This constructs a dynamic optimization loop for service feedback, accurately transforming customer dissatisfaction feedback into the basis for adjusting the weights of fill-in-the-blank features. By amplifying the weights of corresponding features, subsequent service personnel analysis focuses more on the core concerns of customers, effectively avoiding the recurrence of similar service deviations and improving service targeting and repair efficiency. Simultaneously, this iterative mechanism based on real feedback allows the service system to continuously adapt to changes in customer needs, constantly improving the matching logic, ensuring the stability of service quality and the improvement of customer satisfaction in the long term, forming a virtuous cycle of "service-feedback-optimization-reservice".

[0024] An intelligent recognition system for legal documents, the system comprising: a fill-in-the-blank data feature extraction module, a service personnel database construction module, a new task adaptation module, a busyness filtering and selection module, a feature adaptation highest selection module, and a feedback secondary analysis module; The fill-in-the-blank data feature extraction module is used to obtain the fill-in-the-blank data of the element-based two-state task by specifying the format of the element-based two-state task, preprocessing and normalizing the fill-in-the-blank data, and obtaining N fill-in-the-blank features of the element-based two-state task. The service personnel database construction module is used to analyze service personnel, call successfully processed element-based two-state tasks, and build a service personnel database based on the characteristics of successfully processed element-based two-state tasks. The new task adaptation module is used to create a new element-based two-state preset task when a new element-based two-state appears, analyze the degree of adaptation of service personnel to the new element-based two-state preset task, and select the appropriate service personnel. The busyness filtering module is used to select service personnel for new element-based two-state tasks when there are suitable service personnel available. The selection of service personnel for new element-based two-state tasks is based on the number of cases they are busy with. The Feature Fitting Highest Selection Module is used when there are no suitable service personnel for the new element-based two-state task. It comprehensively considers the degree of fit of the fill-in-the-blank features and selects the service personnel with the highest degree of fit as the selected service personnel for the new element-based two-state task. The feedback secondary analysis module is used to analyze the feedback after the service is completed. If the customer is not satisfied with the feedback on the new element-based two-state task, the module will re-analyze and select the service personnel to serve the customer.

[0025] Example 1: A customer needs to submit an element-based complaint due to a contract dispute, thus initiating the implementation process of this method.

[0026] In step S1, the system standardizes the element-based complaint into a fixed table format. The table includes fixed fill-in-the-blank items such as "Dispute Type," "Contract Subject Matter," and "Breach of Contract Circumstances," with clear options for each item for the client to choose from. After the client completes the selection based on their own situation, the system preprocesses the collected fill-in-the-blank data, cleaning up invalid information and standardizing the format. Subsequently, it extracts several fill-in-the-blank features of the complaint task through normalization, laying the data foundation for subsequent operations.

[0027] In step S2, the system analyzes multiple service personnel handling element-based two-state tasks. Taking service personnel Zhang San and Li Si as examples, the system retrieves their historical successful cases of element-based two-state tasks with satisfactory customer feedback, extracts the fill-in-the-blank features from these cases, and uses this to build a service personnel database. Zhang San's successful cases are mostly concentrated in the field of contract disputes, while Li Si excels in tort cases.

[0028] When the contract dispute complaint appears as a new element-based two-file pre-set task, step S3 is initiated. The system accesses the service personnel database and analyzes the matching degree between the successful cases of service personnel such as Zhang San and Li Si and the new task. By comparing the case characteristics with the new task's fill-in-the-blank characteristics one by one, it is determined that Zhang San's multiple successful cases have a high matching degree with the new task. Therefore, Zhang San is listed as a suitable service personnel, while Li Si is listed as a candidate because his area of ​​expertise is not compatible.

[0029] After a user submits a service request, the new element-based two-state preset task is transformed into a new task, proceeding to step S4. The system analyzes the number of suitable service personnel and finds that only Zhang San is available. Further analysis of Zhang San's busy case count, i.e., the number of element-based two-state tasks he has not completed within the preset period, reveals that his current task load is moderate, so Zhang San is selected as the service personnel for this new task.

[0030] Assuming there are no suitable service personnel at this time, proceed to step S5. The system analyzes all service personnel. Taking service personnel Wang Wu as an example, although he has no successful cases of direct contract disputes, the system compares the similarity between the filling characteristics of his historical successful cases and the filling characteristics of the new task, calculates the matching score of each characteristic, and finally Wang Wu has the highest comprehensive score. Therefore, Wang Wu is selected to undertake the task.

[0031] If, after Zhang San completes the service, the customer is dissatisfied with the service feedback regarding the contract dispute complaint, and the dissatisfaction is concentrated on the fill-in-the-blank features related to "interpretation of contract terms," ​​then proceed to step S6. The system calls upon the fill-in-the-blank feature indicating dissatisfaction, amplifies its weight, re-analyzes Zhang San's service capabilities, and adjusts the subsequent task allocation logic to prevent similar dissatisfaction from recurring, thereby achieving dynamic optimization of the service system.

[0032] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary sensing device embodiments described above, and that the invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. An intelligent recognition method for legal documents, characterized in that: The method includes the following steps: S1. Define the format for the two-dimensional element-based form, obtain the fill-in data of the two-dimensional element-based form entered by the customer, preprocess and normalize the fill-in data to obtain the fill-in characteristics of the two-dimensional element-based form task. S2. Analyze the service personnel, call the successful processing of element-based two-state tasks, and establish a service personnel database based on the characteristics of successfully processing element-based two-state tasks. S3. When a new element-based two-state condition appears, establish a preset task for the new element-based two-state condition, analyze the adaptability of service personnel to the preset task for the new element-based two-state condition, and select suitable service personnel. S4. When there are suitable service personnel for the new element-based two-state task, select the selected service personnel for the new element-based two-state task based on the number of cases the service personnel are busy with. S5. When there are no suitable service personnel for the new element-based two-state task, the service personnel with the highest degree of suitability shall be selected as the selected service personnel for the new element-based two-state task, taking into account the suitability of the fill-in-the-blank features. S6. After the service is completed, if the customer is not satisfied with the feedback on the new element-based two-state task, re-analyze and select service personnel to serve the customer.

2. The intelligent recognition method for legal documents according to claim 1, characterized in that: In step S1, the format of the two element-based statements is defined. The two element-based statements include the element-based complaint and the element-based answer. The two element-based statements are defined as a fixed table format with fixed fill-in-the-blank items. The fill-in-the-blank items are provided to the customer as options. The fill-in-the-blank items are selected by the customer to fill in the blanks, thereby obtaining the fill-in-the-blank data in the options of the two element-based statements. The fill-in-the-blank data is preprocessed and normalized. After preprocessing and normalization, N fill-in-the-blank features of the two element-based statement task are obtained.

3. The intelligent recognition method for legal documents according to claim 2, characterized in that: In step S2, the service personnel handling the X element-based two-state tasks are analyzed. For the x-th service personnel (x=1,2,…,X), the number of element-based two-state tasks successfully handled by the x-th service personnel in historical data is Y. The successful handling of an element-based two-state task by the x-th service personnel indicates that the customer feedback on the element-based two-state task was satisfactory after the x-th service personnel provided service. Therefore, the characteristics {A} of the y-th successfully handled element-based two-state task are obtained. y_1 A y_2 ,…,A y_n ,…,A y_N }, where A y_n Let y represent the nth fill-in feature of the yth successfully processed element-type two-state task. Substitute each of these features into y=1,2,…,Y to obtain the features of the Y successfully processed element-type two-state tasks. Establish a service personnel database based on the features of successfully processed element-type two-state tasks.

4. The intelligent recognition method for legal documents according to claim 3, characterized in that: In step S3, when a new element-type two-state condition appears, a preset task for the new element-type two-state condition is established. The adaptability of service personnel to the preset task for the new element-type two-state condition is analyzed. The fill-in-the-blank characteristics of the preset task for the new element-type two-state condition are {A1, A2, ..., A...}. n ,…,A N }, where A n Let n be the fill-in-the-blank feature of the new element-based two-state preset task. For the x-th service personnel, call the service personnel database and analyze the matching degree B between the y-th successfully processed element-based two-state task and the new element-based two-state preset task. y : ； Where k n This represents the preset weight of the nth fill-in-the-blank feature, which is then used to determine the matching relationship between the yth successfully processed element-based two-state task and the new element-based two-state preset task. When B y When >C, determine that the y-th successfully processed element-type two-state task matches the new element-type two-state preset task, and the y-th successfully processed element-type two-state task is a matching element-type two-state task of the new element-type two-state preset task; otherwise, determine that the y-th successfully processed element-type two-state task does not match the new element-type two-state preset task, and the y-th successfully processed element-type two-state task is not a matching element-type two-state task of the new element-type two-state preset task.

5. The intelligent recognition method for legal documents according to claim 4, characterized in that: Substitute y=1,2,…,Y one by one, determine the matching relationship between the Y items of successfully processed element-based two-state tasks and the new element-based two-state preset tasks, and obtain the Y0 item matching element-based two-state task for the x-th service personnel for the new element-based two-state preset task. If Y0 is not 0, list the x-th service personnel as the appropriate service personnel; otherwise, list the x-th service personnel as the candidate service personnel. Substitute x=1,2,…,X one by one to obtain the Q appropriate service personnel for the new element-based two-state preset task, and add the appropriate service personnel to the appropriate service personnel directory for the new element-based two-state preset task.

6. The intelligent recognition method for legal documents according to claim 5, characterized in that: In step S4, when a user issues a service request, the preset task for the new element-based two-state task is transformed into a new element-based two-state task. The number of suitable service personnel is analyzed. If Q is not 0, it is determined that there are suitable service personnel for the new element-based two-state task. The Q-th suitable service personnel are analyzed, and the creation time of the new element-based two-state task is retrieved. The number of busy cases for the q-th suitable service personnel is D. y The number of busy cases refers to the number of element-based two-state tasks that the assigned service personnel have not completed within a preset period starting from the establishment time. Uncompleted element-based two-state tasks represent element-based two-state tasks within a preset period starting from the establishment time, where q = 1, 2, ..., Q. Substituting q = 1, 2, ..., Q into each of these values ​​yields the number of busy cases {E1, E2, ..., E} for Q assigned service personnel. q ,…,E Q }, where E q This represents the number of busy cases for the q-th matching service personnel. The matching service personnel with the fewest busy cases are then selected as the selected service personnel for the new element-type two-state task.

7. The intelligent recognition method for legal documents according to claim 6, characterized in that: In step S5, when Q is 0, it is determined that there are no suitable service personnel for the new element-based two-state task. The X service personnel are analyzed. For the x-th service personnel, in Y successfully processed element-based two-state tasks, the nth fill-in-the-blank feature of Y successfully processed element-based two-state tasks is {A}. 1_n A 2_n ,…,A y_n ,…,A Y_n For the y-th successfully processed element-based two-state task, the degree of difference between the nth fill-in-the-blank feature of the y-th successfully processed element-based two-state task and the nth fill-in-the-blank feature of the new element-based two-state task is F. n_y F n_y =|(A y_n -A n ) / A y_n |, when F n_y <F y When the nth blank feature of the yth successfully processed element-based two-state task is similar to the nth blank feature of the new element-based two-state task, it is determined that the nth blank feature of the yth successfully processed element-based two-state task is not similar to the nth blank feature of the new element-based two-state task. This leads to the number of matching features G for the nth blank feature of the xth service personnel. n_x The number of adaptable features represents the number of fill-in-the-blank features in the nth fill-in-the-blank feature of the new element-based two-state task that are similar to the nth fill-in-the-blank feature of the new element-based two-state task. Substituting these into x=1,2,…,X, we obtain the number of adaptable features {G} for the nth fill-in-the-blank feature of X service personnel. n_1 G n_2 ,…,G n_x ,…,G n_X }, and thus obtain the nth fill-in-the-blank feature fit score H for the xth service personnel. n_x : ； Substitute each of the following into the given values ​​n=1,2,…,N to obtain the N fill-in-the-blank feature matching scores for the x-th service personnel, and then obtain the comprehensive score J for the x-th service personnel. x J x The sum of the fitting scores for the N fill-in-the-blank features of the x-th service personnel is substituted into x=1,2,…,X to obtain the comprehensive scores of the X service personnel. The service personnel with the highest comprehensive scores are selected as the selected service personnel for the new element-based two-state task.

8. The intelligent recognition method for legal documents according to claim 6, characterized in that: In step S6, after the service of the new element-based two-state task is completed, if the customer is dissatisfied with the feedback of the new element-based two-state task, the fill-in-the-blank feature that the customer is dissatisfied with is called. If the nth fill-in-the-blank feature is the fill-in-the-blank feature that the customer is dissatisfied with, the weight of the nth fill-in-the-blank feature is replaced with α*k. n α is the preset weight amplification factor. If α > 1, the selected service personnel will be analyzed again to serve customers.

9. An intelligent recognition system for legal documents, wherein the system is applied to the intelligent recognition method for legal documents as described in any one of claims 1-8, characterized in that: The system includes: a fill-in-the-blank data feature extraction module, a service personnel database construction module, a new task adaptation module, a busyness filtering and selection module, a feature adaptation highest selection module, and a feedback secondary analysis module. The fill-in data feature extraction module is used to obtain the fill-in data of the element-type two-form according to the specified format, and to preprocess and normalize the fill-in data to obtain N fill-in features of the element-type two-form task. The service personnel database construction module is used to analyze service personnel, call successfully processed element-based two-state tasks, and build a service personnel database based on the characteristics of successfully processed element-based two-state tasks. The new task adaptation module is used to create a new element-based two-state preset task when a new element-based two-state appears, analyze the degree of adaptation of service personnel to the new element-based two-state preset task, and select the appropriate service personnel. The busyness filtering and selection module is used to select the appropriate service personnel for the new element-based two-state task based on the number of busy cases of the service personnel when there are suitable service personnel for the new element-based two-state task. The feature adaptation highest selection module is used to select the service personnel with the highest degree of adaptation as the selected service personnel for the new element-type two-state task when there are no adapted service personnel for the new element-type two-state task. The feedback secondary analysis module is used to re-analyze and select service personnel to serve customers if they are not satisfied with the feedback on the new element-based two-state task after the service is completed.

Images