A large model-driven data classification and grading system

Abstract

The present invention relates to a large-model-driven data classification and grading system, belonging to the field of data processing technology. It solves the problem that existing data classification and grading results are inaccurate and consume a lot of time and manpower. It includes a rule base, a comprehensive large-model built based on LangChain, a classification agent, and a grading agent. The comprehensive large-model includes a task tree generation component, a task distribution component, and a task tracking component. The task tree generation component extracts tasks from the rule base to build a task tree. The task distribution component extracts tasks from the task tree and distributes them to the classification agent or the grading agent. The task tracking component receives and evaluates the execution results of the classification agent and the grading agent. The classification agent calls a data classification tool to perform classification and sends the classification to the task tracking component. The grading agent calls a data grading tool to perform grading and sends the grading execution results to the task tracking component. This improves the quality and efficiency of data classification and grading.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a large model driven data classification and grading system. Background Art

[0002] With the development of data classification and grading technology and the improvement of relevant standards, complex and diverse business scenarios and multi-dimensional data classification and grading goals have become difficulties faced by data classification and grading.

[0003] Most of the existing data classification and grading technologies are based on pre-defined static rules. With the development of data classification and grading technologies and related standards, multi-dimensional data classification and grading has become mainstream. When facing new business scenarios and multi-dimensional data classification and grading goals, data classification and grading technologies based on static rules require a lot of time and manpower to add and modify rules.

[0004] Moreover, similar data may have significantly different classification results in different business scenarios. Conventional data classification methods only focus on the data itself, without considering the specificity of the business scenario or multi-dimensional indicators. Summary of the Invention

[0005] In view of the above analysis, an embodiment of the present invention aims to provide a large model-driven data classification and grading system to solve the problem that existing data classification and grading results are inaccurate and consume a lot of time and manpower.

[0006] The embodiment of the present invention provides a large-model driven data classification and grading system, comprising: a rule base, and a comprehensive large-model, a classification agent, and a grading agent built on LangChain;

[0007] The overall model includes a task tree generation component, a task distribution component, and a task tracking component. The task tree generation component extracts tasks from the rule base based on the received task identifiers and constructs a task tree. The task distribution component extracts pending tasks from the task tree and distributes them to the classification agent or hierarchical agent based on the evaluation results of the task tracking component. The task tracking component receives and evaluates the execution results of the classification agent and the hierarchical agent, and sends the evaluation results to the task distribution component.

[0008] The classification agent calls the data classification tool to classify the relevant data according to the received to-do tasks, and sends the classification execution results to the task tracking component;

[0009] The grading agent calls the data grading tool to grade the relevant data according to the received to-do tasks and preset grading indicators, and sends the grading execution results to the task tracking component.

[0010] Based on further improvements of the above system, the rule base includes multiple classification tasks and hierarchical tasks, each task includes: task ID, parent task ID, child task ID, task name, data source, data description, task goal, task description, task requirements, task parameters, tool model and tool hyperparameters.

[0011] Based on the further improvement of the above system, the task tree generation component extracts tasks from the rule base according to the received task identifier and constructs them into a task tree. The task corresponding to the task identifier is taken as the root task, and all subtasks are traversed layer by layer from the rule base according to the subtask identifier of the root task until the subtask identifier is empty to obtain tree-structured data.

[0012] Based on the further improvement of the above system, the task distribution component extracts the to-do tasks from the task tree according to the evaluation results of the task tracking component, including:

[0013] Initially, take the root task from the task tree as the to-do task;

[0014] When the received evaluation result is normal, the next task is taken from the task tree as the to-do task in a breadth-first manner;

[0015] When the received evaluation result is timeout and the number of executions of the to-do task does not exceed the threshold, the to-do task is retrieved again;

[0016] When the received evaluation result is abnormal, or the number of executions of the to-do task exceeds the threshold, the to-do task and all subtasks with the to-do task as the parent task will no longer be executed, and the next task will be taken from the task tree as the to-do task in a breadth-first manner.

[0017] Based on the further improvement of the above system, the classification agent calls the data classification tool to classify the relevant data according to the received to-do tasks, including:

[0018] According to the task ID of the to-do task, it is determined whether the received to-do task is an existing task or a new task. If it is an existing task that has not been modified, the data classification tool is called based on the to-do task to classify the relevant data; if it is a new task or a modified existing task, and the data source of the to-do task is a structured database table, it is determined whether the data description of the to-do task can be aligned with the metadata information of each field in the database table. If so, the data classification tool is called based on the aligned to-do task to classify the relevant data; otherwise, the misaligned exception information is sent to the task tracking component as the classification execution result.

[0019] Based on the further improvement of the above system, the data classification tool is called to classify the relevant data, including:

[0020] Use the data extraction tool bound to the first LLM model to obtain the data to be classified from the data source of the to-be-done tasks;

[0021] When the tool model and tool hyperparameters of the to-do task are not empty, the corresponding tool model is extracted from the data classification tool bound to the first LLM model, and the hyperparameters are obtained from the tool hyperparameters and set to the tool model. The first LLM model calls the tool model to complete the data classification task according to the task objectives of the to-do task for the classified data; otherwise, the classification prompt words are constructed according to the to-do task, and the first LLM model completes the data classification task according to the classification prompt words.

[0022] Based on further improvements to the above system, we can identify whether the data description of the to-do task letter is aligned with the metadata information of each field in the database table, including:

[0023] Filter out the attributes in the data description that do not exist in the metadata information of each field as the attributes to be identified;

[0024] Get the similarity threshold in the task parameters of the to-be-done task, and identify whether there is a field in the metadata information whose similarity with the attribute to be identified exceeds the similarity threshold. If so, alignment is possible, and the field corresponding to the maximum similarity is used to replace the attribute to be identified; otherwise, alignment cannot be performed.

[0025] Based on the further improvement of the above system, the grading agent calls the data grading tool to grade the relevant data according to the received to-do tasks and the preset grading indicators, including:

[0026] Use the data extraction tool bound to the second LLM model to obtain each attribute and its data record from the data source in the to-do task;

[0027] According to the task requirements of the to-do task and the data records of each attribute, the scenario weight vector of each attribute in each grading indicator is obtained;

[0028] According to the embedding vector representation of each attribute, obtain the indicator compliance vector of each attribute on each grading indicator;

[0029] According to the scenario weight vector and indicator compliance vector of each attribute, the data grading tool is called to obtain the level of each attribute.

[0030] Based on the further improvement of the above system, the scenario weight vector of each attribute on each grading indicator is obtained by randomly taking out at least one data record corresponding to the attribute from the data source, constructing a weight distribution prompt word according to the data record, task requirements and each grading indicator, and passing it to the second LLM model.

[0031] Based on the further improvement of the above system, the indicator compliance vector of each attribute on each grading indicator is obtained, which is to pass the embedded vector representation of each attribute into the trained neural network model and output the probability distribution that meets each grading indicator.

[0032] Compared with the prior art, the present invention can achieve at least one of the following beneficial effects:

[0033] 1. By triggering the overall coordination model to plan and distribute pending tasks from the rule base, the classification agent and grading agent are used to complete the classification and grading tasks in the pending tasks respectively during the iterative execution process. The overall coordination model infers and evaluates the execution results. This is completed through multi-party collaboration, which not only ensures the quality of data classification and grading results, but also saves time and labor costs and improves data processing efficiency.

[0034] 2. While classifying and grading, it supports the addition of new tasks and the modification of existing tasks, realizing the automatic expansion and optimization of task rules in the rule library.

[0035] 3. Based on multi-dimensional grading indicators, the scenario weight vector and indicator compliance vector are extracted from task requirements, data content and data attributes. The data grading task is completed by fusing the two vectors, making the grading results more closely aligned with the business scenario, enhancing data security and reducing business risks.

[0036] In the present invention, the above-mentioned technical solutions can be combined with each other to achieve more preferred combinations. Other features and advantages of the present invention will be described in the following description, and some advantages will become apparent from the description or be learned through practice of the present invention. The objectives and other advantages of the present invention can be realized and obtained through the contents particularly pointed out in the description and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The accompanying drawings are only used for the purpose of illustrating specific embodiments and are not to be considered as limiting the present invention. Throughout the drawings, the same reference symbols denote the same components.

[0038] Figure 1 This is a structural diagram of a large model-driven data classification and grading system in an embodiment of the present invention. DETAILED DESCRIPTION

[0039] The preferred embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings, wherein the accompanying drawings constitute a part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, and are not used to limit the scope of the present invention.

[0040] A specific embodiment of the present invention discloses a large model driven data classification and grading system, which is the basis for enterprises to manage data assets. Figure 1 As shown in the figure, the system includes: a rule base, a comprehensive model based on LangChain, a classification agent, and a hierarchical agent;

[0041] The overall model includes a task tree generation component, a task distribution component, and a task tracking component. The task tree generation component extracts tasks from the rule base based on the received task identifiers and constructs a task tree. The task distribution component extracts pending tasks from the task tree and distributes them to the classification agent or hierarchical agent based on the evaluation results of the task tracking component. The task tracking component receives and evaluates the execution results of the classification agent and the hierarchical agent, and sends the evaluation results to the task distribution component.

[0042] The classification agent calls the data classification tool to classify the relevant data according to the received to-do tasks, and sends the classification execution results to the task tracking component;

[0043] The grading agent calls the data grading tool to grade the relevant data according to the received to-do tasks and preset grading indicators, and sends the grading execution results to the task tracking component.

[0044] During implementation, the coordinated large model is triggered to plan and distribute the pending tasks from the rule base. During the iterative execution process, the classification agent and the grading agent are used to complete the classification and grading tasks in the pending tasks respectively. The coordinated large model infers and evaluates the execution results to ensure the quality of the data classification and grading results and optimize the task rules in the rule base.

[0045] It should be noted that the rule base includes multiple classification tasks and grading tasks, each task includes: task ID, parent task ID, child task ID, task name, data source, data description, task goal, task description, task requirements, task parameters, tool model and tool hyperparameters; the task ID is automatically generated when it is saved to the rule base, and the parent task ID and child task ID are automatically associated according to the hierarchical relationship of the tasks; the data description is used to describe data information in multiple dimensions from the attribute name, attribute Chinese name, attribute type, length or precision; the task goal includes the data processing type and the data processing basis; the data processing type includes data classification and data grading; the task requirements are used to indicate the special processing requirements of the task, for example, focusing on data leakage loss, or no special requirements; the task parameters are used to indicate the parameters required by the classification agent or the grading agent during the execution of the task, such as the similarity threshold, the number of sample sampling, etc.; the tool model is used to indicate the tool model used when processing the task, and the tool model comes from the tool library. The tool types in the tool library include: general tools, data classification tools, and data grading tools; general tools include but are not limited to: data extraction tools, data vectorization tools, and image processing tools; data classification tools include but are not limited to: text-CNN classification model, text-SVM classification model, text-decision tree classification model, and image-CNN classification model; data grading tools include but are not limited to: MLP classification model and random forest classification model; tool hyperparameters are the path to the configuration file used to specify the hyperparameters required by the tool model.

[0046] For example, the task rules for “department employee classification” in the rule base are shown in Table 1.

[0047] Table 1 Tasks for “Department Employee Classification”

[0048]

[0049] It should be noted that before data classification and grading, data information from different data sources should be collected first, including structured data and unstructured data. For example, structured data in the telecommunications industry includes master data, transaction data, and analytical data, while unstructured data includes text data and multimedia data.

[0050] Furthermore, the collected data is pre-processed, including data attribute extraction and data cleaning. Data attribute extraction involves obtaining the data attributes of each data table and the relationships between data tables from structured data. Data cleaning includes default value processing, outlier processing, and data format conversion.

[0051] Furthermore, the data attributes are vectorized and stored in a vector database. For example, the Word2Vec model is used to obtain the vector embedding representation of the data attributes and stored in the Chroma or Pinecone vector database to facilitate subsequent retrieval and analysis.

[0052] When performing data classification and grading, the task rules in the rule base are displayed on the interface, or the task rules in the rule base are retrieved by entering the keyword of the task name, and the task rules determined by the operator are submitted and passed into the overall model to start executing the data classification and grading task.

[0053] It should be noted that the coordinated large model is a large model application built on LangChain that combines multiple components to complete the data processing process, including: task tree generation component, task distribution component and task tracking component.

[0054] Among them, the task tree generation component extracts tasks from the rule base according to the received task identifier and constructs them into a task tree. The task corresponding to the task identifier is taken as the root task, and all subtasks are traversed layer by layer from the rule base according to the subtask identifier of the root task until the subtask identifier is empty to obtain tree-structured data.

[0055] Furthermore, the constructed task tree is displayed on the interface, and for each task node on the task tree, the corresponding task and its detailed information are obtained from the rule library according to its corresponding task identifier, and modification operations are provided; at the same time, operators can add new tasks, delete or modify existing tasks according to actual needs, and add corresponding detailed information to the new task rules.

[0056] The adjusted task tree is used as the to-be-processed task tree corresponding to the current operator. After submission, the task tree is received by the task distribution component. The root task is initially taken out from the task tree as the to-be-processed task. According to the data processing type in the task target of the to-be-processed task, the to-be-processed task is distributed to the classification agent or the hierarchical agent.

[0057] It should be noted that the classification agent is a first-level LLM (Large Language Model) model application built on LangChain, which includes a series of processing steps; the hierarchical agent is a second-level LLM model application built on LangChain, which includes a series of processing steps. LLM models include but are not limited to the GPT series model, the LLaMA model, and the Bert model.

[0058] (1) The classification agent calls the data classification tool to classify the relevant data based on the received to-do tasks, including:

[0059] According to the task identifier in the to-do task, identify whether the received to-do task is an existing task or a new task. If it is an existing task that has not been modified, call the data classification tool based on the to-do task to classify the relevant data; if it is a new task or a modified existing task, and the data source of the to-do task is a structured database table, identify whether the data description in the to-do task can be aligned with the metadata information of each field in the database table. If so, call the data classification tool based on the aligned to-do task to classify the relevant data; otherwise, send the misaligned exception information as the classification execution result to the task tracking component.

[0060] It should be noted that since new tasks are added by operators on the interface, the data descriptions entered may not be completely consistent with the data attributes in the data source. Therefore, before performing data classification, the data descriptions of the new tasks must be aligned. For modified existing tasks whose data source is a structured database table, additional identifiers are added to facilitate the classification agent to align the data descriptions of these existing tasks.

[0061] Specifically, we need to identify whether the data description of the to-do task is aligned with the metadata information of each field in the database table, including:

[0062] Attributes that do not exist in the metadata information of each field are screened out from the attributes in the data description as attributes to be identified; illustratively, a preliminary screening is performed based on the attribute name.

[0063] Get the similarity threshold in the task parameters of the to-be-done task, and identify whether there is a field in the metadata information whose similarity with the attribute to be identified exceeds the similarity threshold. If so, alignment is possible, and the field corresponding to the maximum similarity is used to replace the attribute to be identified; otherwise, alignment cannot be performed.

[0064] Specifically, the semantic vectors of the attribute to be identified and each field are obtained respectively. For example, the semantic vector of the attribute to be identified is obtained according to the attribute name and the Chinese description of the attribute, and the semantic vector of the field is obtained according to the field name and the field description; the cosine similarity between the semantic vector of the attribute to be identified and the semantic vector of each field is calculated. When the cosine similarity exceeds the set similarity threshold, it means that it can be aligned, and the field corresponding to the maximum cosine similarity is taken to replace the attribute to be identified; when the cosine similarity does not exceed the similarity threshold, it means that there is a serious error in the new task, and the misaligned exception information is sent as the execution result of the classification to the task tracking component of the overall large model.

[0065] Preferably, when the cosine similarities do not exceed the similarity threshold, the similarity threshold is gradually reduced in proportion and is not less than the minimum similarity threshold until the maximum cosine similarity is greater than the reduced similarity threshold, the field corresponding to the maximum cosine similarity is taken out to replace the attribute to be identified, and the reduced similarity threshold is recorded at the same time; when the task can be correctly executed, the similarity threshold in the task parameters is updated with the reduced similarity threshold, and the task in the rule base is optimized.

[0066] For existing tasks that have not been modified, new tasks that have completed information alignment, and existing tasks, call data classification tools to classify the relevant data, including:

[0067] Using a data extraction tool bound to the first LLM model to obtain the data to be classified from the data source of the to-do task; wherein the data extraction tool is a general tool in the tool library and can be bound to multiple LLM models. The data extraction tool includes but is not limited to: connecting to a relational database and obtaining database table data; accessing a file system and obtaining file data;

[0068] When the tool model and tool hyperparameters of the to-do task are not empty, the corresponding tool model is extracted from the data classification tool bound to the first LLM model, and the hyperparameters are obtained from the tool hyperparameters and set to the tool model. The first LLM model calls the tool model to complete the data classification task according to the task objectives of the to-do task for the classified data; otherwise, the classification prompt words are constructed according to the to-do task, and the first LLM model completes the data classification task according to the classification prompt words.

[0069] It should be noted that the role of the first LLM model is defined as a data classification expert in the classification prompt words, and the data description of the to-be-classified task is used as the context of the data to be classified; the task objectives of the to-be-classified task are used as the goals of the first LLM model; the task requirements of the to-be-classified task are used as the classification constraints of the first LLM model, and the first LLM model is required to automatically select a suitable tool model from the bound data classification tools for classification by analyzing the data to be classified and the task objectives.

[0070] Furthermore, the first LLM model of the classification agent organizes the execution results and classification results according to the set format to form a classification result report, and sends it to the task tracking component of the overall large model.

[0071] The task tracking component obtains the evaluation result by identifying whether there are any exceptions in the classification result report. Specifically, if there are no exceptions, the evaluation result is normal; if there is an exception and the exception type is failure to obtain data or a running error, the evaluation result is abnormal and the relevant information is recorded in the log; if there is an exception and the exception type is no response, the evaluation result is timeout and the execution count increases by 1.

[0072] Preferably, considering that when aligning information between a new task and a modified existing task, in some cases the similarity of multiple attributes may be greater than the similarity threshold and the difference in similarity is relatively small, in this case, the attribute with the greatest similarity is first selected to replace the attribute to be identified. The data extraction tool samples data from the data source according to the number of samples in the task parameters, performs data classification, and sends the classification execution results to the task tracking component. If the evaluation result of the task tracking component is normal, the classification agent extracts all data for data classification and sends the new classification execution results to the task tracking component; if the evaluation result of the feedback component is abnormal, the classification agent selects the attribute with the second greatest similarity to replace the attribute to be identified, and repeats the above process. If the evaluation result of the task tracking component is still normal, the evaluation result is sent to the task distribution component.

[0073] (2) The grading agent calls the data grading tool to grade the relevant data based on the received to-do tasks and the preset grading indicators, including:

[0074] Using the data extraction tool bound to the second LLM model, obtain each attribute and its data record from the data source of the to-do task;

[0075] According to the task requirements of the to-do task and the data records of each attribute, the scenario weight vector of each attribute in each grading indicator is obtained;

[0076] According to the embedding vector representation of each attribute, obtain the indicator compliance vector of each attribute on each grading indicator;

[0077] According to the scenario weight vector and indicator compliance vector of each attribute, the data grading tool is called to obtain the level of each attribute.

[0078] It should be noted that the grading indicators of this embodiment include: confidentiality, availability, timeliness, integrity, data value and leakage loss; among them, confidentiality means that the data has confidentiality requirements and must be encrypted; availability means that the grading and desensitization of data cannot have a significant impact on data availability; timeliness means that the data grading changes over time, such as strong timeliness within a certain period of time; data value means that the grading and desensitization of data cannot have a significant impact on the value of the data; leakage loss refers to the need to consider the economic losses caused to the enterprise by the leaked data.

[0079] Specifically, the scenario weight vector of each attribute on each grading indicator is obtained by randomly extracting at least one data record corresponding to the attribute from the data source, constructing a weight distribution prompt word based on the data record, task requirements and each grading indicator, and passing it to the second LLM model.

[0080] In the weight allocation prompt, the role of the second LLM model is defined as the indicator weight allocation expert. The description of each hierarchical indicator is used as the background, each attribute and at least one data record thereof is used as the context, the task requirements and the value range of each hierarchical indicator weight are used as the allocation constraints, and the appropriate weight value is assigned to each indicator according to the task requirements and the data records in each attribute as the allocation target. The weight is also defined to be output in the order of the hierarchical indicators.

[0081] The scene weight vector output by the second LLM model , where n represents the total number of grading indicators, Represents the weight of the i-th grading indicator, i=1,2,...,n, which is used to indicate the importance of the i-th grading indicator in the current business scenario. The larger the value, the more important the grading indicator.

[0082] For example, the value range of each grading indicator is (0, 2). If the data record of the "telephone number" attribute is "187****0010", then the value of the grading indicator "leakage loss" may be set to 0.5, indicating that data leakage will not cause significant losses.

[0083] Furthermore, obtaining the indicator compliance vector for each attribute on each grading indicator involves passing the embedding vector representation of each attribute into a trained neural network model and outputting the probability distribution of the attribute's compliance with each grading indicator, thereby mapping the attribute vector to each grading indicator. Exemplarily, the neural network model employs a multilayer perceptron neural network.

[0084] Index Conformity Vector , the same dimension as the scene weight vector, Indicates the degree of conformity of the attribute vector to the i-th grading index. The larger the value, the more it satisfies the grading index.

[0085] Furthermore, the scenario weight vector of each attribute is fused on the basis of the indicator compliance vector, and the scenario compliance vector is obtained by the following formula: :

[0086] .

[0087] Finally, the scenario conformity vector is passed to a data classification tool to obtain the level of each attribute. For example, the data classification tool is a trained random forest model that outputs the predicted level.

[0088] It should be noted that, in this embodiment, there are four levels of data classification, representing four levels of data importance: high, relatively high, medium and low.

[0089] It should be noted that the scenario weight vector reflects the importance of each indicator in the business scenario, and the indicator compliance vector reflects the degree to which the data attributes meet each indicator. The scenario weight vector and the indicator compliance vector work together to obtain the scenario compliance vector in the business scenario. Based on the scenario compliance vector, the data classification task based on multi-dimensional classification indicators is completed, which improves the accuracy of data classification. In addition, each attribute is graded to facilitate the control of data access rights based on the attribute level. For example, if the attribute is at a low level of importance, the data of the attribute is fully public; if the attribute is at a high level of importance, the data of the attribute is only accessible or usable internally by personnel with specific permissions.

[0090] The grading agent forms a grading result report with the embedding vector of each attribute in the grading task, the data records extracted from each attribute, the scenario weight vector, the indicator compliance vector and the grading result, and sends it to the task tracking component of the overall model.

[0091] The task tracking component analyzes and evaluates the grading result report according to the above evaluation method for the classification result report, and sends the obtained evaluation results to the task distribution component.

[0092] Preferably, in the task tracking component, for result reports that do not contain exceptions, secondary reasoning is performed by constructing prompt words to set a series of rules and constraints. If it is inferred that the result report is unreasonable, the evaluation result is set to be abnormal.

[0093] It should be noted that the task distribution component extracts tasks from the task tree based on the evaluation results of the task tracking component, including:

[0094] When the received evaluation result is normal, the next task is taken from the task tree as the to-do task in a breadth-first manner;

[0095] When the received evaluation result is timeout and the number of executions of the to-do task does not exceed the threshold, the to-do task is retrieved again;

[0096] When the received evaluation result is abnormal, or the number of executions of the to-do task exceeds the threshold, the to-do task and all subtasks with the to-do task as the parent task will no longer be executed, and the next task will be taken from the task tree as the to-do task in a breadth-first manner.

[0097] When the task distribution component has no pending tasks to retrieve, it completes the data classification and grading task for the current user. New tasks that are executed normally are added to the rule library, and existing tasks that have been modified are updated with the detailed information of normal execution.

[0098] It's important to note that the results of classification and grading are part of the data asset and, like the original collected data, are vectorized and stored to facilitate retrieval and management of data assets, forming a list of important data and a map of sensitive data assets. Combining data grading results not only facilitates the configuration of permissions for various components, roles, and administrators, but also facilitates data desensitization and encryption.

[0099] Preferably, in order to improve the accuracy of data assets, the task rules in the coordinated large model execution rule library are periodically used to timely update the levels of data and data attributes in each category.

[0100] Compared with the existing technology, the large model-driven data classification and grading system provided in this embodiment plans and distributes pending tasks from the rule base by triggering the overall large model, and uses the classification agent and grading agent to complete the classification and grading tasks in the pending tasks respectively during the iterative execution process. The overall large model infers and evaluates the execution results, while ensuring the quality of the data classification and grading results, saving time and manpower costs and improving data processing efficiency; while performing classification and grading, it supports the addition of new tasks and the modification of existing tasks, and realizes the automatic expansion and optimization of task rules in the rule base; based on multi-dimensional grading indicators, it extracts scenario weight vectors and indicator compliance vectors from task requirements, data content and data attributes, and completes the data grading task through the fusion of the two vectors, so that the grading results are better aligned with the business scenario, enhances data security and reduces business risks.

[0101] Those skilled in the art will appreciate that all or part of the process steps of the above-described embodiments can be implemented by instructing related hardware through a computer program, and the program can be stored in a computer-readable storage medium, such as a magnetic disk, an optical disk, a read-only memory, or a random access memory.

[0102] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by any technician familiar with this technical field within the technical scope disclosed by the present invention should be covered by the scope of protection of the present invention.

Claims

1. A large model driven data classification and grading system, characterized by: include: Rule base, as well as the overall model, classification agents and hierarchical agents built on LangChain; The overall planning model includes a task tree generation component, a task distribution component and a task tracking component; the task tree generation component extracts tasks from the rule base according to the received task identifier to construct a task tree, which takes the task corresponding to the task identifier as the root task, and traverses the rule base layer by layer according to the subtask identifier of the root task to extract all subtasks until the subtask identifier is empty to obtain tree-structured data; the task tree is displayed on the interface end and is used to add new tasks, delete or modify existing tasks; the task distribution component extracts to-do tasks from the task tree according to the evaluation results of the task tracking component and distributes them to the classification agent or the hierarchical agent; the task tracking component receives the execution results of the classification agent and the hierarchical agent and evaluates them, and sends the evaluation results to the task distribution component; The classification agent calls the data classification tool to classify the relevant data according to the received pending tasks, and sends the classification execution results to the task tracking component; The grading agent calls the data grading tool to grade the relevant data according to the received pending tasks and the preset grading indicators, and sends the grading execution results to the task tracking component; The task distribution component extracts the to-do task from the task tree according to the evaluation result of the task tracking component, including: initially extracting the root task from the task tree as the to-do task; When the received evaluation result is normal, the next task is taken out from the task tree in a breadth-first manner as a to-do task; when the received evaluation result is a timeout and the number of executions of the to-do task does not exceed the threshold, the to-do task is taken out again; when the received evaluation result is abnormal, or the number of executions of the to-do task exceeds the threshold, the to-do task and all child tasks with the to-do task as the parent task are no longer executed, and the next task is taken out from the task tree in a breadth-first manner as a to-do task; When there are no pending tasks to be retrieved by the task distribution component, the data classification and grading task of the current operating user is completed; new tasks that are executed normally are added to the rule library, and the existing tasks that are modified are updated according to the detailed information during normal execution.

2. The large model-driven data classification and grading system according to claim 1, characterized in that: The rule base includes multiple classification tasks and hierarchical tasks, each task includes: task identifier, parent task identifier, child task identifier, task name, data source, data description, task target, task description, task requirements, task parameters, tool model and tool hyperparameters.

3. The large model driven data classification and grading system according to claim 2, characterized in that: The classification agent calls a data classification tool to classify relevant data according to the received pending tasks, including: According to the task identifier of the to-do task, it is determined whether the received to-do task is an existing task or a new task. If it is an existing task that has not been modified, the data classification tool is called based on the to-do task to classify the relevant data; if it is a new task or a modified existing task, and the data source of the to-do task is a structured database table, it is determined whether the data description of the to-do task can be aligned with the metadata information of each field in the database table. If so, the data classification tool is called based on the aligned to-do task to classify the relevant data; otherwise, the misaligned exception information is sent to the task tracking component as the execution result of the classification.

4. The large model driven data classification and grading system according to claim 2 or 3, characterized in that: The calling of the data classification tool to classify the relevant data includes: Use the data extraction tool bound to the first LLM model to obtain the data to be classified from the data source of the to-be-done tasks; When the tool model and tool hyperparameters of the to-be-done task are not empty, the corresponding tool model is extracted from the data classification tool bound to the first LLM model, and the hyperparameters are obtained from the tool hyperparameters and set to the tool model. The first LLM model calls the tool model to complete the data classification task for the data to be classified according to the task objectives of the to-be-done task; otherwise, a classification prompt word is constructed according to the to-be-done task, and the first LLM model completes the data classification task according to the classification prompt word.

5. The large model driven data classification and grading system according to claim 3, characterized in that: Whether the data description of the identified to-do tasks is aligned with the metadata information of each field in the database table includes: Filtering out attributes that do not exist in the metadata information of each field from the attributes in the data description as attributes to be identified; Get the similarity threshold in the task parameters of the to-be-done task, and identify whether there is a field in the metadata information whose similarity with the attribute to be identified exceeds the similarity threshold. If so, alignment is possible, and the field corresponding to the maximum similarity is used to replace the attribute to be identified; otherwise, alignment cannot be performed.

6. The large model driven data classification and grading system according to claim 2, characterized in that: The grading agent calls the data grading tool to grade the relevant data according to the received pending tasks and the preset grading indicators, including: Using the data extraction tool bound to the second LLM model, obtain each attribute and its data record from the data source of the to-do task; According to the task requirements of the to-do task and the data records of each attribute, the scenario weight vector of each attribute in each grading indicator is obtained; According to the embedding vector representation of each attribute, obtain the indicator compliance vector of each attribute on each grading indicator; According to the scenario weight vector and indicator compliance vector of each attribute, the data grading tool is called to obtain the level of each attribute.

7. The large model driven data classification and grading system according to claim 6, characterized in that: The scenario weight vector of each attribute on each grading indicator is obtained by randomly extracting at least one data record corresponding to the attribute from the data source, constructing a weight distribution prompt word based on the data record, task requirements and each grading indicator, and passing it to the second LLM model.

8. The large model driven data classification and grading system according to claim 6, characterized in that: The method of obtaining the index conformity vector of each attribute on each grading index is to input the embedded vector representation of each attribute into the trained neural network model and output the probability distribution conforming to each grading index.

Images