Intelligent data processing system carrying AI digital human

Abstract

The invention discloses an intelligent data processing system carrying an AI digital human. According to the intelligent data processing system, the knowledge graph sub-module carries out deep reasoning on the knowledge graph by combining a rule engine and a machine learning algorithm, and remarkable benefits are brought to the whole intelligent data processing system. Firstly, due to the introduction of the rule engine, the system can automatically generate new knowledge or verify existing knowledge based on a predefined inference rule, and the accuracy and efficiency of knowledge management are improved. The system can quickly infer a new attribute relationship and enrich the content of the knowledge base. Through the application of a support vector machine (SVM), the autonomous learning and reasoning capabilities of the system are enhanced. The SVM model can perform accurate reasoning on unknown knowledge, generate new knowledge or verify existing knowledge by training and learning potential modes in the knowledge graph, so that the intelligent level of data analysis and decision is improved, the data value is improved, and powerful support is provided for intelligent upgrading and business expansion of the system.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent data processing, and specifically relates to an intelligent data processing system equipped with an AI digital human. Background Art

[0002] Data processing refers to a series of processes of collecting, storing, organizing, transforming, and analyzing various data. It is the basic work in the fields of information science and information technology. The goal of data processing is to extract valuable information from a large amount of raw data for decision-making support and knowledge discovery. It involves multiple links such as data collection, cleaning, transformation, loading, analysis, and visualization. By using technical means such as database management systems, data mining, statistical analysis, and machine learning, in-depth understanding and effective utilization of data are achieved. The data processing process emphasizes the accuracy, integrity, and timeliness of data, aiming to improve data quality, reduce data redundancy, ensure data security and privacy, and provide data support and intellectual services for all walks of life. With the rapid development of information technology, data has become an important asset of enterprises and society. However, in the face of massive and complex data, how to efficiently manage and utilize this data has become an urgent problem to be solved. Traditional data processing systems have many deficiencies in data source management, data verification, data analysis, etc., and are difficult to meet the intelligent and efficient requirements of modern enterprises for data processing.

[0003] However, existing data processing systems often rely on manual input and update in knowledge management, which is prone to errors and low in efficiency. The knowledge graph sub-module automatically generates and verifies knowledge through a rule engine, greatly improving the accuracy and efficiency of knowledge management. Summary of the Invention

[0004] The purpose of the present invention is to provide an intelligent data processing system equipped with an AI digital human in order to solve the above-mentioned problems.

[0005] The technical solution adopted by the present invention is as follows: An intelligent data processing system equipped with an AI digital human includes: a data source management module, a data verification module, a data scraping module, a data analysis module, an AI digital human module, an interactive display module, and a system management module;

[0006] The internal of the AI digital human module is provided with a natural language processing sub-module, a knowledge graph sub-module, a dialogue generation sub-module, a speech synthesis sub-module, and an image display sub-module;

[0007] The output end of the data source management module is connected to the input end of the data verification module through a data interface to realize the transmission and reception of raw data.

[0008] The output end of the data verification module is connected to the input end of the data scraping module through a data stream, ensuring that the verified valid data enters the preprocessing process.

[0009] The output end of the data scraping module is connected to the input end of the data analysis module through a data stream, and the preprocessed data is transmitted to the analysis link.

[0010] The output end of the data analysis module is connected to the input ends of the AI digital human module and the interactive display module through a data interface or direct transmission method, and is used to generate intelligent responses and visually display analysis results respectively.

[0011] The interaction interface of the AI digital human module is connected to the input end of the interactive display module to realize the synchronous display of the natural language responses, actions and expressions of the AI digital human and the user interface; at the same time, the input end of the AI digital human module also receives user inputs from the interactive display module through the interaction interface to realize two-way interaction.

[0012] In a preferred embodiment, the internal settings of the data source management module include multiple components such as a data connector, a data extraction engine, a data source monitor, a data source configuration management, a data source permission control, a data source cache mechanism, and a data source metadata management. The data connector is responsible for establishing connections with various data sources, supporting multiple connection protocols to ensure compatibility with different data sources; the data extraction engine realizes automatic data extraction, supports full-volume extraction and incremental extraction, and provides data conversion functions; the data source monitor monitors the status of the data source in real time, alarms in time and records logs in case of anomalies; the data source configuration management provides a configuration interface, allowing users to add, modify, and delete data source information and store the configuration information; the data source permission control manages users' access permissions to the data source, supporting multi-level permission settings; the data source cache mechanism caches frequently accessed data to improve access efficiency and updates the cache data regularly; the data source metadata management stores and manages the metadata of the data source and provides query interfaces to facilitate other modules to obtain data source information. Through the collaborative work of these components, the data source management module ensures the smooth access, effective management, and efficient use of data;

[0013] The internal settings of the data verification module cover multiple functional aspects, including data format verification, data integrity check, data consistency verification, outlier detection and handling, and data quality report generation. Data format verification is responsible for checking whether the data conforms to the predefined format standards, such as field types, lengths, etc.; data integrity check identifies and fills in missing values to ensure the completeness of the data; data consistency verification ensures the consistency of data between different data sources and avoids data conflicts; outlier detection and handling can identify possible incorrect data or abnormal situations and take corresponding measures for processing; data quality report generation regularly generates data quality reports to provide data quality assessment and analysis. Through the comprehensive application of these functions, the data verification module ensures the accuracy and reliability of the data entering the system for analysis, laying a solid foundation for subsequent data processing and analysis.

[0014] In a preferred embodiment, the internal settings of the data scraping module include key components such as a web crawler, an API caller, a data parser, a scraping task scheduler, a scraping result storage, and a scraping log recorder. The web crawler is responsible for scraping data from specified web pages on the Internet, supporting various crawling strategies and anti-crawling mechanisms; the API caller obtains data through API interfaces, supporting various API protocols and authentication methods; the data parser parses the scraped data, extracts the required information, and converts it into a unified format; the scraping task scheduler is responsible for scheduling and managing scraping tasks, supporting scheduled scraping and real-time scraping; the scraping result storage stores the scraped data in a specified location, such as a database or a file system; the scraping log recorder records the scraping process and results for subsequent analysis and troubleshooting. Through the collaborative work of these components, the data scraping module realizes the automatic and efficient scraping and parsing of data.

[0015] In a preferred embodiment, the internal settings of the data analysis module include core functions such as data preprocessing, statistical analysis, data mining, model training, predictive analysis, and result visualization. Data preprocessing performs operations such as data cleaning, transformation, and normalization to prepare for analysis; statistical analysis provides statistical methods such as descriptive statistics, hypothesis testing, and correlation analysis to help understand data characteristics; data mining applies algorithms such as clustering, classification, and association rule mining to discover potential patterns and relationships in the data; model training trains machine learning models based on historical data, such as regression models, decision trees, neural networks, etc.; predictive analysis uses the trained models for future trend prediction and decision support; result visualization displays the analysis results in the form of charts, reports, etc. for easy user understanding and decision-making. Through the comprehensive application of these functions, the data analysis module realizes in-depth analysis and value mining of the data.

[0016] In a preferred embodiment, the natural language processing sub-module is a core component of the AI digital human module, responsible for processing and analyzing the natural language input by users. This module includes a speech recognition unit, a semantic understanding unit, an intent recognition unit, and a slot filling unit. The speech recognition unit uses deep neural network technology to convert the user's speech input into text; the semantic understanding unit understands the meaning of the text through means such as syntactic analysis and semantic role labeling; the intent recognition unit determines which type of predefined intent the user's input belongs to; the slot filling unit extracts key information from the user's input and fills it into predefined slots, providing the necessary data support for subsequent dialogue generation.

[0017] In a preferred embodiment, the knowledge graph sub-module is used to store and manage the knowledge information required by the AI digital human to support intelligent question answering and reasoning functions. This module includes a knowledge extraction unit, a knowledge storage unit, a knowledge query unit, and a knowledge reasoning unit. The knowledge extraction unit extracts structured and unstructured knowledge from various data sources; the knowledge storage unit uses technologies such as graph databases to store knowledge and form a knowledge graph; the knowledge query unit provides an efficient query interface for quickly retrieving relevant knowledge; the knowledge reasoning unit uses a rule engine and machine learning algorithms to perform reasoning on the knowledge graph, generating new knowledge or verifying existing knowledge, and enhancing the intelligence level of the AI digital human;

[0018] The knowledge graph sub-module combines a rule engine and machine learning algorithms to perform reasoning on the knowledge graph, aiming to generate new knowledge or verify existing knowledge;

[0019] The specific algorithm content includes:

[0020] Rule engine part:

[0021] Rule definition: Define a series of reasoning rules, such as "if A is a subclass of B and B has property C, then A also has property C".

[0022] Rule application: Apply the rules to the entities and relationships in the knowledge graph to generate new knowledge or verify existing knowledge.

[0023] Machine learning algorithm part:

[0024] Algorithm selection: Select the support vector machine (SVM) as the reasoning algorithm, which is suitable for classification and regression tasks.

[0025] Feature extraction: Extract features from the knowledge graph, such as entity attributes, relationship types, etc.

[0026] Model training: Use known knowledge to train the SVM model and learn the latent patterns in the knowledge graph.

[0027] Inference application: Use the trained model to infer unknown knowledge, generate new knowledge or verify existing knowledge;

[0028] The training formula for the SVM model is:

[0029] minimize1 / 2||w||^2+C*Σξ_i

[0030] subjecttoy_i*(w·x_i+b)≥1-ξ_i,ξ_i≥0

[0031] The inference formula for the SVM is:

[0032] f(x)=sign(w·x+b)

[0033] Where w is the weight vector, representing the importance of the features learned by the model; b is the bias term, representing the threshold of the model; C is the regularization parameter, controlling the balance between the model complexity and the training error; ξ_i is the slack variable, allowing the model to misclassify some samples during training; y_i is the sample label, representing the positive or negative class; x_i is the feature vector, representing the entity or relationship features in the knowledge graph; sign is the sign function, converting the model output into a classification result.

[0034] In a preferred embodiment, the dialogue generation sub-module is responsible for generating natural and fluent dialogue responses according to the user input and context information. This module includes a dialogue management unit, a language generation unit, and a response selection unit. The dialogue management unit maintains the dialogue state, tracks the dialogue history and user intentions; the language generation unit uses natural language generation techniques, such as sequence-to-sequence models, to generate candidate responses; the response selection unit then selects the best response from the candidate responses according to indicators such as context relevance and language fluency. In addition, the dialogue generation sub-module also considers emotional factors to make the response more user-friendly;

[0035] The speech synthesis sub-module converts the generated text response into natural and realistic speech output. This module includes a text analysis unit, a prosody model unit, a vocoder unit, and a speech synthesis engine. The text analysis unit performs linguistic analysis on the text, extracting information such as pronunciation, stress, and intonation; the prosody model unit generates prosody parameters, such as speech rate, pitch, and volume, according to the text analysis results; the vocoder unit converts the prosody parameters into a speech waveform; the speech synthesis engine then integrates the outputs of the above units to generate the final speech signal. This module also supports multiple speech styles and dialects to meet the needs of different users.

[0036] In a preferred embodiment, the image display sub-module is responsible for the visual presentation of the AI digital human, including an image modeling unit, a motion generation unit, an expression generation unit, and a rendering output unit. The image modeling unit creates a 3D model of the AI digital human according to the design requirements; the motion generation unit generates natural human motions using motion capture or animation technology; the expression generation unit generates corresponding facial expressions based on the conversation content and the results of sentiment analysis; and the rendering output unit integrates the model, motions, and expressions and displays them on the user interface through real-time rendering technology. This module also supports multi-modal interactions, such as gesture recognition and eye tracking, to enhance the interaction experience between the user and the AI digital human.

[0037] In a preferred embodiment, the internal settings of the interactive display module include multiple parts such as user interface design, interactive logic processing, data visualization display, user input processing, and feedback mechanism. The user interface design is responsible for designing an intuitive and user-friendly interface, including layout, color, icons, etc.; the interactive logic processing implements the interactive logic between the user and the system, such as click, drag, zoom, etc.; the data visualization display presents data in the form of charts, graphs, dashboards, etc., supporting multiple visualization methods; the user input processing receives and processes user inputs, such as query conditions, parameter settings, etc.; the feedback mechanism provides user operation feedback, such as prompt messages, loading animations, etc., to enhance the user experience. Through the coordinated work of these parts, the interactive display module realizes a natural and intuitive interaction between the user and the system, improving the user's usage experience.

[0038] In a preferred embodiment, the internal settings of the system management module include multiple submodules such as user management, authority management, system monitoring, log management, backup and recovery, and configuration management. User management is responsible for managing the information and authority of system users, including user registration, login, role allocation and authority setting, to ensure data security and privacy protection; authority management implements authority allocation and control for different user roles, defines the user's access level to system resources, functions and data, and ensures the safe operation of the system; system monitoring monitors the operating status of the system in real time, including server load, database performance, network status, etc., and timely discovers and handles abnormalities by setting thresholds and alarm mechanisms to ensure stable operation of the system; log management records the operation and change history of the system, including user operation logs, system operation logs and error logs, to provide a basis for system troubleshooting, performance analysis and security auditing; backup and recovery implements the data backup and recovery function of the system, supports scheduled backup, manual backup and disaster recovery, ensures that data is not lost, and ensures business continuity; configuration management provides a system configuration interface, allowing administrators to set and adjust system parameters, including database connection configuration, service port configuration, mail server configuration, etc., to adapt to different operating environments and business needs. Through the comprehensive management of these sub-modules, the system management module ensures the stable operation, safe controllability and flexible configuration of the system, providing a solid foundation and strong support for the entire intelligent data processing system.

[0039] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0040] 1. In the present invention, the knowledge graph submodule performs deep reasoning on the knowledge graph by combining the rule engine and the machine learning algorithm, which brings significant benefits to the entire intelligent data processing system. First, the introduction of the rule engine enables the system to automatically generate new knowledge or verify existing knowledge based on predefined reasoning rules, thereby improving the accuracy and efficiency of knowledge management. The system can quickly infer new attribute relationships and enrich the content of the knowledge base. Through the application of support vector machines (SVM), the autonomous learning and reasoning capabilities of the system are enhanced. The SVM model can accurately reason about unknown knowledge, generate new knowledge or verify existing knowledge by training and learning the potential patterns in the knowledge graph, thereby improving the intelligent level of data analysis and decision-making. In addition, this combination method also improves the system's ability to handle complex knowledge relationships, enables the system to better cope with diverse business scenarios, and provides users with more accurate and comprehensive information services. These algorithm contents significantly enhance the system's knowledge processing capabilities, improve the value of data, and provide strong support for the intelligent upgrade and business expansion of the system.

[0041] 2. In the present invention, the AI digital human module realizes highly bionic interaction with users through advanced algorithms such as natural language processing, speech recognition, and image recognition. This interaction method not only greatly improves the user experience, enabling users to communicate with the system in a more natural and intuitive way, but also significantly enhances the efficiency of data query and obtaining analysis results. Users can quickly retrieve specific data through voice commands or intuitively understand complex data analysis results through the visual interface provided by the digital human. In addition, the AI digital human can also provide personalized services based on the user's historical behavior and preferences, further optimizing the interaction process, reducing the user's operation cost, and thus improving the overall work efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 is the overall system block diagram of the present invention;

[0043] Figure 2 is the schematic diagram of the AI digital human module system in the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0044] In order to make the objectives, technical solutions, and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0045] Refer to Figure 1-2 ,

[0046] An intelligent data processing system equipped with an AI digital human, the system includes: a data source management module, a data verification module, a data scraping module, a data analysis module, an AI digital human module, an interaction display module, and a system management module;

[0047] Inside the AI digital human module, there are a natural language processing sub-module, a knowledge graph sub-module, a dialogue generation sub-module, a speech synthesis sub-module, and an image display sub-module;

[0048] The output end of the data source management module is connected to the input end of the data verification module through a data interface to realize the transfer and reception of raw data.

[0049] The output end of the data verification module is connected to the input end of the data scraping module through a data stream to ensure that the verified valid data enters the preprocessing process.

[0050] The output end of the data scraping module is connected to the input end of the data analysis module through a data stream to transfer the preprocessed data to the analysis link.

[0051] The output end of the data analysis module is connected to the input ends of the AI digital human module and the interactive display module through a data interface or a direct transmission method, respectively used to generate intelligent responses and visually display analysis results.

[0052] The interaction interface of the AI digital human module is connected to the input end of the interactive display module to synchronously display the natural language responses, actions, and expressions of the AI digital human with the user interface; at the same time, the input end of the AI digital human module also receives user input from the interactive display module through the interaction interface to achieve two-way interaction.

[0053] The management interfaces of the system management module are respectively connected to the management ends of the data source management module, the data verification module, the data scraping module, the data analysis module, the AI digital human module, and the interactive display module to achieve unified management operations such as configuration, monitoring, maintenance, and upgrade of each module.

[0054] 2. An intelligent data processing system with an AI digital human as claimed in claim 1, characterized in that: the internal settings of the data source management module include multiple components such as a data connector, a data extraction engine, data source monitoring, data source configuration management, data source permission control, a data source caching mechanism, and data source metadata management. The data connector is responsible for establishing connections with various data sources, supporting multiple connection protocols to ensure compatibility with different data sources; the data extraction engine realizes automatic data extraction, supports full-volume extraction and incremental extraction, and provides a data conversion function; the data source monitoring monitors the status of the data source in real time, alarms in time and records logs in case of anomalies; the data source configuration management provides a configuration interface, allowing users to add, modify, and delete data source information and store the configuration information; the data source permission control manages users' access permissions to the data source, supporting multi-level permission settings; the data source caching mechanism caches frequently accessed data to improve access efficiency and periodically updates the cached data; the data source metadata management stores and manages the metadata of the data source, providing a query interface for other modules to obtain data source information. Through the collaborative work of these components, the data source management module ensures the smooth access, effective management, and efficient use of data;

[0055] The internal settings of the data validation module cover multiple functional aspects, including data format verification, data integrity check, data consistency validation, outlier detection and handling, and data quality report generation. Data format verification is responsible for checking whether the data conforms to the predefined format standards, such as field types and lengths; data integrity check identifies and fills in missing values to ensure the completeness of the data; data consistency validation ensures the consistency of data between different data sources and avoids data conflicts; outlier detection and handling can identify potential incorrect data or abnormal situations and take corresponding measures for processing; data quality report generation regularly generates data quality reports to provide data quality assessment and analysis. Through the comprehensive application of these functions, the data validation module ensures the accuracy and reliability of the data entering the system for analysis, laying a solid foundation for subsequent data processing and analysis;

[0056] 3. An intelligent data processing system with an AI digital human as claimed in claim 1, wherein the internal settings of the data scraping module include key components such as a web crawler, an API caller, a data parser, a scraping task scheduler, a scraping result storage, and a scraping log recorder. The web crawler is responsible for scraping data from specified web pages on the Internet, supporting various crawling strategies and anti-crawling mechanisms; the API caller obtains data through API interfaces, supporting various API protocols and authentication methods; the data parser parses the scraped data, extracts the required information, and converts it into a unified format; the scraping task scheduler is responsible for scheduling and managing scraping tasks, supporting scheduled scraping and real-time scraping; the scraping result storage stores the scraped data in a specified location, such as a database or a file system; the scraping log recorder records the scraping process and results for subsequent analysis and troubleshooting. Through the collaborative work of these components, the data scraping module realizes the automatic and efficient scraping and parsing of data.

[0057] The internal settings of the data analysis module include core functions such as data preprocessing, statistical analysis, data mining, model training, predictive analysis, and result visualization. Data preprocessing performs operations such as cleaning, transforming, and normalizing the data to prepare for analysis; statistical analysis provides statistical methods such as descriptive statistics, hypothesis testing, and correlation analysis to help understand data characteristics; data mining applies algorithms such as clustering, classification, and association rule mining to discover potential patterns and relationships in the data; model training trains machine learning models based on historical data, such as regression models, decision trees, and neural networks; predictive analysis uses the trained models for future trend prediction and decision support; result visualization displays the analysis results in the form of charts, reports, etc. for easy user understanding and decision-making. Through the comprehensive application of these functions, the data analysis module realizes in-depth analysis and value mining of the data.

[0058] The natural language processing sub-module is a core component of the AI digital human module, responsible for processing and analyzing the natural language input by users. This module includes a speech recognition unit, a semantic understanding unit, an intention recognition unit, and a slot filling unit. The speech recognition unit uses deep neural network technology to convert the user's speech input into text; the semantic understanding unit understands the meaning of the text through means such as syntactic analysis and semantic role labeling; the intention recognition unit determines which type of predefined intention the user's input belongs to; the slot filling unit extracts key information from the user input and fills it into predefined slots, providing necessary data support for subsequent dialogue generation.

[0059] The knowledge graph sub-module is used to store and manage the knowledge information required by the AI digital human to support intelligent question answering and reasoning functions. This module includes a knowledge extraction unit, a knowledge storage unit, a knowledge query unit, and a knowledge reasoning unit. The knowledge extraction unit extracts structured and unstructured knowledge from various data sources; the knowledge storage unit uses technologies such as graph databases to store knowledge and form a knowledge graph; the knowledge query unit provides an efficient query interface for quickly retrieving relevant knowledge; the knowledge reasoning unit uses a rule engine and machine learning algorithms to perform reasoning on the knowledge graph, generate new knowledge or verify existing knowledge, and enhance the intelligence level of the AI digital human.

[0060] The knowledge graph sub-module combines a rule engine and machine learning algorithms to perform reasoning on the knowledge graph, aiming to generate new knowledge or verify existing knowledge.

[0061] The specific algorithm content includes:

[0062] Rule engine part:

[0063] Rule definition: Define a series of reasoning rules, such as "if A is a subclass of B and B has attribute C, then A also has attribute C".

[0064] Rule application: Apply the rules to the entities and relationships in the knowledge graph to generate new knowledge or verify existing knowledge.

[0065] Machine learning algorithm part:

[0066] Algorithm selection: Select the support vector machine (SVM) as the reasoning algorithm, which is suitable for classification and regression tasks.

[0067] Feature extraction: Extract features from the knowledge graph, such as entity attributes, relationship types, etc.

[0068] Model training: Use known knowledge to train the SVM model to learn the latent patterns in the knowledge graph.

[0069] Inference application: Use the trained model to perform inference on unknown knowledge to generate new knowledge or verify existing knowledge.

[0070] The training formula of the SVM model is as follows:

[0071] minimize 1 / 2||w||^2 + C * Σξ_i

[0072] subject to y_i * (w·x_i + b) ≥ 1 - ξ_i, ξ_i ≥ 0

[0073] The inference formula of the SVM is as follows:

[0074] f(x) = sign(w·x + b)

[0075] Among them, w is the weight vector, representing the importance of the features learned by the model; b is the bias term, representing the threshold of the model; C is the regularization parameter, controlling the balance between the model complexity and the training error; ξ_i is the slack variable, allowing the model to misclassify some samples during training; y_i is the sample label, representing the positive class or the negative class; x_i is the feature vector, representing the entity or relationship features in the knowledge graph; sign is the sign function, converting the model output into the classification result.

[0076] The dialogue generation sub-module is responsible for generating natural and fluent dialogue responses according to the user input and context information. This module includes a dialogue management unit, a language generation unit, and a response selection unit. The dialogue management unit maintains the dialogue state, tracks the dialogue history and the user intention; the language generation unit uses natural language generation techniques, such as the sequence-to-sequence model, to generate candidate responses; the response selection unit then selects the best response from the candidate responses according to indicators such as context relevance and language fluency. In addition, the dialogue generation sub-module also considers emotional factors to make the response more user-friendly;

[0077] The speech synthesis sub-module converts the generated text response into natural and realistic speech output. This module includes a text analysis unit, a prosody model unit, a vocoder unit, and a speech synthesis engine. The text analysis unit performs linguistic analysis on the text, extracting information such as pronunciation, stress, and intonation; the prosody model unit generates prosody parameters, such as speech rate, pitch, and volume, according to the text analysis results; the vocoder unit converts the prosody parameters into a speech waveform; the speech synthesis engine then integrates the outputs of the above units to generate the final speech signal. This module also supports multiple speech styles and dialects to meet the needs of different users;

[0078] The image display sub-module is responsible for the visual presentation of the AI digital human, including the image modeling unit, motion generation unit, expression generation unit, and rendering output unit. The image modeling unit creates a 3D model of the AI digital human according to the design requirements; the motion generation unit generates natural human motions using motion capture or animation techniques; the expression generation unit generates corresponding facial expressions based on the conversation content and sentiment analysis results; the rendering output unit integrates the model, motions, and expressions and displays them on the user interface through real-time rendering technology. This module also supports multi-modal interactions such as gesture recognition and eye tracking to enhance the interaction experience between the user and the AI digital human.

[0079] The internal settings of the interaction display module include multiple parts such as user interface design, interaction logic processing, data visualization display, user input processing, and feedback mechanism. The user interface design is responsible for designing an intuitive and user-friendly interface, including layout, color, icons, etc.; the interaction logic processing implements the interaction logic between the user and the system, such as click, drag, zoom, etc.; the data visualization display presents data in the form of charts, graphs, dashboards, etc., supporting multiple visualization methods; the user input processing receives and processes user inputs such as query conditions, parameter settings, etc.; the feedback mechanism provides user operation feedback such as prompt messages, loading animations, etc. to enhance the user experience. Through the collaborative work of these parts, the interaction display module realizes a natural and intuitive interaction between the user and the system, improving the user's usage experience.

[0080] The internal settings of the system management module cover multiple sub-modules such as user management, permission management, system monitoring, log management, backup and recovery, and configuration management. User management is responsible for managing the information and permissions of system users, including user registration, login, role allocation and permission setting, to ensure data security and privacy protection; permission management implements permission allocation and control for different user roles, defines the user's access level to system resources, functions and data, and ensures the safe operation of the system; system monitoring monitors the operating status of the system in real time, including server load, database performance, network status, etc., and promptly discovers and handles abnormalities by setting thresholds and alarm mechanisms to ensure stable operation of the system; log management records the operation and change history of the system, including user operation logs, system operation logs and error logs, providing a basis for system troubleshooting, performance analysis and security auditing; backup and recovery implements the system's data backup and recovery functions, supports scheduled backup, manual backup and disaster recovery, ensures data is not lost, and ensures business continuity; configuration management provides a system configuration interface, allowing administrators to set and adjust system parameters, including database connection configuration, service port configuration, mail server configuration, etc., to adapt to different operating environments and business needs. Through the comprehensive management of these sub-modules, the system management module ensures the stable operation, safe controllability and flexible configuration of the system, providing a solid foundation and strong support for the entire intelligent data processing system.

[0081] In the present invention, the knowledge graph submodule performs deep reasoning on the knowledge graph by combining the rule engine and the machine learning algorithm, which brings significant benefits to the entire intelligent data processing system. First, the introduction of the rule engine enables the system to automatically generate new knowledge or verify existing knowledge based on predefined reasoning rules, thereby improving the accuracy and efficiency of knowledge management. The system can quickly infer new attribute relationships and enrich the content of the knowledge base. Through the application of support vector machines (SVM), the autonomous learning and reasoning capabilities of the system are enhanced. The SVM model can accurately reason about unknown knowledge, generate new knowledge or verify existing knowledge by training and learning the potential patterns in the knowledge graph, thereby improving the intelligent level of data analysis and decision-making. In addition, this combination method also improves the system's ability to handle complex knowledge relationships, enables the system to better cope with diverse business scenarios, and provides users with more accurate and comprehensive information services. These algorithm contents significantly enhance the system's knowledge processing capabilities, improve data value, and provide strong support for the intelligent upgrade and business expansion of the system.

[0082] In the present invention, the AI digital human module realizes highly biomimetic interaction with users through advanced algorithms such as natural language processing, speech recognition, and image recognition. This interaction method not only greatly improves the user experience, enabling users to communicate with the system in a more natural and intuitive way, but also significantly improves the efficiency of data query and obtaining analysis results. Users can quickly retrieve specific data through voice commands or intuitively understand complex data analysis results through the visualization interface provided by the digital human. In addition, the AI digital human can also provide personalized services based on the user's historical behavior and preferences, further optimizing the interaction process, reducing the user's operation cost, and thus improving the overall work efficiency.

[0083] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising a..." does not exclude the existence of additional identical elements in the process, method, article or device comprising the element.

[0084] The above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. An intelligent data processing system equipped with AI digital human, characterized by: The system includes: a data source management module, a data verification module, a data capture module, a data analysis module, an AI digital human module, an interactive display module and a system management module; The AI ​​digital human module is internally provided with a natural language processing submodule, a knowledge graph submodule, a dialogue generation submodule, a speech synthesis submodule and an image display submodule; The output end of the data source management module is connected to the input end of the data verification module through a data interface; The output end of the data verification module is connected to the input end of the data capture module through a data stream; The output end of the data capture module is connected to the input end of the data analysis module through a data stream; The output end of the data analysis module is connected to the input end of the AI ​​digital human module and the interactive display module through a data interface or direct transmission; The interactive interface of the AI ​​digital human module is connected to the input end of the interactive display module; The management interface of the system management module is respectively connected to the management ends of the data source management module, the data verification module, the data capture module, the data analysis module, the AI ​​digital human module and the interactive display module.

2. The intelligent data processing system equipped with an AI digital human as claimed in claim 1, characterized in that: The data source management module includes data connectors, data extraction engines, data source monitoring, data source configuration management, data source authority control, data source cache mechanism and data source metadata management components; The internal settings of the data verification module include a data format verification component, a data integrity check component, a data consistency verification component, an outlier detection and processing component, and a data quality report generation component.

3. The intelligent data processing system equipped with an AI digital human as claimed in claim 1, characterized in that: The internal configuration of the data capture module includes a web crawler submodule, an API caller submodule, a data parser submodule, a capture task scheduler submodule, a capture result storage submodule and a capture log recorder submodule.

4. The intelligent data processing system equipped with an AI digital human as claimed in claim 1, characterized in that: The internal configuration of the data analysis module includes a data preprocessing submodule, a statistical analysis submodule, a data mining submodule, a model training submodule, a prediction analysis submodule and a result visualization submodule.

5. The intelligent data processing system equipped with an AI digital human as claimed in claim 1, characterized in that: The natural language processing submodule includes a speech recognition unit, a semantic understanding unit, an intention recognition unit and a slot filling unit.

6. The intelligent data processing system equipped with an AI digital human as claimed in claim 1, characterized in that: The knowledge graph submodule includes a knowledge extraction unit, a knowledge storage unit, a knowledge query unit and a knowledge reasoning unit; The SVM model training formula of the knowledge graph submodule is: minimize1 / 2||w||^2+C*Σξ_i subjecttoy_i*(w·x_i+b)≥1-ξ_i,ξ_i≥0 The SVM inference formula is: f(x)=sign(w·x+b) Where w is the weight vector, which indicates the importance of the features learned by the model; b is the bias term, which indicates the threshold of the model; C is the regularization parameter, which controls the balance between model complexity and training error; ξ_i is a slack variable, which allows the model to misclassify some samples during training; y_i is the sample label, which indicates positive or negative class; x_i is the feature vector, which indicates the entity or relationship feature in the knowledge graph; sign is the symbolic function, which converts the model output into the classification result.

7. The intelligent data processing system equipped with an AI digital human as claimed in claim 1, characterized in that: The dialogue generation submodule includes a dialogue management unit, a language generation unit and a response selection unit; The speech synthesis submodule includes a text analysis unit, a prosody model unit, a vocoder unit and a speech synthesis engine.

8. The intelligent data processing system equipped with an AI digital human as claimed in claim 1, characterized in that: The image display submodule includes an image modeling unit, an action generating unit, an expression generating unit and a rendering output unit.

9. The intelligent data processing system equipped with an AI digital human as claimed in claim 1, characterized in that: The internal configuration of the interactive display module includes a user interface design submodule, an interactive logic processing submodule, a data visualization display submodule, a user input processing submodule and a feedback mechanism submodule.

10. The intelligent data processing system equipped with an AI digital human as claimed in claim 1, characterized in that: The system management module is internally provided with a user management submodule, a permission management submodule, a system monitoring submodule, a log management submodule, a backup and recovery submodule and a configuration management submodule.

Images