Intelligent management assistant system and method based on large model

Abstract

The invention relates to the technical field of computer application, in particular to an intelligent management assistant system and method based on a large model, and the system comprises a user interaction module, a large model core processing module, a task analysis module, a function execution module and a knowledge base module. The method has the beneficial effects that an intelligent assistant system integrated with multiple management functions is constructed, and the powerful natural language understanding and processing capability of a large model is utilized, so that accurate analysis and efficient processing of a user instruction are realized; one-stop services such as weather query, schedule management, work order management, answer based on a knowledge base and answer based on networking search results can be provided for the user, the management efficiency and experience of the user are improved, and the requirements of the user for intelligent and convenient management are met.

Description

Technical Field

[0001] The present invention relates to the field of computer application technology, and in particular to a large model-based intelligent management assistant system and method. Background Art

[0002] In the existing technology, there are a variety of different tools and systems for daily management for individuals and businesses. For example, there are dedicated weather apps for weather queries, calendar software for schedule management, internal work order systems for work order management, knowledge Q&A systems that rely on specific knowledge base systems, and online searches that rely on search engines. However, these tools are often independent, requiring users to switch between different applications, which is cumbersome and inefficient.

[0003] Furthermore, traditional management tools lack intelligent interaction capabilities. Often, they can only accept simple command inputs and have limited understanding of natural language, making it difficult to accurately grasp users' true needs. For example, when managing their schedule, users might describe tasks in natural language, such as "I'm meeting a client at a café next Wednesday at 3:00 PM. Remind me to bring the contract." Traditional systems may not be able to accurately interpret key information such as time, location, and event.

[0004] Furthermore, while some existing integrated management tools attempt to integrate multiple functions, they lack robust large-scale model support and therefore perform poorly when handling complex tasks and multi-turn conversations. For answers based on knowledge bases, they lack a good understanding of the connections and context between knowledge, resulting in inaccurate and incomplete responses. For answers based on online search results, they also fail to efficiently filter and integrate information to provide valuable content to users.

[0005] In summary, existing technologies suffer from the following main problems: fragmented functionality and inconvenient user operation; insufficiently intelligent interaction and limited natural language understanding and processing capabilities; and insufficient capabilities for complex task processing and information integration. These issues make it difficult for existing management tools to meet users' growing demand for intelligent and convenient management. Therefore, a large-scale model-based intelligent management assistant system and method are urgently needed to address these issues. Summary of the Invention

[0006] The purpose of the present invention is to provide an intelligent management assistant system and method based on a large model to solve the problems raised in the above background technology.

[0007] To achieve the above objectives, the present invention provides the following technical solutions: an intelligent management assistant system based on a large model, comprising:

[0008] The user interaction module, serving as the interface for user interaction with the system, is used to receive user input commands in various forms, including text input, voice input, and gesture input, and to provide feedback to the user based on the system processing results. For voice input, speech recognition technology is used to convert the speech into text and then pass it to subsequent modules. When providing feedback, speech synthesis technology can be used to output the results in speech form.

[0009] The large model core processing module, based on a pre-trained large language model, receives user instructions from the user interaction module, performs semantic analysis on the instructions, understands user intent and needs, identifies key information in the instructions, including task type and specific content, and maintains contextual information in multiple rounds of dialogue to ensure accurate understanding of user intent;

[0010] The task parsing module decomposes user instructions into specific executable tasks based on the user intentions and key information analyzed by the core processing module of the large model, determines the execution order of the tasks and the required functional modules, and provides clear task instructions for subsequent functional execution modules.

[0011] Preferably, it further includes: a function execution module, which includes multiple functional units corresponding to different management functions, specifically including:

[0012] The weather query unit is connected to the external weather data interface, obtains real-time weather data based on the geographic location and time information provided by the task analysis module, and returns the data to the core processing module of the large model for processing and generating answers;

[0013] The schedule management unit is responsible for managing the user's schedule. It can create, modify, and delete schedule events, store schedule management task information in the schedule table in the data storage module, and send reminder notifications to users through the user interaction module according to the set reminder time, including pop-up reminders, SMS reminders, and email reminders;

[0014] The work order management unit is used to handle work order tasks for enterprises or individuals. It supports the creation, assignment, tracking and processing status updates of work orders. It connects with the enterprise's internal work order system or external work order management platform to realize the full process management of work orders. When creating a work order, it automatically fills in the relevant fields of the work order based on the information provided by the task parsing module.

[0015] The knowledge question-answering unit is connected to the knowledge base module. It searches and matches the knowledge base based on user questions. It supports keyword search and semantic search, obtains relevant knowledge content, and returns the answers to the core processing module of the large model.

[0016] The network search unit receives instructions from the task analysis module, calls the network search module to obtain and process relevant information, and returns the processed information to the large model core processing module.

[0017] Preferably, it also includes: a knowledge base module for storing and managing knowledge and information related to various management functions, including answers to frequently asked questions, business process descriptions, and industry knowledge; the data in the knowledge base is updated and maintained through manual entry and data import, and the knowledge is organized in a combination of structured and unstructured methods to facilitate knowledge retrieval and utilization.

[0018] Preferably, the network search module includes:

[0019] The search request generation module generates reasonable search keywords and search statements based on the user's questions and the information provided by the task analysis module;

[0020] Search result acquisition module, which obtains search results by calling the search engine API;

[0021] The search result processing module cleans, filters and integrates the obtained search results, removes irrelevant information, retains valuable content, and converts it into a format suitable for system processing.

[0022] Preferably, it also includes: a data storage module, which uses database technology for data management and storage, and is used to store various data generated during the operation of the system, including user information, schedule data, work order data, knowledge base data, and search history data, to ensure the security, integrity and accessibility of the data, and support data backup and recovery to deal with possible data loss problems.

[0023] A method for an intelligent management assistant system based on a large model, comprising the following steps:

[0024] User command input step: The user inputs a command through the user interaction module, and the command form is text, voice or other forms; if the command is in voice form, the user interaction module converts it into text and passes it to the subsequent processing module;

[0025] Core processing steps of the large model: The core processing module of the large model performs natural language processing operations such as word segmentation, part-of-speech tagging, and syntactic analysis on user instructions, extracts key information from the instructions, including task type, time, location, and object, and combines contextual information to understand the user's true intentions.

[0026] Preferably, after the large model core processing step, the following steps are also included: task parsing step: the task parsing module decomposes the instructions into specific tasks according to the user intention and key information output by the large model core processing module; for the schedule management task, the event name, time, and reminder setting information are parsed; for the work order management task, the work order type, problem description, and priority information are parsed; at the same time, the functional unit that needs to be called for the task is determined, such as the schedule management task calls the schedule management unit, and the weather query task calls the weather query unit.

[0027] Preferably, after the task analysis step, the following steps are further included:

[0028] Function execution step: The function execution module calls the corresponding functional unit to process the task according to the instructions of the task analysis module;

[0029] For weather query tasks, the weather query unit sends a request to the external weather data interface to obtain weather information at the specified location and time, and returns the information to the core processing module of the large model;

[0030] For the schedule management task, the schedule management unit stores the schedule events in the data storage module and sets the reminder time. When the reminder time arrives, the user interaction module sends a reminder to the user;

[0031] For work order management tasks, the work order management unit connects with the work order system to create or update work order information and track the processing status of work orders;

[0032] For answering tasks based on the knowledge base, the knowledge question answering unit retrieves relevant knowledge in the knowledge base module and returns the answer to the core processing module of the large model;

[0033] For answering tasks based on online search results, the online search unit calls the online search module, the online search module generates a search request, obtains and processes the search results, and returns useful information to the online search unit, which then passes it to the large model core processing module.

[0034] Preferably, after the function execution step, the following steps are also included: result generation and feedback step: the large model core processing module receives the processing results returned by the function execution module, and generates appropriate answers or feedback information according to the user's needs and interaction scenarios; for example, for weather query results, a description of the weather conditions is generated; for schedule management tasks, confirmation information is generated; then, the generated results are passed to the user interaction module and fed back to the user through text display and voice playback.

[0035] Preferably, after the result generation and feedback steps, the following steps are also included: data storage and update steps: during the entire processing process, the data storage module stores relevant data, including the user's input history, schedule data, and work order data; at the same time, the new information obtained by the knowledge base module and the network search module is updated to the data storage module as needed to enrich the system's knowledge and data.

[0036] Compared with the prior art, the present invention has the following beneficial effects:

[0037] The intelligent management assistant system and method based on the big model proposed in the present invention, by building an intelligent assistant system integrating multiple management functions and utilizing the powerful natural language understanding and processing capabilities of the big model, can achieve accurate parsing and efficient processing of user instructions, and can provide users with one-stop services such as weather query, schedule management, work order management, answers based on the knowledge base and answers based on network search results, thereby improving users' management efficiency and experience and meeting users' needs for intelligent and convenient management. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 This is a system block diagram of the present invention. DETAILED DESCRIPTION

[0039] In order to clearly and completely describe the objectives and technical solutions of the present invention and make the advantages more clearly understood, the embodiments of the present invention are further described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are part of the embodiments of the present invention, not all of them, and are only used to explain the embodiments of the present invention, not to limit the embodiments of the present invention. All other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0040] For example 1, please refer to Figure 1 The present invention provides a technical solution: an intelligent management assistant system based on a large model, comprising:

[0041] 1. User interaction module

[0042] The user interaction module is the interface for users to interact with the system. It is responsible for receiving user input commands and providing feedback on the system's processing results. Input methods can include text, voice, gesture input, and other methods to meet different user habits. For voice input, this module first converts the speech into text using speech recognition technology, and then passes the text to the core processing module of the large model. When providing feedback, in addition to text display, speech synthesis technology can also be used to output the results in speech form, achieving a more natural interactive experience.

[0043] 2. Large model core processing module

[0044] The large model core processing module is the core of the entire system. It is based on a pre-trained large language model and has powerful natural language understanding, generation and reasoning capabilities. It receives user instructions from the user interaction module, performs semantic analysis on the instructions, and understands the user's intentions and needs. Specifically, the module can identify key information in the instructions, such as task type (weather query, schedule management, etc.) and specific content (time, place, object, etc.). At the same time, in multiple rounds of dialogue, it can maintain contextual information to ensure accurate understanding of user intentions. For example, the user first asks "What's the weather like tomorrow?" and then says "What clothes are suitable to wear?" The large model core processing module can combine the context to clarify that the user's question is based on the weather conditions of the previous day and asks for clothing recommendations.

[0045] 3. Task parsing module

[0046] The task parsing module decomposes user instructions into specific, executable tasks based on the user intent and key information analyzed by the core processing module of the large model. For example, for a calendar management instruction, the task parsing module extracts information such as the event name, time, location, and participants, and generates the corresponding calendar task. For a work order management instruction, it parses information such as the work order type, problem description, and priority to form a work order task. This module also determines the execution order of the tasks and the required functional modules, providing clear task instructions to the functional execution module.

[0047] 4. Function execution module

[0048] The function execution module contains multiple functional units, each corresponding to a different management function, such as weather query unit, schedule management unit, work order management unit, knowledge question and answer unit, and network search unit.

[0049] Weather query unit: connected to the external weather data interface, obtains real-time weather data, including temperature, humidity, weather conditions, etc., based on the geographic location and time information provided by the task analysis module, and returns the data to the large model core processing module for processing and generating answers.

[0050] The Schedule Management Unit is responsible for managing the user's schedule and can create, modify, and delete schedule events. When a schedule management task is received, the unit stores the task information in the schedule table in the data storage module and sends a reminder notification to the user through the user interaction module based on the set reminder time. Reminders can include pop-up notifications, SMS reminders, and email reminders.

[0051] The Work Order Management Unit handles work order tasks for both enterprises and individuals, supporting the creation, assignment, tracking, and status updates of work orders. This unit integrates with the enterprise's internal work order system or an external work order management platform to enable comprehensive work order management. When creating a work order, relevant fields such as the problem description, priority, and responsible person are automatically populated based on information provided by the Task Parsing Module.

[0052] The Knowledge Question and Answer unit is connected to the knowledge base module. Based on the user's question, it searches and matches the knowledge base, obtains relevant knowledge content, and returns the answer to the core processing module of the large model. The Knowledge Question and Answer unit supports multiple search methods such as keyword search and semantic search, and can accurately find knowledge related to the user's question.

[0053] Network search unit: receives instructions from the task parsing module, calls the network search module to obtain and process relevant information, and returns the processed information to the large model core processing module.

[0054] 5. Knowledge Base Module

[0055] The knowledge base module is used to store and manage knowledge and information related to various management functions, including FAQs, business process descriptions, and industry knowledge. Data in the knowledge base can be updated and maintained through manual entry and data import. Knowledge is organized using a combination of structured and unstructured methods to facilitate retrieval and utilization. For example, business process descriptions can be stored using a combination of flowcharts and text descriptions, while FAQs can be stored using question-answer pairs.

[0056] 6. Online search module

[0057] The online search module is primarily responsible for interacting with external search engines to obtain information from the internet. It includes a search request generation module, a search result acquisition module, and a search result processing module. The search request generation module generates appropriate search keywords and search statements based on the user's question and information provided by the task analysis module. The search result acquisition module retrieves search results by calling the search engine's API. The search result processing module cleans, filters, and integrates the obtained search results, removing irrelevant information while retaining valuable content. The module then converts the results into a format suitable for system processing.

[0058] 7. Data storage module

[0059] The data storage module is used to store various data generated during system operation, including user information, schedule data, work order data, knowledge base data, and search history data. Database technology is used to manage and store data, ensuring data security, integrity, and accessibility. Furthermore, the data storage module supports data backup and recovery to address potential data loss.

[0060] In the second embodiment, based on the first embodiment, a method for an intelligent management assistant system based on a large model is proposed, comprising the following steps:

[0061] 1. User command input

[0062] The user enters instructions through the user interaction module, which can be in text form, voice form or other forms. If the instruction is in voice form, the user interaction module first converts it into text and then passes the text to the large model core processing module.

[0063] 2. Large model core processing

[0064] The core processing module of the large model processes user commands, first performing natural language processing operations such as word segmentation, part-of-speech tagging, and syntactic analysis to extract key information from the command, such as the task type, time, location, and object. It then combines contextual information (if there are multiple rounds of dialogue) to understand the user's true intent. For example, if a user enters "Remind me for a meeting tomorrow morning at 8:00 AM," the core processing module of the large model can identify the task type as calendar management, the time as tomorrow morning at 8:00 AM, and the event as a meeting.

[0065] 3. Task Analysis

[0066] The task parsing module breaks down instructions into specific tasks based on the user intent and key information output by the core processing module of the large model. For calendar management tasks, this module parses information such as the event name, time, and reminder settings; for work order management tasks, this module parses information such as the work order type, problem description, and priority. It also determines the functional units that the task will call, such as the calendar management unit for a calendar management task and the weather query unit for a weather query task.

[0067] 4. Function Execution

[0068] The function execution module calls the corresponding functional unit to process the task according to the instructions of the task analysis module.

[0069] For weather query tasks, the weather query unit sends a request to the external weather data interface to obtain weather information at the specified location and time, and returns the information to the core processing module of the large model.

[0070] For the schedule management task, the schedule management unit stores the schedule events in the data storage module and sets the reminder time. When the reminder time arrives, the user interaction module sends a reminder to the user.

[0071] For work order management tasks, the work order management unit connects to the work order system, creates or updates work order information, and tracks the processing status of work orders.

[0072] For answering tasks based on the knowledge base, the knowledge question answering unit retrieves relevant knowledge in the knowledge base module and returns the answer to the core processing module of the large model.

[0073] For answering tasks based on online search results, the online search unit calls the online search module, the online search module generates a search request, obtains and processes the search results, and returns useful information to the online search unit, which then passes it to the large model core processing module.

[0074] 5. Result Generation and Feedback

[0075] The core processing module of the large model receives the processing results returned by the function execution module and generates appropriate responses or feedback based on the user's needs and interaction scenario. For example, for weather query results, a description of the weather conditions is generated; for schedule management tasks, a confirmation message is generated. The generated results are then passed to the user interaction module, which provides feedback to the user through text display, voice playback, and other means.

[0076] 6. Data Storage and Update

[0077] Throughout the entire processing process, the data storage module stores relevant data, such as user input history, schedule data, work order data, etc. At the same time, new information obtained by the knowledge base module and the network search module can also be updated to the data storage module as needed to enrich the system's knowledge and data.

[0078] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent management assistant system based on a large model, characterized by: include: The user interaction module, serving as the interface for user interaction with the system, is used to receive user input commands in various forms, including text input, voice input, and gesture input, and to provide feedback to the user based on the system processing results. For voice input, speech recognition technology is used to convert the speech into text and then pass it to subsequent modules. When providing feedback, speech synthesis technology can be used to output the results in speech form. The large model core processing module, based on a pre-trained large language model, receives user instructions from the user interaction module, performs semantic analysis on the instructions, understands user intent and needs, identifies key information in the instructions, including task type and specific content, and maintains contextual information in multiple rounds of dialogue to ensure accurate understanding of user intent; The task parsing module decomposes user instructions into specific executable tasks based on the user intentions and key information analyzed by the core processing module of the large model, determines the execution order of the tasks and the required functional modules, and provides clear task instructions for subsequent functional execution modules.

2. The intelligent management assistant system based on a large model according to claim 1, characterized in that: Also includes: The function execution module contains multiple functional units, each corresponding to a different management function, including: The weather query unit is connected to the external weather data interface, obtains real-time weather data based on the geographic location and time information provided by the task analysis module, and returns the data to the core processing module of the large model for processing and generating answers; The schedule management unit is responsible for managing the user's schedule. It can create, modify, and delete schedule events, store schedule management task information in the schedule table in the data storage module, and send reminder notifications to users through the user interaction module according to the set reminder time, including pop-up reminders, SMS reminders, and email reminders; The work order management unit is used to handle work order tasks for enterprises or individuals. It supports the creation, assignment, tracking and processing status updates of work orders. It connects with the enterprise's internal work order system or external work order management platform to realize the full process management of work orders. When creating a work order, it automatically fills in the relevant fields of the work order based on the information provided by the task parsing module. The knowledge question-answering unit is connected to the knowledge base module. It searches and matches the knowledge base based on user questions. It supports keyword search and semantic search, obtains relevant knowledge content, and returns the answers to the core processing module of the large model. The network search unit receives instructions from the task analysis module, calls the network search module to obtain and process relevant information, and returns the processed information to the large model core processing module.

3. The intelligent management assistant system based on a large model according to claim 2, characterized in that: Also includes: Knowledge base module, used to store and manage knowledge and information related to various management functions, including FAQs, business process descriptions, and industry knowledge; The data in the knowledge base is updated and maintained through manual entry and data import. The knowledge is organized in a combination of structured and unstructured methods to facilitate knowledge retrieval and utilization.

4. The intelligent management assistant system based on a large model according to claim 3, characterized in that: The network search module includes: The search request generation module generates reasonable search keywords and search statements based on the user's questions and the information provided by the task analysis module; Search result acquisition module, which obtains search results by calling the search engine API; The search result processing module cleans, filters and integrates the obtained search results, removes irrelevant information, retains valuable content, and converts it into a format suitable for system processing.

5. The intelligent management assistant system based on a large model according to claim 1, characterized in that: Also includes: The data storage module uses database technology for data management and storage. It is used to store various data generated during the operation of the system, including user information, schedule data, work order data, knowledge base data, and search history data, to ensure the security, integrity, and accessibility of the data. It also supports data backup and recovery to deal with possible data loss problems.

6. A method for a large model-based intelligent management assistant system according to claim 5, characterized in that: The following steps are involved: User command input step: The user inputs a command through the user interaction module, and the command form is text, voice or other forms; if the command is in voice form, the user interaction module converts it into text and passes it to the subsequent processing module; Core processing steps of the large model: The core processing module of the large model performs natural language processing operations such as word segmentation, part-of-speech tagging, and syntactic analysis on user instructions, extracts key information from the instructions, including task type, time, location, and object, and combines contextual information to understand the user's true intentions.

7. A method according to claim 6, characterized in that: After the core processing step of the large model, the following steps are also included: Task parsing step: The task parsing module decomposes the instructions into specific tasks based on the user intention and key information output by the core processing module of the large model; for the schedule management task, the event name, time, and reminder setting information are parsed; for the work order management task, the work order type, problem description, and priority information are parsed; at the same time, the functional unit that needs to be called for the task is determined, such as the schedule management task calls the schedule management unit, and the weather query task calls the weather query unit.

8. A method according to claim 7, characterized in that: After the task parsing step, the following steps are also included: Function execution step: The function execution module calls the corresponding functional unit to process the task according to the instructions of the task analysis module; For weather query tasks, the weather query unit sends a request to the external weather data interface to obtain weather information at the specified location and time, and returns the information to the core processing module of the large model; For the schedule management task, the schedule management unit stores the schedule events in the data storage module and sets the reminder time. When the reminder time arrives, the user interaction module sends a reminder to the user; For work order management tasks, the work order management unit connects with the work order system to create or update work order information and track the processing status of work orders; For answering tasks based on the knowledge base, the knowledge question answering unit retrieves relevant knowledge in the knowledge base module and returns the answer to the core processing module of the large model; For answering tasks based on online search results, the online search unit calls the online search module, the online search module generates a search request, obtains and processes the search results, and returns useful information to the online search unit, which then passes it to the large model core processing module.

9. A method according to claim 8, characterized in that: After the function execution step, the following steps are also included: result generation and feedback step: the large model core processing module receives the processing results returned by the function execution module, and generates appropriate answers or feedback information based on the user's needs and interaction scenarios; for example, for weather query results, a description of the weather conditions is generated; for schedule management tasks, confirmation information is generated; then, the generated results are passed to the user interaction module and fed back to the user through text display and voice playback.

10. A method according to claim 9, characterized in that: After the result generation and feedback steps, the following steps are also included: Data storage and update steps: During the entire processing process, the data storage module stores relevant data, including user input history, schedule data, and work order data; at the same time, the new information obtained by the knowledge base module and the network search module is updated to the data storage module as needed to enrich the system's knowledge and data.

Images