An office collaboration system and method based on artificial intelligence

By using AI-based collaborative office systems and technologies such as natural language processing, machine learning, and deep learning, intelligent information analysis and decision support are achieved, solving the problem of low efficiency in existing collaborative office systems and improving the operational efficiency and innovation capabilities of enterprises.

CN118521272BActive Publication Date: 2024-11-15深圳市裕熙科技有限公司
View PDF 0 Cites 2 Cited by

Patent Information

Application Number
CN202410971368.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-07-19
Publication Date
2024-11-15
Estimated Expiration
2044-07-19

AI Technical Summary

Technical Problem

Existing collaborative office systems are not intelligent enough, are inefficient, and cannot meet enterprises' needs for efficient, low-cost, and intelligent management.

Method used

An AI-based collaborative office system is adopted, which utilizes technologies such as natural language processing, machine learning, deep learning, and predictive analytics through cloud servers, IoT servers, and control terminals to achieve real-time information analysis, extraction and push of key information, decision support, project risk prediction, and project display in a virtual environment.

Benefits of technology

It improved information processing efficiency, enhanced decision support capabilities, optimized resource allocation, promoted knowledge sharing and innovation, improved collaboration quality, realized an intelligent office environment, and improved enterprise operational efficiency and innovation capabilities.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118521272B_ABST
    Figure CN118521272B_ABST
Patent Text Reader

Abstract

The present invention proposes an office collaboration system and method based on artificial intelligence. By establishing an integrated intelligent collaboration platform, natural language processing, machine learning and other technologies are used to analyze documents, emails, etc., extract key information and build an enterprise knowledge graph; through an intelligent collaborative control system, task reminders, decision-making suggestions, etc. are pushed to relevant personnel; predictive analysis models are used to predict project risks, resource requirements, etc., and VR technology is used to display project progress; image recognition technology is used to automatically identify unstructured information, and natural language generation technology is used to generate report summaries. The scheme of the present invention realizes the intelligence, automation and personalization of office collaboration by integrating multiple artificial intelligence technologies, greatly improving the operational efficiency, decision-making quality and innovation ability of enterprises, not only optimizing daily work processes, but also providing strong technical support for the long-term development of enterprises.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to an office collaboration system and method based on artificial intelligence. Background Art

[0002] Enterprise collaborative office refers to a way of achieving efficient cross-departmental collaboration by integrating and sharing information, resources and work processes. As enterprises' demand for collaborative office continues to increase, its definition has been expanded to intelligent office. In addition to solving routine functions such as daily office and asset management, enterprises also need more advanced needs such as instant communication, data sharing, and mobile office. At the same time, enterprises are also looking for low-cost, high-performance, and highly integrated intelligent management platforms, and their requirements for collaborative office systems are getting higher and higher. In short, enterprise collaborative office aims to improve office efficiency, meet the needs of enterprise digital transformation, and become an important tool to promote enterprise innovation and development. However, the existing collaborative office system is not smart enough and inefficient. Summary of the invention

[0003] Based on the above problems, the present invention proposes an office collaboration system and method based on artificial intelligence. Through the scheme of the present invention, it is possible to improve information processing efficiency, enhance decision-making support capabilities, optimize resource allocation, promote knowledge sharing and innovation, improve collaboration quality, automate daily tasks, and realize an intelligent office environment.

[0004] In view of this, one aspect of the present invention proposes an office collaboration system based on artificial intelligence, comprising: a cloud server, a plurality of Internet of Things servers connected to the cloud server in communication, and a control terminal connected to the Internet of Things servers in communication; wherein,

[0005] The cloud server is configured as follows:

[0006] Establish an integrated intelligent collaboration platform to connect the information systems of various work departments within the enterprise;

[0007] Using natural language processing technology, real-time analysis is performed on documents, emails and communication records in the intelligent collaboration platform to extract key information;

[0008] Based on a machine learning algorithm, the extracted key information is classified, associated and evaluated for importance to obtain key information analysis data;

[0009] Establishing logical associations between the key information analysis data to construct an enterprise knowledge graph;

[0010] Through the intelligent collaborative control system, task reminders, decision-making suggestions and knowledge resources are pushed to relevant personnel in various work departments within the enterprise, and various work departments are controlled to collaboratively handle work tasks;

[0011] Obtain historical and current work data for each work department;

[0012] Using a predictive analysis model, based on the historical work data and the current work data, potential project risks, resource requirements, and schedule delays are predicted to obtain a project prediction result;

[0013] Based on the current work data and the project forecast results, using augmented reality technology, the project progress, team collaboration and expected project progress are displayed in a virtual environment to support remote collaboration and decision-making;

[0014] Image recognition technology based on deep learning can automatically identify and classify unstructured information such as meeting minutes and whiteboard content;

[0015] Use natural language generation technology to automatically generate meeting minutes, project reports, and work summaries.

[0016] Another aspect of the present invention provides an office collaboration method based on artificial intelligence, comprising:

[0017] Establish an integrated intelligent collaboration platform to connect the information systems of various work departments within the enterprise;

[0018] Using natural language processing technology, real-time analysis is performed on documents, emails and communication records in the intelligent collaboration platform to extract key information;

[0019] Based on a machine learning algorithm, the extracted key information is classified, associated and evaluated for importance to obtain key information analysis data;

[0020] Establishing logical associations between the key information analysis data to construct an enterprise knowledge graph;

[0021] Through the intelligent collaborative control system, task reminders, decision-making suggestions and knowledge resources are pushed to relevant personnel in various work departments within the enterprise, and various work departments are controlled to collaboratively handle work tasks;

[0022] Obtain historical and current work data for each work department;

[0023] Using a predictive analysis model, based on the historical work data and the current work data, potential project risks, resource requirements, and schedule delays are predicted to obtain a project prediction result;

[0024] Based on the current work data and the project forecast results, using augmented reality technology, the project progress, team collaboration and expected project progress are displayed in a virtual environment to support remote collaboration and decision-making;

[0025] Image recognition technology based on deep learning can automatically identify and classify unstructured information such as meeting minutes and whiteboard content;

[0026] Use natural language generation technology to automatically generate meeting minutes, project reports, and work summaries.

[0027] Optionally, the steps of pushing task reminders, decision suggestions and knowledge resources to relevant personnel of various work departments in the enterprise through the intelligent collaborative control system, and controlling various work departments to collaboratively process work tasks include:

[0028] Obtain product order data for the first product, historical design data for similar products, and market trend data for similar products;

[0029] Generate initial design data for the first product using machine learning and computer vision technology based on the product order data, the historical design data, and the market trend data;

[0030] Importing the initial design data into a cloud-based 3D collaborative design module;

[0031] The 3D collaborative design module generates a plurality of collaborative design tasks and sends the collaborative design tasks to corresponding design terminals;

[0032] Each of the design terminals generates a corresponding initial design model according to the corresponding collaborative design task;

[0033] Merging the multiple initial design models to obtain an initial product design model;

[0034] Using VR technology to create an initial product virtual model corresponding to the initial product design model;

[0035] Use natural language processing technology to automatically analyze the product technical information of the corresponding products and obtain technology development trend data;

[0036] Modifying the initial product virtual model according to the evaluation feedback of the initial product virtual model and the technology development trend data to obtain a basic product virtual model;

[0037] Using artificial intelligence to simulate and optimize the basic product virtual model to obtain a finalized product model;

[0038] Based on the product finalization model and the product technical information, a research and development knowledge graph is constructed to support cross-departmental knowledge sharing and rapid problem location;

[0039] According to the product finalization model and the product ordering data, a machine learning algorithm is used to predict demand, optimize inventory levels, and generate raw material procurement data;

[0040] According to the raw material procurement data, based on multi-dimensional data analysis, real-time evaluation and selection of the best suppliers are carried out, and raw material procurement is carried out;

[0041] Use blockchain technology to achieve full traceability of raw materials to ensure quality and compliance;

[0042] In the process of manufacturing the first product using raw materials, an optimization algorithm is used to realize intelligent scheduling of the production line to improve production efficiency;

[0043] During the manufacturing process of the first product, deep learning and computer vision technology are used to perform real-time quality inspection on the product being produced to obtain a first inspection result;

[0044] Based on the first detection result, using an intelligent collaborative robot, the manufacturing process is controlled to improve production flexibility and efficiency;

[0045] Using machine learning algorithms to generate defect prediction models based on historical data to predict product defects;

[0046] The manufactured first product is tested according to the defect prediction model.

[0047] Optionally, the step of establishing an integrated intelligent collaboration platform to connect the information systems of various work departments within the enterprise includes:

[0048] Obtain the first attribute information of the existing information system of each work department and the second attribute information of the work process;

[0049] Determine an integration target and an integration scope according to the first attribute information and the second attribute information;

[0050] Develop an integration strategy and technology roadmap based on the integration goals and integration scope;

[0051] Designing a distributed system architecture for the intelligent collaboration platform according to the integration strategy and the technical roadmap, planning a data storage and processing solution for the intelligent collaboration platform, and designing a security architecture and access control mechanism for the intelligent collaboration platform;

[0052] Formulate unified data standards for the intelligent collaboration platform, establish data mapping relationships between various information systems, and design data conversion and cleaning processes;

[0053] Develop standardized APIs for the information systems of various work departments, implement real-time data synchronization mechanisms, and develop middleware to handle communications between different information systems;

[0054] Build a unified user interface for the intelligent collaboration platform, develop a workflow engine, implement unified identity authentication and permission management, and develop data visualization and reporting functions;

[0055] Integrate natural language processing modules for intelligent search and automatic classification into the intelligent collaboration platform;

[0056] Embed machine learning algorithms for data analysis and prediction in the intelligent collaboration platform, develop intelligent recommendation systems, and integrate knowledge graph construction and management functions;

[0057] Conduct unit testing, integration testing, system testing, performance testing, and load testing on the intelligent collaboration platform to optimize system response speed and stability;

[0058] Develop data migration strategies for the intelligent collaboration platform, perform data cleansing and conversion, and migrate data in batches;

[0059] Generate user manuals and technical documentation for the intelligent collaboration platform, train users from various work departments, develop and execute a phased deployment plan;

[0060] Collect user feedback, monitor the system performance of the intelligent collaboration platform, and regularly update and optimize system functions.

[0061] Optionally, the step of using natural language processing technology to perform real-time analysis on documents, emails and communication records in the intelligent collaboration platform to extract key information includes:

[0062] Establish a data interface to obtain documents, emails and communication records in the intelligent collaboration platform in real time and integrate them into the first text;

[0063] Performing text cleaning on the first text, removing irrelevant characters and punctuation marks, and then performing text standardization to obtain a second text;

[0064] Using a word segmentation algorithm to segment the second text, and performing part-of-speech tagging to obtain a third text;

[0065] Identifying and annotating entities and professional terms and keywords in a specific field in the third text;

[0066] Using a preset text classification model to perform topic classification on the third text;

[0067] extracting a text summary of the classified third text, and identifying key sentences and keywords as key information from the text summary;

[0068] Constructing a sentiment analysis model to identify the sentiment tendency of the third text to perform sentiment evaluation and obtain a sentiment evaluation result;

[0069] Develop an intent recognition model to understand the purpose or demand behind the third text and identify user intentions such as task assignment, problem feedback, and decision requests;

[0070] Identify the relationship between entities, build the corresponding knowledge graph, and display the logical association between entities;

[0071] Annotate sentiment evaluation results, user intentions, and logical associations between entities as key information;

[0072] The annotated key information is stored in a structured database, and a full-text search engine is established for rapid retrieval;

[0073] Develop data visualization interfaces to display analysis results and implement automatic report generation to summarize key findings;

[0074] Desensitize data, protect sensitive information, establish access control mechanisms, and ensure information security.

[0075] Optionally, the step of classifying, associating and evaluating the importance of the extracted key information based on a machine learning algorithm to obtain key information analysis data includes:

[0076] After data cleaning and standardization of the key information, text features are extracted, metadata features are generated, and context features are created;

[0077] Select the corresponding classification algorithm, train the multi-label classification model, and classify the key information by topic;

[0078] Use content-based recommendation algorithms to build association models, and apply graph neural networks to analyze the associations between key information;

[0079] Develop a multi-factor scoring model that considers the novelty, relevance, and influence of key information to evaluate the importance of key information in the overall text network.

[0080] Optionally, the step of using a predictive analysis model to predict potential project risks, resource requirements, and schedule delays based on the historical work data and the current work data to obtain a project prediction result includes:

[0081] Integrate the historical working data and the current working data, perform data cleaning to handle missing values ​​and outliers, and perform feature engineering;

[0082] Use correlation analysis to identify important features from the output of the previous step, apply principal component analysis to reduce the dimensionality of the data, and use domain knowledge to select key predictors;

[0083] According to the key prediction factors, select corresponding prediction algorithms for different prediction tasks;

[0084] In combination with the prediction algorithm, develop a deep learning model for complex pattern recognition;

[0085] Training and validating the deep learning model to obtain a predictive analysis model;

[0086] Develop automated data input and preprocessing pipelines and implement model inference engines to support batch processing and real-time prediction;

[0087] Design a visual interface to display prediction results;

[0088] Develop interactive tools for model interpretation exploration, explain model decisions, and generate feature importance reports;

[0089] Design a risk scoring system to determine overall project risk and set dynamic thresholds to determine different risk levels;

[0090] The historical work data and the current work data are input into a predictive analysis model to predict potential project risks, resource requirements and schedule delays to obtain a project prediction result.

[0091] Optionally, the step of displaying the project progress, team collaboration status and expected project progress in a virtual environment using augmented reality technology based on the current work data and the project forecast results to support remote collaboration and decision-making includes:

[0092] Select an appropriate AR development platform, configure the development environment, set up version control and collaborative development tools, and build an AR system;

[0093] Developing a data API for ensuring that the AR system can access the current working data and the project forecast results in real time;

[0094] Create 3D visualization models that represent project progress, virtual models that represent team members and resources, and develop interactive charts and data visualization components in the AR system;

[0095] Create a virtual project manager in the AR system and design the user interface and interaction method;

[0096] Generate project progress data, team collaboration data and project expected progress data based on the current work data and the project forecast results;

[0097] In combination with the project progress data, the team collaboration data and the project expected progress data, a dynamic Gantt chart for displaying the project progress, a 3D network diagram for displaying the team collaboration situation, a visual model for displaying resource allocation and utilization, a risk heat map or early warning indicator for indicating risks and a visual expected progress comparison chart for displaying the expected progress of the project are created in the AR system;

[0098] Configure gesture and voice control functions, virtual sticky notes and annotation functions in AR systems, and create an operational 3D control panel;

[0099] Create multi-user synchronization modules and virtual meeting room modules in AR systems to support real-time collaboration, and create shared whiteboards and drawing tools to support remote discussions;

[0100] Develop interactive analytical tools in AR systems to create resource allocation optimizers, risk assessments, and mitigation strategy generators;

[0101] Develop scenario simulation capabilities in AR systems to display the potential impact of different decisions based on input user decisions.

[0102] Optionally, the step of automatically identifying and classifying unstructured information such as meeting minutes and whiteboard contents by using deep learning-based image recognition technology includes:

[0103] Collect historical meeting records and whiteboard images, annotate the text, charts, and formulas in the whiteboard images, and annotate the types of content;

[0104] Using a deep learning convolutional neural network model to perform feature extraction on the whiteboard image, extracting a first feature that can reflect the content of the image;

[0105] Using the first features and annotations to perform supervised learning on a deep learning model, training a first model that can automatically identify and classify image content;

[0106] The trained first model is used to automatically identify and classify new meeting records and whiteboard contents to obtain a first recognition result.

[0107] Optionally, the step of automatically generating meeting minutes, project reports and work summaries using natural language generation technology includes:

[0108] Collect historical meeting minutes texts, historical project report texts, and historical work summary texts as the first text sample;

[0109] Processing the first text sample to obtain a second text sample;

[0110] Choose the first deep learning model suitable for natural language processing;

[0111] Using the second text sample to train the first deep learning model to learn the statistical laws and structural features of the language, thereby obtaining a first report generation model;

[0112] Extracting meeting discussion records, project task records and / or work logs from the first recognition result;

[0113] The meeting discussion records, project task records and / or work logs are input into the first report generation model to obtain the meeting minutes, project reports and / or work summaries in a standard format.

[0114] By adopting the technical solution of the present invention, the artificial intelligence-based office collaboration system realizes the intelligence, automation and personalization of office collaboration by integrating multiple artificial intelligence technologies, which greatly improves the enterprise's operational efficiency, decision-making quality and innovation ability. It not only optimizes daily work processes, but also provides strong technical support for the long-term development of the enterprise. BRIEF DESCRIPTION OF THE DRAWINGS

[0115] Figure 1 is a schematic block diagram of an office collaboration system based on artificial intelligence provided by an embodiment of the present invention;

[0116] Figure 2 It is a flow chart of an office collaboration method based on artificial intelligence provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0117] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0118] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0119] The terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices.

[0120] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0121] Refer to the following Figure 1 to Figure 2 To describe an artificial intelligence-based office collaboration system and method provided according to some embodiments of the present invention.

[0122] like Figure 1 As shown, an embodiment of the present invention provides an office collaboration system based on artificial intelligence, including: a cloud server, a plurality of Internet of Things servers connected to the cloud server in communication, and a control terminal connected to the Internet of Things server in communication; wherein,

[0123] The cloud server is configured as follows:

[0124] Establish an integrated intelligent collaboration platform to connect the information systems of various work departments within the enterprise;

[0125] Using natural language processing technology, real-time analysis is performed on documents, emails and communication records in the intelligent collaboration platform to extract key information;

[0126] Based on a machine learning algorithm, the extracted key information is classified, associated and evaluated for importance to obtain key information analysis data;

[0127] Establishing logical associations between the key information analysis data to construct an enterprise knowledge graph;

[0128] Through the intelligent collaborative control system, task reminders, decision-making suggestions and knowledge resources are pushed to relevant personnel in various work departments within the enterprise, and various work departments are controlled to collaboratively handle work tasks;

[0129] Obtain historical and current work data for each work department;

[0130] Using a predictive analysis model, based on the historical work data and the current work data, potential project risks, resource requirements, and schedule delays are predicted to obtain a project prediction result;

[0131] Based on the current work data and the project forecast results, using augmented reality technology, the project progress, team collaboration and expected project progress are displayed in a virtual environment to support remote collaboration and decision-making;

[0132] Image recognition technology based on deep learning can automatically identify and classify unstructured information such as meeting minutes and whiteboard content;

[0133] Use natural language generation technology to automatically generate meeting minutes, project reports, and work summaries.

[0134] It should be known that Figure 1 The block diagram of the office collaboration system based on artificial intelligence is for illustration only, and the number of modules shown therein does not limit the protection scope of the present invention.

[0135] See also Figure 2 Another embodiment of the present invention provides an office collaboration method based on artificial intelligence, comprising:

[0136] Establish an integrated intelligent collaboration platform to connect the information systems of various work departments within the enterprise;

[0137] Using natural language processing technology, real-time analysis is performed on documents, emails and communication records in the intelligent collaboration platform to extract key information;

[0138] Based on a machine learning algorithm, the extracted key information is classified, associated and evaluated for importance to obtain key information analysis data;

[0139] Establishing logical associations between the key information analysis data to construct an enterprise knowledge graph;

[0140] In this step, you can define the core concepts and relationship types in the enterprise field, design the structure and hierarchy of the knowledge graph, and formulate unified naming rules and attribute standards; use named entity recognition (NER) technology to identify key entities, use rule matching and machine learning methods to extract entity attributes, remove duplicate entities, and unify entity representations; apply dependency syntax analysis to identify direct relationships between entities, use remote supervised learning methods to discover potential relationships, and use graph pattern matching algorithms to identify complex relationships; achieve entity alignment and merge different representations of the same entity; perform relationship alignment and unify relationship expressions from different sources; resolve knowledge conflicts and ensure knowledge consistency; implement rule-based reasoning mechanisms, apply graph neural networks for knowledge completion, and develop probabilistic reasoning models to handle uncertain knowledge; add a time dimension to the knowledge graph to achieve knowledge version control and tracking Knowledge evolution; select appropriate graph databases (such as Neo4j, JanusGraph, etc.), design efficient storage structures and indexing strategies, and implement distributed storage to support large-scale knowledge graphs; develop SPARQL-based query interfaces, implement semantic search functions, support natural language queries, and develop knowledge graph visualization tools; design evaluation indicators (such as coverage, accuracy, consistency, etc.), conduct manual sampling evaluations, and continuously optimize knowledge graphs based on evaluation results; design incremental update strategies, implement automated knowledge extraction and graph update processes, and develop knowledge change monitoring and conflict resolution mechanisms; implement fine-grained access control, design sensitive information protection mechanisms, and establish knowledge traceability and audit systems; design external system integration API interfaces, develop SDKs to support calls from various programming languages, and provide graph editing and maintenance tools.Through this step, implicit knowledge associations can be revealed and cross-domain knowledge integration can be promoted; complex knowledge reasoning and exploration can be supported to discover new insights; semantic search can be realized to improve the accuracy and efficiency of information retrieval; complex natural language question and answer can be supported to enhance user interaction experience; comprehensive knowledge background can be provided to support smarter decision making; experts and resources can be quickly located to optimize decision-making processes; structured representation and unified management of knowledge can be realized; knowledge silos can be reduced to promote knowledge sharing within the organization; cross-domain knowledge associations can be discovered to stimulate innovative thinking; knowledge-based innovation project planning and management can be supported; the associations between risk factors can be comprehensively analyzed to improve the accuracy of risk prediction; risk sources can be quickly traced to support effective risk control; personalized knowledge maps can be provided for new employees to accelerate the learning curve; and based on Adaptive learning system of knowledge graph; identify process bottlenecks by analyzing the relationship between business entities; support intelligent business process reconstruction and optimization; build a comprehensive customer knowledge graph to provide personalized services; support complex customer behavior analysis and prediction; comprehensively analyze the relationship between products, technologies, and markets to guide product innovation; optimize product life cycle management and improve product competitiveness; build a regulatory knowledge graph to achieve intelligent compliance inspection; quickly adapt to new regulations and reduce compliance risks; form a dynamically updated enterprise knowledge base to precipitate organizational wisdom; support long-term accumulation and inheritance of knowledge; provide a unified knowledge base for various information systems of the enterprise; promote data interoperability and improve data utilization efficiency; provide accurate personalized recommendations based on the association analysis of the knowledge graph; support multi-dimensional resource matching and recommendation. In this step, by building an enterprise knowledge graph, the organization can transform scattered information into a structured knowledge network, greatly improving the efficiency of knowledge management and utilization. This not only optimizes daily operations, but also provides strong support for the company's strategic decision-making, innovative development, and risk management. As the "brain" of the enterprise, the knowledge graph can continue to learn and evolve, providing key technical support for enterprises to maintain their competitiveness in the digital age.

[0141] Through the intelligent collaborative control system, task reminders, decision suggestions and knowledge resources are pushed to relevant personnel in various work departments within the enterprise (or smart terminals associated with relevant personnel), and various work departments are controlled to collaboratively handle work tasks;

[0142] In this step, you can specifically: design a distributed microservice architecture to ensure the scalability and maintainability of the system; plan data flow and control flow, and determine the interfaces between modules; design a security architecture, including identity authentication, permission management, and data encryption; develop data interfaces for various department systems, implement real-time data synchronization mechanisms, and establish a unified data warehouse and data lake; implement task analysis and priority sorting algorithms based on machine learning; develop decision support models (including predictive analysis and scenario simulation); build a knowledge recommendation system based on user portraits and context; design configurable workflow templates, implement dynamic workflow generation and adjustment mechanisms, and develop workflow monitoring and exception handling modules; design multi-channel push mechanisms (such as email, SMS, in-app notifications), implement personalized push strategies based on user preferences, and develop push effect tracking and optimization modules; develop responsive Web interfaces and Mobile applications, implement customizable dashboards and report generation functions, design intuitive task management and collaboration interfaces; develop task allocation and progress tracking systems, implement cross-departmental collaborative management functions, and design conflict detection and resolution mechanisms; connect with enterprise knowledge graphs, implement context-related knowledge recommendations, and develop knowledge contribution and evaluation mechanisms; implement caching strategies, optimize data access performance, use asynchronous processing to improve system response speed, and implement load balancing and automatic expansion mechanisms; conduct unit testing, integration testing, and system testing, implement continuous integration and continuous deployment (CI / CD) processes, and perform phased deployment and grayscale releases; implement comprehensive system monitoring and logging, develop automated operation and maintenance tools, and establish user feedback mechanisms and rapid response processes; develop user training materials and online help systems, implement phased user training plans, and establish incentive mechanisms to promote the widespread use of the system.This step can realize intelligent task allocation and priority management, reduce task delays, and improve overall work efficiency; provide data-driven decision-making suggestions and reduce human bias; support rapid response to market changes and internal needs; realize accurate knowledge push and reduce information overload; promote organizational learning, accelerate knowledge dissemination and innovation; break down departmental barriers and promote seamless collaboration; improve project execution efficiency and success rate; optimize human and resource allocation based on real-time data; improve resource utilization and reduce waste; reduce manual intervention and reduce error rate; accelerate process execution and improve business response speed; provide personalized work support and reduce work pressure; enhance employee participation and satisfaction; provide objective and comprehensive performance data; support fairer and more transparent performance evaluation; monitor business risks in real time and provide early warning; support rapid risk response and mitigation strategy formulation; promote cross-departmental knowledge sharing and creative collision, provide comprehensive resource and knowledge support for innovative projects, improve product and service quality through efficient collaboration, and accelerate response to customer needs and problem solving; automate compliance checks and reduce the risk of violations; provide comprehensive audit tracking to support regulatory requirements; support rapid organizational structure and business process adjustments to improve response speed to market changes; promote data-based decision-making and management methods. In this step, by implementing this intelligent collaborative control system, enterprises can achieve a shift from traditional hierarchical management to a networked, intelligent collaborative model. It not only improves daily operational efficiency, but also provides strong support for the company's strategic decision-making, innovative development, and risk management. As the core engine of the company's digital transformation, this system can continuously optimize and evolve, providing key technical support for the company to maintain its leading position in the fiercely competitive market.

[0143] Obtain historical and current work data for each work department;

[0144] In this step, it includes: sorting out the existing information systems and databases of various departments, evaluating data quality, format and access methods, determining key data indicators and necessary historical spans; designing standardized APIs for each data source, implementing data extraction, transformation, and loading (ETL) processes, and developing real-time data synchronization mechanisms (such as change data capture CDC); implementing data desensitization and encryption strategies, designing fine-grained data access control mechanisms, and establishing data audit tracking systems; building a centralized data warehouse for historical data, designing real-time data processing pipelines for current data, and implementing data lake architecture to support diversified data storage; formulating unified data formats and encoding standards, developing automated data cleaning processes, and implementing data quality monitoring and reporting mechanisms; establishing a comprehensive metadata warehouse, implementing data lineage analysis functions, and developing metadata search and visualization tools ; Design a phased data migration plan, perform data validation and consistency checks, and establish a rollback mechanism for data migration; implement real-time data collection based on message queues; develop edge computing nodes for distributed data preprocessing, design data buffering and exception handling mechanisms, implement cross-departmental data association and integration, resolve data conflicts and inconsistencies, and establish a master data management (MDM) system; design a unified data access interface, implement a data caching mechanism to improve access performance, and develop data service components to support microservice architecture; formulate data management policies and processes, establish a data responsibility system and a data management team, and implement a continuous data quality improvement plan; implement a distributed computing framework (such as Spark) for large-scale data processing; optimize query performance, such as using columnar storage and index optimization, and design automated data partitioning and storage expansion strategies.This step can eliminate data silos and achieve a unified view of data; improve data quality and reduce decision-making errors caused by data errors; support real-time business decisions and operations, and improve the response speed to market changes and internal conditions; support long-term trend analysis and predictive modeling, and provide historical basis and insights for strategic decision-making; promote the formation of a global perspective and avoid local optimization; support cross-departmental collaboration and optimal resource allocation; reduce the risk of data leakage; ensure compliance and meet data protection regulations; reduce data retrieval and processing time, and support large-scale concurrent data access; improve the manageability and traceability of data assets, and support the formation of a data-driven organizational culture; provide a variety of data analysis and AI Applications provide a foundation to support agile development and rapid innovation; reduce duplicate data storage and processing costs; improve data utilization and increase data value; provide comprehensive and accurate data support for decision-making, reduce decision delays, and improve decision quality; support the construction of advanced analysis and machine learning models, and improve the accuracy of business forecasts and risk assessments; simplify the compliance report generation process and provide comprehensive data audit tracking; integrate customer-related data to form a 360-degree customer view to support personalized services and precision marketing; automate data collection and report generation, reduce manual data processing work, and improve work efficiency; provide a foundation for data-driven innovation and support the rapid verification of new business models and services. In this step, by systematically acquiring and integrating historical and current work data from various departments, enterprises are able to build a strong data infrastructure. This not only improves the efficiency and accuracy of daily operations, but also provides solid data support for advanced analysis, intelligent decision-making, and innovative development. Although this process is complex and challenging, the long-term benefits it brings will significantly enhance the competitiveness and adaptability of enterprises, enabling them to thrive in a data-driven business environment.

[0145] Using a predictive analysis model, based on the historical work data and the current work data, potential project risks, resource requirements, and schedule delays are predicted to obtain a project prediction result;

[0146] Based on the current work data and the project forecast results, using augmented reality technology, the project progress, team collaboration and expected project progress are displayed in a virtual environment to support remote collaboration and decision-making;

[0147] Image recognition technology based on deep learning can automatically identify and classify unstructured information such as meeting minutes and whiteboard content;

[0148] Use natural language generation technology to automatically generate meeting minutes, project reports, and work summaries.

[0149] The technical solution of this embodiment can improve information processing efficiency, enhance decision-making support capabilities, optimize resource allocation, promote knowledge sharing and innovation, improve collaboration quality, automate daily tasks, and realize an intelligent office environment. In general, this solution integrates a variety of artificial intelligence technologies to achieve intelligent, automated, and personalized office collaboration, greatly improving the operational efficiency, decision-making quality, and innovation capabilities of the enterprise, not only optimizing daily work processes, but also providing strong technical support for the long-term development of the enterprise.

[0150] In some possible implementations of the present invention, the steps of pushing task reminders, decision suggestions and knowledge resources to relevant personnel of various work departments in the enterprise through the intelligent collaborative control system, and controlling various work departments to collaboratively process work tasks include:

[0151] Obtain product order data for the first product, historical design data for similar products, and market trend data for similar products;

[0152] In this step, the enterprise's internal ERP system or customer relationship management (CRM) system can be used to extract the product ordering data of the first product, including order quantity, order amount, delivery date and other information; the sales data of the product on the third-party e-commerce platform can be obtained through crawler technology to understand the performance of the product in different sales channels; the product ordering data can be analyzed to understand the product's sales cycle, sales peak period, sales area distribution, etc. The specific implementation steps for obtaining the historical design data of similar products can be: using 3D modeling technologies such as CAD / BIM to obtain 3D model data of similar products of the first product (such as products of different brands / models from the first product); extracting the size parameters, structural design, material selection and other information of similar products to establish a product design database; using machine learning algorithms to analyze historical design data to discover the laws and trends of product design. The specific implementation steps for obtaining market trend data of similar products can be: using Web crawler technology to collect sales information, comment information, price information, etc. of similar products on e-commerce platforms, industry websites and other channels; using natural language processing technology to analyze the collected text information to extract market insights such as product popularity, consumer preferences, and price trends; combining external data such as economic indicators and industry reports to predict the future market trends of similar products. Through the above steps, comprehensive product data can be obtained to provide valuable data support for product design, marketing strategy formulation, etc.; product design can be optimized to improve the market appeal and competitiveness of products; product demand can be accurately predicted to improve the efficiency of production and inventory management; targeted marketing strategies can be formulated to improve product sales performance; new market opportunities can be discovered and product lines can be expanded.

[0153] Generate initial design data for the first product using machine learning and computer vision technology based on the product order data, the historical design data, and the market trend data;

[0154] In this step, the collected product ordering data, historical design data and market trend data are sorted and cleaned; effective features (such as sales volume, sales amount, product size parameters, material properties, consumer evaluation, etc.) are extracted; the data is pre-processed by standardization and normalization; the sales forecasting model is trained based on the product ordering data to predict future product demand; the product design optimization model is trained using historical design data to learn the laws of product design; the consumer preference prediction model is trained using market trend data to understand future market trends; the sales forecasting, design optimization, consumer preference and other models are integrated to generate an initial product design that meets market demand; computer vision technology is used to automatically generate a draft of a three-dimensional model based on historical design samples; design parameters such as size, material, structure, etc. are optimized to make the design meet the predicted needs; the generated initial design scheme is compared and analyzed with the historical design; the market potential of the design scheme is evaluated based on indicators such as sales forecasts and consumer preferences; design iterations are performed based on the evaluation results to continuously optimize product design. Through the above steps, we can quickly generate initial product designs that meet market demand in a data-driven manner, significantly shorten the product development cycle, and increase the speed of product launch; the design plan is closer to market demand, improving the market competitiveness of the product; reducing the cost and time of design rework, improving R&D efficiency; and providing data support for subsequent product iteration and optimization. In general, this product design method based on machine learning and computer vision can help companies launch products that better meet market demand in a shorter time.

[0155] Importing the initial design data into a cloud-based 3D collaborative design module;

[0156] In this step, select a suitable cloud-based 3D design platform, such as a browser-side 3D engine based on WebGL or WebAssembly; deploy a 3D design engine on the cloud to provide 3D model rendering, interaction, editing and other functions; build a cloud-based data storage service to support the storage and management of 3D models and their associated data; convert the previously generated initial product design data into a standard 3D model format; upload the 3D model data to the cloud-based 3D design platform through an API or SDK, create a product design project on the cloud-based platform, and import the 3D model into it; provide design team members with access to the cloud-based design platform; support team members to access and edit 3D models in real time on multiple terminals such as browsers and mobile terminals; realize real-time communication and collaboration between designers, such as chat, annotation, and perspective synchronization; enable version control for 3D model data to record each modification operation; set up an approval process, and have designated design supervisors or experts review the modifications; support backtracking historical versions, comparing differences, and selecting the appropriate version for the next design step; provide cloud-based design data backup and recovery functions; support permission control and access management of design data, and integrate other enterprise systems, such as PLM and ERP, to achieve unified management of design data. Through the above steps, 3D collaborative design based on cloud technology can be realized, the collaborative efficiency of the design team can be improved, the product development progress can be accelerated, the investment cost of design tools and infrastructure can be reduced, the security and reliability of design data can be enhanced, the visual management of the entire design life cycle can be realized, and a reliable data foundation can be provided for subsequent product optimization and iteration. In short, through the cloud 3D collaborative design platform, enterprises can manage the product design process more efficiently and improve the overall efficiency of product development.

[0157] The 3D collaborative design module generates a plurality of collaborative design tasks and sends the collaborative design tasks to corresponding design terminals;

[0158] In this step, on the cloud 3D design platform, the product design project is divided into multiple subtasks; each subtask corresponds to a certain part or functional module of the product design, and the goals, requirements, deadlines, etc. of each subtask are determined, and a unique task ID and metadata information are generated for each subtask; tasks are assigned to appropriate designers based on the professional expertise and load of the design team members, and task details are sent to each design terminal through the cloud message push service, and pending collaborative tasks are displayed on the mobile APP or Web design tool; designers receive and view the assigned collaborative tasks on their own terminals; execute tasks, and design and modify 3D models on the cloud 3D design platform; after completing the task, submit the design results on the platform and update the task status; the design supervisor or expert reviews the design results of each subtask on the cloud platform; evaluates and scores the design results, and provides feedback to the designer on modification suggestions; the reviewed design results are merged into the overall product design; the progress and status of each collaborative task are tracked, reminders and processing suggestions are issued for delayed or problematic tasks, and the entire collaborative process is recorded to provide a basis for subsequent review and optimization. Through the above steps, cloud-based 3D collaborative design process management can be realized, which can improve the efficiency of division of labor and collaboration in design tasks and shorten the product development cycle; enhance the transparency and traceability of the design process, which is conducive to quality control; provide designers with the convenience of participating in collaboration on mobile terminals; promote communication and collaboration between design teams and improve innovation capabilities; and accumulate collaborative design experience and best practices for enterprises. In short, through the task management function of the cloud-based 3D collaborative design platform, enterprises can better organize and coordinate the product design process and promote continuous improvement in product development. It should be understood that in this step, tasks can also be directly assigned to the corresponding multiple design terminals based on the functions and performance of the intelligent design modules configured by each design terminal and the load conditions of the design terminal, and these design terminals automatically complete the design tasks, reducing manual participation and improving work efficiency.

[0159] Each of the design terminals generates a corresponding initial design model according to the corresponding collaborative design task;

[0160] In this step, designers receive the assigned collaborative design tasks on their own terminals (such as PCs, tablets, mobile phones, etc.), view the detailed requirements of the tasks, including design goals, reference information, deadlines, etc.; preview the existing 3D model data associated with the task on the cloud 3D design platform; according to the task requirements, designers start to create the initial 3D design model on the cloud 3D design platform; use the 3D modeling tools provided by the platform, such as the Web-based 3D editor; according to the task requirements, designers can create new 3D models from scratch or modify existing models; designers perform detailed modeling, texture mapping, Edit material settings, etc.; use the real-time preview function provided by the cloud-based 3D design platform to dynamically view the model effect and optimize it; iteratively design and adjust the model's size, structure, performance, etc.; after the designer completes the initial design model, he saves it to the cloud-based 3D design platform; the platform automatically records the model's version information and associates it with the task status; other designers participating in the collaboration can also see the changes to the model in real time; the design supervisor or expert views and reviews the initial models submitted by each designer on the cloud-based platform; puts forward modification opinions and suggestions, and the designer further optimizes based on the feedback. Only after the model passes the review can it enter the next round of collaborative design. Through the above steps, each design terminal can independently generate an initial design model on the cloud 3D design platform according to the assigned collaborative tasks, and effectively collaborate with other designers, which can improve the work efficiency and creativity of designers and shorten the design cycle; enhance the visualization and collaboration of the design process and promote communication among design teams; simplify the management and synchronization of design data to avoid version conflicts and data loss; provide a more transparent and reliable data foundation for design review and feedback; and provide reliable digital design assets for subsequent product optimization and iteration. In short, through the cloud 3D collaborative design platform, designers can better play their expertise and collaboratively complete product design tasks, thereby improving the product development capabilities of the entire enterprise. It is understandable that for design terminals equipped with intelligent design modules, the initial design model of the design task can be completed independently and automatically without human intervention, which can greatly improve work efficiency and reduce errors.

[0161] Merging the multiple initial design models to obtain an initial product design model;

[0162] In this step, the initial design models submitted by various designers / design terminals are collected on the cloud-based 3D design platform to ensure that each model has passed the review and meets the fusion standard; the collected 3D model data in different formats are uniformly converted to ensure that all model data can be processed on the same 3D modeling software or platform; each initial design model is placed in a unified 3D coordinate system, and the position, orientation, etc. of the model are calibrated and adjusted according to design requirements; the calibrated initial design models are geometrically merged using 3D modeling software or professional model fusion tools; overlaps, gaps and other issues between models are handled to ensure the geometric coherence of the overall fused model; materials and texture maps are uniformly applied to the fused overall model to ensure visual coordination and beauty between components; the performance of the fused initial product design model is optimized, such as simplifying details and reducing the number of faces; the optimized model is exported to a unified 3D model format for subsequent use. Through the above steps, multiple initial design models can be effectively integrated to obtain a complete initial product design model, realize the seamless integration of the work results of different designers / design terminals, improve the efficiency of collaboration, ensure the overall consistency and visual coordination of product design, provide a reliable digital design foundation for subsequent product evolution, simulation analysis, etc., reduce design rework and duplication of work, shorten the product development cycle, and provide a more intuitive and comprehensive digital product model for design review and decision-making. In short, through the model fusion function provided by the cloud-based 3D collaborative design platform, enterprises can better manage and integrate the product design process and improve the overall R&D efficiency and product quality.

[0163] Use VR technology to create an initial product virtual model corresponding to the initial product design model (to facilitate internal team evaluation and early customer feedback);

[0164] In this step, the initial product design 3D model generated by the above fusion is imported into the VR modeling software or platform to ensure that the model's geometric information, material mapping and other data can be fully imported; a virtual scene suitable for product display is created in the VR software, including environmental settings, lighting arrangements, display stands and other elements to ensure that the scene can create a good product display effect; interactive functions are added to the initial product virtual model, such as rotation, zooming in, viewing details, etc., using VR controllers or gestures to enhance the user's immersive experience; rendering of virtual scenes and models is optimized to improve image quality and smoothness, and adaptation is made according to the performance of the target VR device; the optimized virtual scenes and product models are exported as publishable VR content packages to support mainstream VR platforms and devices, such as Oculus, HTC Vive, PlayStation VR, etc.; design team members, customers, etc. are invited to use VR devices to experience the initial product virtual model; user feedback is collected and fed back to the product design team. Through the above steps, the initial product design model can be converted into an immersive virtual reality experience, and the visualization and interactivity of product design can be improved to enhance the user experience; feedback from internal teams and customers can be obtained in the early design stage; design problems can be discovered and optimized in a timely manner to shorten the product development cycle; a new immersive display method can be provided for product marketing and display; experience and technology in VR product display can be accumulated to lay the foundation for future development. In short, using VR technology to create an initial product virtual model can help companies better design, evaluate and display products, and improve the overall efficiency of product development and user experience.

[0165] Use natural language processing technology to automatically analyze the product technical information of the corresponding products and obtain technology development trend data;

[0166] In this step, technical information related to the target product is collected, such as patent documents, industry reports, and technology news. Ensure that the collected information covers the historical development and latest developments of the product. Perform preprocessing operations such as cleaning, word segmentation, and part-of-speech tagging on the collected unstructured text materials to remove irrelevant content and extract keywords and phrases related to product technology. Use topic model algorithms such as Latent Dirichlet Allocation (LDA) to perform topic clustering on the preprocessed text. Identify the main technical themes and concepts contained in the text. Based on the topic division, further extract key terms and phrases related to product technology. Filter and classify the extracted terms in combination with background knowledge. Analyze the frequency and change trend of themes and terms in time series to identify the hot spots and potential trends in product technology development. Visually display the analysis results in the form of charts, interactive dashboards, etc., to facilitate technical decision makers to understand and gain insight into the development direction of product technology. Through the above steps, NLP technology can be used to automatically extract and analyze key information from a large amount of product technical data, improve the efficiency and comprehensiveness of technical intelligence mining, and reduce the cost of manual analysis; timely grasp the latest developments and trends in product technology development; provide data support and basis for product design, R&D investment and other decisions; discover potential technical opportunities and innovation directions to enhance product competitiveness; accumulate technical analysis experience and methods to lay the foundation for future intelligence analysis. In short, using NLP technology to automatically analyze and mine trends in product technical data can help companies better understand the dynamics of technological development and improve the pertinence and foresight of product R&D.

[0167] Modifying the initial product virtual model according to the evaluation feedback of the initial product virtual model and the technology development trend data to obtain a basic product virtual model;

[0168] In this step, the evaluation opinions of the design team and customers on the initial product virtual model are sorted and analyzed to identify the key issues and user needs that need to be improved; combined with the product technology development trend data obtained by the aforementioned natural language processing, the optimization direction and innovation points that meet the future technology direction are determined; according to the evaluation feedback and technology trends, a modification plan for the virtual model is formulated, including structural optimization, function enhancement, and interactive experience improvement; the initial virtual model is modified and optimized in the VR modeling software to achieve improvements such as structural adjustment, material beautification, and function expansion, and a richer user interaction method is designed to enhance immersion; the optimized virtual model is rendered with high quality, and simulation tests are performed using the physical engine to verify product performance; the virtual model that has been optimized for multiple rounds is exported as a displayable content package to support preview and interactive experience on mainstream VR platforms. Through the above steps, the initial product virtual model can be transformed into a basic product virtual model, which can fully absorb user feedback, be close to actual needs, and improve the pertinence of product design; combined with forward-looking technology trends, the innovation and competitiveness of the product are enhanced; the interactivity and simulation capabilities of the virtual model are improved, and the user experience is enhanced; the iterative optimization process is more efficient, shortening the product development cycle, and laying a good foundation for further improvement of the product in the future. In short, through continuous optimization and improvement of the initial product virtual model, a basic product virtual model that meets user needs and conforms to the direction of technological development is eventually formed.

[0169] Using artificial intelligence to simulate and optimize the basic product virtual model (accelerate product performance testing and optimization process) to obtain a finalized product model;

[0170] In this step, select a suitable physical simulation engine, such as Unreal Engine, Unity, etc., and import the optimized basic product virtual model into the simulation environment; determine the type of simulation test to be performed based on product functions and performance indicators, including structural strength, dynamic performance, fluid / heat conduction and other tests; use programming languages ​​such as Python and C++ to develop automated simulation scripts to achieve batch simulation testing of virtual models, greatly improving testing efficiency; during the simulation process, collect test data of various performance indicators of the product, including stress, deformation, flow field, temperature distribution and other multi-dimensional data; use machine learning algorithms to analyze and model the collected performance data; identify key parameters that affect product performance and perform intelligent optimization; adjust the parameters of the virtual model based on the results of artificial intelligence optimization; repeat the simulation test and optimization process until the product performance requirements are met, and output the fully simulated and optimized virtual model as the final product finalization model to provide an accurate design reference for subsequent physical product development. Through the above steps, using artificial intelligence technology to simulate and optimize product virtual models can greatly improve the automation and intelligence level of product performance testing; shorten the product development cycle and accelerate the conversion process from virtual to physical; improve product reliability and performance indicators to meet higher user expectations; accumulate experience data for product optimization to lay the foundation for future product iterations; reduce the cost and resource consumption of physical sample manufacturing and improve R&D efficiency. In short, using artificial intelligence technology to accurately simulate and intelligently optimize product virtual models can significantly improve the scientificity and reliability of product development and bring technical and commercial advantages to enterprises.

[0171] Based on the product finalization model and the product technical information, a research and development knowledge map is constructed to support cross-departmental knowledge sharing and rapid problem location;

[0172] In this step, various R&D materials such as product finalization models, design documents, test reports, etc. are collected, and these unstructured data are sorted and standardized; natural language processing technology is used to automatically extract key entities and concepts from the R&D materials; based on the extracted knowledge elements, a knowledge graph containing information such as product structure, function, and performance is constructed; knowledge reasoning and graph algorithms are applied to optimize and supplement the constructed knowledge graph; multiple knowledge graphs are integrated to form a comprehensive R&D knowledge graph; an intuitive graphical interface is designed to support users to browse and query the knowledge graph; a question-and-answer system based on natural language is developed to support users to quickly retrieve the required knowledge; a system and process for knowledge sharing is formulated to encourage cross-departmental knowledge accumulation and exchange; knowledge graphs are used to support the collaborative work of design, production, service and other departments; a regular update mechanism for the knowledge graph is established to ensure the timeliness and accuracy of information; user feedback is collected to continuously optimize the functions and interactive experience of the knowledge graph. Through the above steps, building an R&D knowledge graph can achieve systematic management of R&D knowledge, improve the efficiency of knowledge acquisition and reuse; support cross-departmental knowledge sharing and collaboration, and promote the overall R&D capabilities of the enterprise; a natural language-based question-answering system can help R&D personnel quickly locate the required information; use graph algorithm analysis to discover hidden knowledge connections and potential innovation opportunities; and continuously optimize the knowledge graph to support the management of the entire product life cycle. In short, building an R&D knowledge graph can effectively integrate and utilize the R&D resources within the enterprise, promote cross-departmental knowledge sharing and collaboration, and assist in the management of the entire product development process.

[0173] According to the product finalization model and the product ordering data, a machine learning algorithm is used to predict demand, optimize inventory levels, and generate raw material procurement data;

[0174] In this step, relevant technical parameters of the product finalization model are collected, historical product ordering data, including order quantity, delivery time, etc., are collected, and the data is cleaned and organized to ensure its integrity and availability; appropriate machine learning algorithms are selected, such as time series prediction, regression analysis, etc.; historical ordering data is used to train a model that can accurately predict future demand; production capacity is analyzed in combination with the manufacturing process parameters in the product finalization model, and a production plan that meets customer needs is formulated based on the demand forecast results; the supply-demand characteristics of the product are analyzed to determine a reasonable safety stock level; inventory management strategies are dynamically adjusted based on production plans and safety stocks; based on production plans and inventory conditions, the required raw material purchase quantities are calculated, and a raw material purchase plan is generated, including purchase time, quantity, supplier, etc.; the accuracy of the prediction model is continuously tracked and iterative optimization is performed; the execution of the production plan is monitored and possible problems are analyzed. Through the above steps, using machine learning algorithms to predict production demand, optimize inventory levels, and generate raw material procurement data can greatly improve the accuracy of demand forecasts and reduce overproduction or shortages caused by forecast deviations; optimize inventory management, reduce capital occupation and storage costs, and improve asset utilization efficiency; generate accurate raw material procurement plans, reduce raw material procurement costs and supply risks; achieve a dynamic balance between production, inventory, and procurement, and improve overall operational efficiency; and continuously improve the intelligence level of the system through continuous monitoring and optimization. In short, using machine learning algorithms to analyze product finalization models and ordering data can help companies achieve intelligent management of the entire process from demand forecasting to raw material procurement, greatly improving the agility and efficiency of production operations.

[0175] According to the raw material procurement data, based on multi-dimensional data analysis, real-time evaluation and selection of the best suppliers are carried out, and raw material procurement is carried out;

[0176] In this step, historical data on raw material procurement are collected, including price, delivery time, quality, etc.; relevant information about suppliers, such as enterprise size, credit rating, service level, etc., are collected; these data are cleaned and integrated to form a data set for analysis; key indicators for evaluating suppliers, such as price, delivery time, quality, service, etc., are determined; a comprehensive supplier evaluation model is established using methods such as multivariate regression and hierarchical analysis; various indicator data in the raw material procurement process are monitored in real time, and the performance of suppliers is evaluated and ranked in real time based on the evaluation model; based on the real-time evaluation results, the best supplier that meets the current procurement needs is automatically selected; considering factors such as safety stock and procurement cycle, the best supplier selection plan is generated; the supplier selection plan is connected with the procurement system to realize automatic ordering; the order execution is continuously tracked to promptly discover and solve problems; procurement execution data is collected to analyze procurement efficiency and cost change trends; based on the analysis results, the supplier evaluation model and procurement strategy are optimized. Through the above steps, real-time evaluation and screening of the best suppliers based on multi-dimensional data analysis can improve procurement cost control capabilities and reduce raw material procurement costs; optimize supply chain response speed, improve the timeliness and reliability of supply; continuously improve the supplier evaluation system to ensure continuous improvement of supplier quality; realize automation and intelligence of procurement processes to improve procurement efficiency; data-driven performance analysis and optimization to promote continuous improvement of procurement management. In short, real-time evaluation and screening of the best suppliers through multi-dimensional data analysis can help companies establish an agile and efficient raw material procurement system, improve the level of supply chain management, and ultimately achieve the goals of cost control and delivery optimization.

[0177] Use blockchain technology to achieve full traceability of raw materials to ensure quality and compliance;

[0178] In the process of manufacturing the first product using raw materials, an optimization algorithm is used to realize intelligent scheduling of the production line to improve production efficiency;

[0179] In this step, collect the real-time status data of key equipment of the production line, such as equipment operating parameters, product process parameters, etc.; collect real-time production progress data, such as current product output, process time, etc.; establish a production scheduling optimization model based on production goals, process constraints, equipment performance and other factors; select appropriate optimization algorithms, such as genetic algorithms, simulated annealing algorithms, etc., input real-time production data into the optimization model, and generate the optimal production scheduling plan; automatically push the scheduling plan to the production line control system to achieve intelligent scheduling; monitor the operation status of the production line in real time, discover and solve production problems in a timely manner, collect production process data, and provide a basis for subsequent optimization; analyze indicators such as production efficiency and product quality, and evaluate the optimization effect; according to the analysis results, continuously improve the optimization model and scheduling strategy. Through the above steps, the use of optimization algorithms to achieve intelligent scheduling of production lines can improve equipment utilization and production line efficiency, shorten product production cycles; reduce waste and rework caused by production problems, and improve product quality; optimize production plans, shorten delivery time, and meet customer needs; realize the intelligence and automation of the production process and reduce labor costs; accumulate production data and analyze and optimize, and continuously improve production capacity. In short, through the steps of production line data collection, optimization algorithm modeling, real-time scheduling execution and performance analysis, the intelligent scheduling of the production line can be achieved, greatly improving production efficiency and product quality.

[0180] During the manufacturing process of the first product, deep learning and computer vision technology are used to perform real-time quality inspection on the product being produced to obtain a first inspection result;

[0181] In this step, collect multi-angle high-definition images or video data of products in the production process, manually annotate these data, and mark various defects or flaws of the products; select appropriate deep learning network architecture, such as convolutional neural network (CNN), target detection network, etc.; use the annotated data set to train the model so that it can learn the characteristics of product defects; deploy the trained deep learning model to the computing equipment of the production line, capture product images or video data in real time, and input them into the trained model for reasoning; the deep learning model will output the quality inspection results of the product, including defect type, location, etc.; feed these results back to the production management system as the basis for subsequent quality control; continue to collect new product image data and annotate the defects that appear; use the new data set to fine-tune and optimize the deep learning model to improve its detection accuracy. Through the above steps, using deep learning and computer vision technology to achieve real-time product quality inspection can improve the accuracy and consistency of quality inspection and reduce the error of manual sampling; realize real-time monitoring of product quality, timely discover and deal with quality problems in the production process; collect a large amount of product quality data to provide a basis for further optimizing production processes and procedures; lay the foundation for intelligent manufacturing and quality management informatization, and promote enterprise transformation and upgrading. In general, applying deep learning and computer vision technology to product quality inspection can greatly improve inspection efficiency and accuracy, realize intelligent quality control of the production process, and create conditions for enterprises to continuously improve product quality and production management level.

[0182] Based on the first detection result, using an intelligent collaborative robot, the manufacturing process is controlled to improve production flexibility and efficiency;

[0183] In this step, the quality inspection system is integrated with the production management system to achieve real-time transmission of inspection results; the intelligent collaborative robot is configured to exchange information with the production line equipment to implement command issuance and feedback; the quality inspection results are analyzed in real time, and an early warning is triggered immediately if an abnormal situation is found; based on the early warning information, the intelligent collaborative robot is automatically called to handle the exception, such as adjusting process parameters, performing rework, etc.; the high flexibility and precision of the intelligent collaborative robot are used to intelligently adjust the production line; based on real-time monitoring data, process parameters, material delivery, equipment operating status, etc. are automatically adjusted; a human-machine collaborative mechanism is established, with manual supervision and participation in key decisions, and robots performing specific operations; production process data is collected, the process flow is analyzed and optimized, and production efficiency is continuously improved; various types of production, quality, equipment, etc. data are integrated to establish a global production management database; big data analysis technology is used to explore the laws and problems in the production process to provide a basis for further optimization. Through the above steps, using intelligent collaborative robots to control and optimize the manufacturing process in real time can improve the flexibility and adaptability of the production line, meet the production needs of multiple varieties and small batches; reduce rework and waste caused by quality problems, improve production efficiency and product yield; realize automation and intelligence of the production process, reduce labor costs and management difficulties; accumulate a large amount of production data to provide a foundation for future intelligent production management; promote human-machine collaboration, give full play to the respective advantages of humans and machines, and improve overall production capacity. In short, the deep integration of intelligent collaborative robots with quality inspection systems and production management systems can realize intelligent control and optimization of the production process, greatly improving production flexibility and efficiency.

[0184] Using machine learning algorithms to generate defect prediction models based on historical data to predict product defects;

[0185] In this step, various types of relevant data accumulated during the production process are collected, including raw materials, process parameters, historical defect records, etc.; the data are cleaned, integrated and feature-engineered to ensure data integrity and availability; appropriate machine learning algorithms are selected, such as decision trees, random forests, neural networks, etc.; the preprocessed data sets are used to train and optimize the algorithms to establish defect prediction models; cross-validation and other methods are used to evaluate the prediction accuracy and generalization ability of the model on new data; based on the evaluation results, the model parameters and features are further adjusted to continuously improve the prediction performance; the trained defect prediction model is deployed to the production management system to achieve real-time prediction; an interface with the quality inspection system is established to compare and feedback the prediction results with the actual inspection results; new production data is continuously collected, and the prediction model is retrained and iterated in combination with the actual inspection results; in this way, the accuracy and robustness of the prediction model are continuously improved. Through the above steps, the use of machine learning algorithms to generate product defect prediction models can improve the predictability of the production process, timely warn of possible quality problems; optimize production plans and resource allocation, reduce rework and waste caused by quality problems; provide decision support for production managers, and guide targeted quality improvement measures; with data accumulation and model optimization, the prediction accuracy continues to improve, laying the foundation for intelligent manufacturing, providing a basis for the formulation of preventive maintenance plans, and improving equipment reliability and service life. In general, the use of machine learning technology to establish a product defect prediction model can greatly improve the predictability and controllability of the production process, and lay the foundation for enterprises to achieve lean production and intelligent manufacturing.

[0186] The manufactured first product is tested according to the defect prediction model.

[0187] In this step, first develop an intelligent and modular automatic test system to improve test coverage and efficiency. Collect various types of data related to the defect prediction model on the production line, such as raw material parameters, process parameters, product characteristics, etc.; ensure that the time, location, and method of data collection are consistent with the model training stage; input the collected product data into the trained defect prediction model; the model automatically calculates the probability of defect occurrence of the product; confirm the actual defect situation of the product through the automatic test system, compare and analyze the test results with the model prediction results, and evaluate the prediction accuracy; feedback the verification results to the model training team for continuous optimization and iteration of the model; at the same time, pass the prediction results to the production management system to guide subsequent production plans and quality control; for products predicted by the model as high-probability defects, take timely remedial measures, such as rework or scrapping; adjust the corresponding process parameters and production plans according to the actual defect situation. Through the above steps, using the defect prediction model to test the manufactured products can improve the predictability of product quality, timely discover and eliminate potential quality problems; reduce rework and waste caused by quality problems, and significantly improve production efficiency; provide data support for quality improvement measures and promote continuous quality optimization; lay the foundation for establishing a complete quality management system and enhance the overall competitiveness of the enterprise; accumulate more practical application data and continuously improve and optimize the defect prediction model. In short, using the defect prediction model to test the manufactured products can effectively monitor and control product quality, which is a key step in the transformation of intelligent manufacturing.

[0188] In some possible implementations of the present invention, it also includes: remote monitoring and data collection of product operating status for continuous improvement; using natural language processing and machine learning technology to provide personalized customer service and sales recommendations; using augmented reality technology to provide customers with an immersive product display experience; adjusting the optimal pricing strategy in real time based on market data and competitor analysis; using optimization algorithms to achieve intelligent planning of logistics routes and improve distribution efficiency; introducing intelligent warehousing robots to improve warehouse management efficiency; using Internet of Things technology to achieve full product tracking and provide accurate distribution information; based on product usage data and machine learning models, predict possible failures and proactively arrange maintenance; through the Internet of Things and artificial intelligence technology, achieve remote product diagnosis and online resolution of some problems; use augmented reality technology to provide visual maintenance guidance for on-site technicians; develop an integrated collaboration platform to achieve seamless information sharing and collaboration among departments; use artificial intelligence to assist in project planning, resource allocation and progress tracking; provide management with a real-time, intuitive data visualization interface to assist in decision making; use natural language processing technology to automatically check whether documents and processes comply with relevant regulations; and use machine learning algorithms to monitor and analyze network security threats in real time.

[0189] The solution of this embodiment integrates multiple advanced technologies such as artificial intelligence, Internet of Things, virtual reality, augmented reality, etc., which improves the efficiency and coordination of the entire product life cycle. It can not only optimize the operation of each link, but also achieve seamless collaboration across departments, thereby improving overall production efficiency and product quality.

[0190] In some possible implementations of the present invention, the step of establishing an integrated intelligent collaboration platform to connect the information systems of various work departments within the enterprise includes:

[0191] Obtain the first attribute information of the existing information system of each work department and the second attribute information of the work process;

[0192] In this step, by obtaining the information system currently used by each work department, collect the first attribute information such as system name, system type, user, application scenario, etc.; obtain the specific work content and process of each department, record the second attribute information such as work tasks, information transmission, and approval links; organize and classify the collected information system and workflow data, and establish a unified information system and workflow ledger to facilitate subsequent analysis; confirm and supplement the organized information to ensure the accuracy and completeness of the information; based on the collected information system and workflow data, identify the connection and coordination needs between work departments, and discover problems such as information islands, process redundancy, and resource waste; propose solutions to optimize information systems and workflows, including integration, automation, etc., to provide decision support for the digital transformation of enterprises. Through the above steps, detailed information on the information systems and workflows of each work department can be obtained, and the current status of the enterprise's IT infrastructure and business operations can be fully understood, the pain points and problems in the information system and business processes can be discovered, and practical optimization suggestions can be provided for the digital transformation of enterprises, laying the foundation for the subsequent construction of enterprise-level application systems, and laying a good foundation for cross-departmental collaboration and information sharing. In general, systematically sorting out the information systems and work processes of various departments within an enterprise is a key step in promoting the digital transformation of the enterprise. It can not only diagnose problems with current IT infrastructure and business operations, but also provide decision-making support for future system integration and process optimization.

[0193] Determine an integration target and an integration scope according to the first attribute information and the second attribute information;

[0194] In this step, analyze the current pain points of each department's information systems and work processes, determine the core issues that need to be resolved by building an intelligent collaboration platform, clarify the expected goals and effects of platform construction, obtain the specific business needs and collaboration needs of each department, collect the functional requirements and integration requirements of various information systems, and identify the key links and pain points of cross-departmental collaboration; based on the goals and needs, determine the scope of information systems that need to be integrated, identify key business processes and collaboration scenarios, clarify the focus of integration, weigh factors such as system complexity and integration difficulty, and reasonably define the integration boundaries; design an integrated architecture solution that meets the current status of enterprise IT, determine the technical path and implementation method of integration, formulate a detailed system integration plan and implementation plan, and gradually complete the integration of each system according to the design plan to ensure the stability and reliability of the integration process, continuously optimize platform functions, and enhance user experience. Through the above steps, the integration goals and scope of the intelligent collaboration platform are determined according to the attribute information of the information system and workflow, which can eliminate information islands between departments and realize cross-system and cross-department information sharing; optimize business processes, eliminate duplication, improve work efficiency and coordination; enhance decision-making support capabilities, and make more accurate decisions based on a global perspective; improve employee work experience, empower employees, and stimulate innovation; lay a solid foundation for the digital transformation of enterprises and enhance market competitiveness. In short, determining the integration goals and scope of the intelligent collaboration platform based on system and process attribute information is a key step in promoting the digital transformation of enterprises. It helps to build an efficient and collaborative digital operation system and provide basic support for the future intelligent development of enterprises.

[0195] Develop an integration strategy and technology roadmap based on the integration goals and integration scope;

[0196] In this step, we fully analyze the current status of the enterprise IT architecture and the digital transformation goals, combine the integration goals and integration scope, determine the overall integration strategy, select the appropriate integration mode, such as API integration, middleware integration, etc., determine the integration standards and technical specifications, and ensure the interoperability between systems; design the integrated technical architecture according to the integration goals and strategies; determine the key technical components, including the integration platform, data center, etc.; plan the phased implementation and launch plan of the integration project; formulate detailed technical solutions such as system integration and data migration; formulate relevant systems for data asset management and data standardization, establish a unified data architecture, and ensure the quality and security of data; develop a mechanism for data sharing and exchange to promote cross-system collaboration; evaluate the mainstream integration platform products on the market according to the integration needs; compare product functions, technical architecture, deployment costs and other indicators; and select the most suitable integration platform for the enterprise as the infrastructure. Through the above steps, we can formulate an integration strategy and technical roadmap, establish an enterprise-level integration architecture, realize the effective collaboration of IT systems, improve data quality and analysis capabilities, and enhance the level of decision support; shorten the system launch cycle and accelerate the progress of digital transformation; reduce IT maintenance costs and improve the flexibility of system operation and maintenance. In short, formulating integration strategies and technology roadmaps based on integration goals and scope is a key link in promoting enterprise digital transformation.

[0197] Designing a distributed system architecture for the intelligent collaboration platform according to the integration strategy and the technical roadmap, planning a data storage and processing solution for the intelligent collaboration platform, and designing a security architecture and access control mechanism for the intelligent collaboration platform;

[0198] In this step, analyze business needs and system performance requirements to determine the appropriate distributed architecture; select key technical components such as microservices, containers, and message queues; define service boundaries and inter-service communication mechanisms to achieve loose coupling; design horizontal expansion and high-availability deployment topology structures; sort out the characteristics and usage scenarios of various types of business data; select appropriate distributed databases, data lakes, data warehouses, etc.; design data collection, cleaning, storage, and analysis processes; use real-time computing, batch processing and other technologies to achieve high-performance data processing; formulate a comprehensive security protection strategy based on business needs and compliance requirements; design security mechanisms such as identity authentication, permission management, and data encryption; use security equipment and technical means such as WAF and IDPS; establish a closed-loop management system for security monitoring and incident response; formulate a comprehensive testing plan to ensure system functionality and performance; formulate emergency plans and rollback mechanisms to ensure smooth launch; establish a continuous optimization mechanism to improve system operation and maintenance capabilities. Through the above steps, the distributed system architecture, data solution and security architecture of the intelligent collaboration platform can be designed to achieve business agility and system scalability to meet future demand changes; improve data processing capabilities and analytical insights to support accurate decision-making; build a comprehensive security protection system to ensure the security of information assets; shorten the launch cycle, improve product delivery efficiency and user experience; reduce IT operation and maintenance costs, and improve system reliability and stability. In short, designing the technical architecture of the intelligent collaboration platform based on the integration strategy and technology roadmap is a key link in achieving the goal of digital transformation; it lays the foundation for the reliability, security and high performance of the platform, and provides strong support for the company's future business innovation.

[0199] Formulate unified data standards for the intelligent collaboration platform, establish data mapping relationships between various information systems, and design data conversion and cleaning processes;

[0200] Develop standardized APIs for the information systems of various work departments, implement real-time data synchronization mechanisms, and develop middleware to handle communications between different information systems;

[0201] Build a unified user interface for the intelligent collaboration platform, develop a workflow engine, implement unified identity authentication and permission management, and develop data visualization and reporting functions;

[0202] Integrate natural language processing modules for intelligent search and automatic classification into the intelligent collaboration platform;

[0203] Embed machine learning algorithms for data analysis and prediction in the intelligent collaboration platform, develop intelligent recommendation systems, and integrate knowledge graph construction and management functions;

[0204] Conduct unit testing, integration testing, system testing, performance testing, and load testing on the intelligent collaboration platform to optimize system response speed and stability;

[0205] Develop data migration strategies for the intelligent collaboration platform, perform data cleansing and conversion, and migrate data in batches;

[0206] Generate user manuals and technical documentation for the intelligent collaboration platform, train users from various work departments, develop and execute a phased deployment plan;

[0207] Collect user feedback, monitor the system performance of the intelligent collaboration platform, and regularly update and optimize system functions.

[0208] In this embodiment, it can realize real-time information sharing across departments, eliminate information islands, and improve data consistency; reduce duplication of work and data entry, accelerate information flow and decision-making process; support cross-departmental collaboration and project management, and improve team communication efficiency; discover business insights through centralized data analysis, and support data-driven decision-making; realize standardization and automation of business processes, and improve process visibility and controllability; introduce AI-assisted decision-making and predictive analysis, and realize intelligent search and personalized recommendations; unified identity authentication and access control, centralized data protection and auditing; facilitate the integration of new systems and functions, and support the rapid development and change of business; unified user interface reduces learning costs, and personalized workbenches improve user satisfaction; reduce the complexity of system maintenance and upgrades; optimize resource utilization and reduce operating costs; provide data and tool support for cross-domain innovation, and accelerate the verification and implementation of new ideas. In this embodiment, through this integrated intelligent collaboration platform, enterprises can achieve seamless information flow, effective knowledge sharing, and optimal resource allocation, thereby greatly improving overall operational efficiency and decision-making quality, and laying a solid foundation for the digital transformation and intelligent upgrading of enterprises.

[0209] In some possible implementations of the present invention, the step of using natural language processing technology to perform real-time analysis on documents, emails, and communication records in the intelligent collaboration platform to extract key information includes:

[0210] Establish a data interface to obtain documents, emails and communication records in the intelligent collaboration platform in real time and integrate them into the first text;

[0211] Performing text cleaning on the first text, removing irrelevant characters and punctuation marks, and then performing text standardization (such as unifying uppercase and lowercase letters, expanding abbreviations, etc.) to obtain a second text;

[0212] Use a word segmentation algorithm (such as Jieba word segmentation) to segment the second text, and perform part-of-speech tagging (identify nouns, verbs, adjectives, etc.) to obtain a third text;

[0213] Identify and annotate entities (such as names of people, places, names of organizations, etc.) and professional terms and keywords in specific fields in the third text;

[0214] Using a preset text classification model (such as a pre-trained model such as BERT or XLNet) to classify the third text into topics (such as project reports, technical documents, customer feedback, etc.);

[0215] Extract the text summary of the classified third text (using TextRank or BERT-based algorithm), and identify key sentences and keywords as key information (such as time, place, person, etc.) from the text summary;

[0216] Constructing a sentiment analysis model to identify the sentiment tendency of the third text to perform sentiment evaluation (on customer feedback, employee communication, etc.) and obtain a sentiment evaluation result;

[0217] Develop an intent recognition model to understand the purpose or demand behind the third text and identify user intentions such as task assignment, problem feedback, and decision requests;

[0218] Identify the relationships between entities (such as person relationships, project dependencies, etc.), build corresponding knowledge graphs, and display the logical associations between entities;

[0219] Annotate sentiment evaluation results, user intentions, and logical associations between entities as key information;

[0220] Store the annotated key information in a structured database and build a full-text search engine (such as Elasticsearch) for fast retrieval;

[0221] Develop data visualization interfaces to display analysis results and implement automatic report generation to summarize key findings;

[0222] Desensitize data, protect sensitive information, establish access control mechanisms, and ensure information security.

[0223] In this embodiment, key information can be automatically extracted to reduce manual screening time; rapid processing and analysis of large-scale texts can be achieved; an enterprise knowledge base can be automatically constructed to facilitate knowledge precipitation and sharing; the accuracy and efficiency of knowledge retrieval can be improved; data support can be provided for management decision-making and potential problems and opportunities can be quickly identified; project progress and risks can be automatically tracked to improve the organization and utilization efficiency of project documents; customer feedback can be deeply analyzed to improve customer satisfaction; customer demand trends can be identified to support product optimization; meeting content can be automatically summarized to reduce information omissions; tasks can be quickly identified and assigned; potential risk signals can be identified early to improve the efficiency of compliance monitoring; cross-domain knowledge associations can be discovered to stimulate innovative thinking and quickly evaluate the feasibility of new ideas; personalized information recommendations can be provided to reduce repetitive work and improve job satisfaction; valuable insights can be mined from unstructured data to support a data-driven corporate culture; real-time alerts can be issued for important information to support rapid decision-making and action. Through this NLP-based real-time analysis system, enterprises can manage and utilize their information assets more effectively, improve overall operational efficiency, enhance decision-making quality, and provide data support for continuous innovation and improvement; this not only optimizes daily work processes, but also provides strong technical support for enterprises to maintain their competitiveness in the digital age.

[0224] In some possible implementations of the present invention, the step of classifying, associating and evaluating the importance of the extracted key information based on a machine learning algorithm to obtain key information analysis data includes:

[0225] After data cleaning and standardization of the key information, text features are extracted, metadata features (such as source, timestamp, author, etc.) are generated, and context features (such as document location, surrounding text, etc.) are created;

[0226] Select the corresponding classification algorithm (such as random forest, SVM, deep learning model), train the multi-label classification model, and classify the key information by topic;

[0227] Use content-based recommendation algorithms to build association models, and apply graph neural networks to analyze the associations between key information;

[0228] Develop a multi-factor scoring model that takes into account the novelty, relevance, and influence of key information, and evaluates the importance of key information in the overall text network (using a PageRank-like algorithm) (a time decay factor can also be applied to balance the weight of new and old information).

[0229] In this embodiment, it is possible to realize automatic information classification and improve information retrieval efficiency; establish a multi-dimensional information index to support complex queries; reveal implicit associations between information and promote knowledge innovation; identify key knowledge points and knowledge gaps; provide prioritized information to support quick decision-making; automatically identify key risks and opportunities; predict future trends based on historical data and current information; identify potential problems and development opportunities in advance; optimize the allocation of manpower and funds according to the importance of information; improve the execution efficiency of key projects and tasks; provide customized information services for users of different roles, improve information consumption efficiency, and reduce information overload; quickly identify abnormal patterns and potential risks and improve enterprise risk management capabilities; automatically build and update enterprise knowledge graphs to support complex knowledge reasoning and exploration; identify relevant experts and resources to promote cross-departmental collaboration; reduce duplication of work and improve overall efficiency; discover cross-domain knowledge associations, stimulate innovative thinking, and support data-based innovative decisions; the system can learn from user interactions and continuously improve accuracy; adapt to the specific needs of different industries and organizations; provide objective value assessment for enterprise information assets, support ROI-based information management strategies, automatically identify sensitive information, ensure data compliance, and improve the accuracy of information security management. Through this machine learning-based key information analysis system, enterprises can more effectively use their information assets, improve decision-making quality, enhance risk management capabilities, and provide data support for continuous innovation. This not only optimizes the information management process, but also provides strong technical support for enterprises to gain competitive advantages in digital transformation. The system's adaptability and scalability also ensure that it can be continuously optimized and improved as the enterprise develops, and continue to provide value.

[0230] In some possible implementations of the present invention, the step of using a predictive analysis model to predict potential project risks, resource requirements, and schedule delays based on the historical work data and the current work data to obtain a project prediction result includes:

[0231] Integrate the historical working data and the current working data, perform data cleaning to handle missing values ​​and outliers, and perform feature engineering;

[0232] Use correlation analysis to identify important features from the output of the previous step, apply principal component analysis (PCA) to reduce the data dimensionality, and use domain knowledge to select key predictors;

[0233] According to the key prediction factors, select corresponding prediction algorithms for different prediction tasks;

[0234] In this step, the corresponding prediction algorithms are selected for different prediction tasks, which can be: project risk (such as random forest, gradient boosting tree), resource requirements (such as time series analysis, regression model), progress delay (such as survival analysis, decision tree);

[0235] In combination with the prediction algorithm, develop a deep learning model for complex pattern recognition;

[0236] Training and validating the deep learning model (e.g., dividing the deep learning model into a training set, a validation set, and a test set; optimizing model parameters using cross-validation; and implementing regularization techniques to prevent overfitting) to obtain a predictive analysis model;

[0237] Develop automated data input and preprocessing pipelines and implement model inference engines to support batch processing and real-time prediction;

[0238] Design a visual interface to display prediction results;

[0239] Develop interactive tools for model interpretation exploration, explain model decisions, and generate feature importance reports;

[0240] Design a risk scoring system to determine overall project risk and set dynamic thresholds to determine different risk levels;

[0241] The historical work data and the current work data are input into a predictive analysis model to predict potential project risks, resource requirements and schedule delays to obtain a project prediction result.

[0242] In this embodiment, it can also include: implementing an online learning mechanism to dynamically update the model; developing an A / B testing framework to evaluate model improvements; establishing a feedback loop to integrate manual corrections and audits; integrating with a project management system to achieve seamless data flow; developing an API interface for external system calls to achieve docking with a decision support system; implementing a distributed computing framework for large-scale data processing to optimize prediction speed, such as model quantization and GPU acceleration; and designing an automated model update and deployment process.

[0243] Through the solution of this embodiment, potential project risks can be identified in advance and the project failure rate can be reduced; proactive risk management can be supported to reduce the impact of emergencies; resource requirements can be accurately predicted to avoid resource waste or shortages; dynamic resource scheduling can be supported to improve resource utilization efficiency; possible schedule delays can be accurately predicted to support early intervention; project timelines can be optimized to improve on-time delivery rates; data-driven decision-making basis can be provided for project managers; decisions based on intuition can be reduced and the scientific nature of decisions can be improved; project costs can be accurately predicted to reduce budget overruns; financial planning and risk mitigation strategy formulation can be supported; the overall risks and resource requirements of multiple projects can be evaluated to support project priority sorting and resource allocation decisions.

[0244] In this embodiment, by implementing this predictive analysis model, the enterprise can transform project management from passive response to active prevention. It not only improves the success rate of individual projects, but also optimizes the overall project portfolio management. This data-driven approach can significantly improve the enterprise's project execution capabilities, resource utilization efficiency, and risk management level, providing key technical support for enterprises to maintain their leading position in the highly competitive market. At the same time, the continuous learning and optimization characteristics of this approach ensure that enterprises can continuously improve their prediction and management capabilities and adapt to the ever-changing business environment.

[0245] In some possible implementations of the present invention, the step of displaying the project progress, team collaboration status and expected project progress in a virtual environment using augmented reality technology based on the current work data and the project forecast results to support remote collaboration and decision-making includes:

[0246] Choose an appropriate AR development platform (such as Unity with ARCore / ARKit), configure the development environment (including SDKs and necessary plug-ins), set up version control and collaborative development tools, and build the AR system;

[0247] Developing a data API for ensuring that the AR system can access the current working data and the project forecast results in real time;

[0248] Create 3D visualization models that represent project progress, virtual models that represent team members and resources, and develop interactive charts and data visualization components in the AR system;

[0249] Create a virtual project manager in the AR system and design the user interface and interaction method;

[0250] Generate project progress data, team collaboration data and project expected progress data based on the current work data and the project forecast results;

[0251] In combination with the project progress data, the team collaboration data and the project expected progress data, a dynamic Gantt chart for displaying the project progress, a 3D network diagram for displaying the team collaboration situation, a visual model for displaying resource allocation and utilization, a risk heat map or early warning indicator for indicating risks and a visual expected progress comparison chart for displaying the expected progress of the project are created in the AR system;

[0252] Configure gesture and voice control functions, virtual sticky notes and annotation functions in AR systems, and create an operational 3D control panel;

[0253] Create multi-user synchronization modules and virtual meeting room modules in AR systems to support real-time collaboration, and create shared whiteboards and drawing tools to support remote discussions;

[0254] Develop interactive analytical tools in AR systems to create resource allocation optimizers, risk assessments, and mitigation strategy generators;

[0255] Develop scenario simulation capabilities in AR systems to display the potential impact of different decisions based on input user decisions.

[0256] In an embodiment of the present invention, it may also include: optimizing 3D rendering performance to ensure a smooth user experience; implementing an efficient data flow and state synchronization mechanism; optimizing network latency to improve the response speed of remote collaboration; implementing end-to-end encryption for data transmission, developing a role-based access control system, and implementing a dynamic blurring function for sensitive information; creating an interactive AR tutorial, developing an online help system and user manual, and designing a progressive learning curve to facilitate user adaptation.

[0257] This embodiment can provide an intuitive 3D project status overview to accelerate information absorption; support spatial display of complex data relationships to enhance the depth of understanding; create an immersive virtual collaborative environment to reduce communication barriers; support real-time sharing and interaction to accelerate the decision-making process; support more scientific decision-making through scenario simulation; provide multi-dimensional data visualization to comprehensively evaluate the impact of decisions; visualize risk status in real time to improve risk awareness; support interactive risk assessment and mitigation strategy formulation; visualize the location and work status of team members to promote collaboration; provide a virtual shared space to enhance team cohesion; track project progress in real time and quickly identify deviations; support parallel monitoring of multiple projects to improve management efficiency; intuitively display resource utilization and support efficient allocation; and quickly adjust and optimize resource plans through interactive tools.

[0258] In this embodiment, by implementing this AR-based project management and collaboration system, enterprises are able to create a highly intuitive and interactive project management environment. This can not only significantly improve the efficiency of remote collaboration and the quality of decision-making, but also bring innovative methods and perspectives to project management. The application of AR technology makes complex project data easier to understand and operate, while also providing team members with a more immersive and interactive working environment. This approach is not only suitable for current project management needs, but also lays the foundation for a more distributed and digital working model in the future.

[0259] In some possible implementations of the present invention, the step of automatically identifying and classifying unstructured information such as meeting minutes and whiteboard content using deep learning-based image recognition technology includes:

[0260] Collect historical meeting records and whiteboard images, annotate the text, charts, and formulas in the whiteboard images, and annotate the types of content;

[0261] Using a deep learning convolutional neural network model to perform feature extraction on the whiteboard image, extracting a first feature that can reflect the content of the image;

[0262] Use the first feature and annotation to supervise the deep learning model (commonly used models include CNN, RNN, etc.) to train a first model that can automatically identify and classify the content of the image;

[0263] The trained first model is used to automatically identify and classify new meeting records and whiteboard contents to obtain a first recognition result.

[0264] This embodiment can realize high-accuracy image content recognition and classification, and has good application effects on meeting record management, whiteboard content extraction, etc. It can replace the traditional manual classification method and improve work efficiency.

[0265] In some possible implementations of the present invention, the step of automatically generating meeting minutes, project reports and work summaries using natural language generation technology includes:

[0266] Collect historical meeting minutes texts, historical project report texts, and historical work summary texts as the first text sample;

[0267] Processing the first text sample (such as word segmentation, stop word removal, etc.) to obtain a second text sample;

[0268] Choose the first deep learning model suitable for natural language processing (such as RCNN, Transformer, etc.);

[0269] Using the second text sample to train the first deep learning model to learn the statistical laws and structural features of the language, thereby obtaining a first report generation model;

[0270] Extracting meeting discussion records, project task records and / or work logs from the first recognition result;

[0271] The meeting discussion records, project task records and / or work logs are input into the first report generation model to obtain the meeting minutes, project reports and / or work summaries in a standard format.

[0272] In this embodiment, through training with a large amount of data, the obtained model has high complexity and high execution efficiency, and the generated text format is standardized, and the content completely covers the input information.

[0273] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

[0274] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0275] In the several embodiments provided in the present application, it should be understood that the disclosed device can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of the above-mentioned units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0276] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0277] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0278] If the above-mentioned integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a memory, including a number of instructions to enable a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the above-mentioned methods of each embodiment of the present application. The aforementioned memory includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or CD-ROM and other media that can store program codes.

[0279] A person skilled in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable memory, and the memory can include: a flash drive, a read-only memory (English: Read-Only Memory, abbreviated as: ROM), a random access memory (English: Random Access Memory, abbreviated as: RAM), a magnetic disk or an optical disk, etc.

[0280] The embodiments of the present application are introduced in detail above. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method and core idea of ​​the present application. At the same time, for general technical personnel in this field, according to the idea of ​​the present application, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

[0281] Although the present invention is disclosed as above, the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions without departing from the spirit and scope of the present invention, and can make various changes and modifications, including the combination of the above-mentioned different functions and implementation steps, including software and hardware implementation methods, all of which are within the scope of protection of the present invention.

Claims

1. An office collaboration system based on artificial intelligence, characterized in that: include: A cloud server, a plurality of IoT servers connected to the cloud server in communication, and a control terminal connected to the IoT server in communication; wherein, The cloud server is configured as follows: Establish an integrated intelligent collaboration platform to connect the information systems of various work departments within the enterprise; Using natural language processing technology, real-time analysis is performed on documents, emails and communication records in the intelligent collaboration platform to extract key information; Based on a machine learning algorithm, the extracted key information is classified, associated and evaluated for importance to obtain key information analysis data; Establishing logical associations between the key information analysis data to construct an enterprise knowledge graph; Through the intelligent collaborative control system, task reminders, decision-making suggestions and knowledge resources are pushed to relevant personnel in various work departments within the enterprise, and various work departments are controlled to collaboratively handle work tasks; Obtain historical and current work data for each work department; Using a predictive analysis model, based on the historical work data and the current work data, potential project risks, resource requirements, and schedule delays are predicted to obtain a project prediction result; Based on the current work data and the project forecast results, using augmented reality technology, the project progress, team collaboration and expected project progress are displayed in a virtual environment to support remote collaboration and decision-making; Image recognition technology based on deep learning can automatically identify and classify unstructured information such as meeting minutes and whiteboard content; Use natural language generation technology to automatically generate meeting minutes, project reports and work summaries; The steps of pushing task reminders, decision suggestions and knowledge resources to relevant personnel of various work departments in the enterprise through the intelligent collaborative control system and controlling various work departments to collaboratively process work tasks include: Obtain product order data for the first product, historical design data for similar products, and market trend data for similar products; Generate initial design data for the first product using machine learning and computer vision technology based on the product order data, the historical design data, and the market trend data; Importing the initial design data into a cloud-based 3D collaborative design module; The 3D collaborative design module generates a plurality of collaborative design tasks and sends the collaborative design tasks to corresponding design terminals; Each of the design terminals generates a corresponding initial design model according to the corresponding collaborative design task; Merging the multiple initial design models to obtain an initial product design model; Using VR technology to create an initial product virtual model corresponding to the initial product design model; Use natural language processing technology to automatically analyze the product technical information of the corresponding products and obtain technology development trend data; Modifying the initial product virtual model according to the evaluation feedback of the initial product virtual model and the technology development trend data to obtain a basic product virtual model; Using artificial intelligence to simulate and optimize the basic product virtual model to obtain a finalized product model; Based on the product finalization model and the product technical information, a research and development knowledge map is constructed to support cross-departmental knowledge sharing and rapid problem location; According to the product finalization model and the product ordering data, a machine learning algorithm is used to predict demand, optimize inventory levels, and generate raw material procurement data; According to the raw material procurement data, based on multi-dimensional data analysis, real-time evaluation and selection of the best suppliers are carried out, and raw material procurement is carried out; Use blockchain technology to achieve full traceability of raw materials to ensure quality and compliance; In the process of manufacturing the first product using raw materials, an optimization algorithm is used to realize intelligent scheduling of the production line to improve production efficiency; During the manufacturing process of the first product, deep learning and computer vision technology are used to perform real-time quality inspection on the product being produced to obtain a first inspection result; Based on the first detection result, using an intelligent collaborative robot, the manufacturing process is controlled to improve production flexibility and efficiency; Using machine learning algorithms to generate defect prediction models based on historical data to predict product defects; The manufactured first product is tested according to the defect prediction model.

2. An office collaboration method based on artificial intelligence, characterized in that: include: Establish an integrated intelligent collaboration platform to connect the information systems of various work departments within the enterprise; Using natural language processing technology, real-time analysis is performed on documents, emails and communication records in the intelligent collaboration platform to extract key information; Based on a machine learning algorithm, the extracted key information is classified, associated and evaluated for importance to obtain key information analysis data; Establishing logical associations between the key information analysis data to construct an enterprise knowledge graph; Through the intelligent collaborative control system, task reminders, decision-making suggestions and knowledge resources are pushed to relevant personnel in various work departments within the enterprise, and various work departments are controlled to collaboratively handle work tasks; Obtain historical and current work data for each work department; Using a predictive analysis model, based on the historical work data and the current work data, potential project risks, resource requirements, and schedule delays are predicted to obtain a project prediction result; Based on the current work data and the project forecast results, using augmented reality technology, the project progress, team collaboration and expected project progress are displayed in a virtual environment to support remote collaboration and decision-making; Image recognition technology based on deep learning can automatically identify and classify unstructured information such as meeting minutes and whiteboard content; Use natural language generation technology to automatically generate meeting minutes, project reports and work summaries; The steps of pushing task reminders, decision suggestions and knowledge resources to relevant personnel of various work departments in the enterprise through the intelligent collaborative control system and controlling various work departments to collaboratively process work tasks include: Obtain product order data for the first product, historical design data for similar products, and market trend data for similar products; Generate initial design data for the first product using machine learning and computer vision technology based on the product order data, the historical design data, and the market trend data; Importing the initial design data into a cloud-based 3D collaborative design module; The 3D collaborative design module generates a plurality of collaborative design tasks and sends the collaborative design tasks to corresponding design terminals; Each of the design terminals generates a corresponding initial design model according to the corresponding collaborative design task; Merging the multiple initial design models to obtain an initial product design model; Using VR technology to create an initial product virtual model corresponding to the initial product design model; Use natural language processing technology to automatically analyze the product technical information of the corresponding products and obtain technology development trend data; Modifying the initial product virtual model according to the evaluation feedback of the initial product virtual model and the technology development trend data to obtain a basic product virtual model; Using artificial intelligence to simulate and optimize the basic product virtual model to obtain a finalized product model; Based on the product finalization model and the product technical information, a research and development knowledge map is constructed to support cross-departmental knowledge sharing and rapid problem location; According to the product finalization model and the product ordering data, a machine learning algorithm is used to predict demand, optimize inventory levels, and generate raw material procurement data; According to the raw material procurement data, based on multi-dimensional data analysis, real-time evaluation and selection of the best suppliers are carried out, and raw material procurement is carried out; Use blockchain technology to achieve full traceability of raw materials to ensure quality and compliance; In the process of manufacturing the first product using raw materials, an optimization algorithm is used to realize intelligent scheduling of the production line to improve production efficiency; During the manufacturing process of the first product, deep learning and computer vision technology are used to perform real-time quality inspection on the product being produced to obtain a first inspection result; Based on the first detection result, using an intelligent collaborative robot, the manufacturing process is controlled to improve production flexibility and efficiency; Using machine learning algorithms to generate defect prediction models based on historical data to predict product defects; The manufactured first product is tested according to the defect prediction model.

3. The office collaboration method based on artificial intelligence according to claim 2, characterized in that: The steps of establishing an integrated intelligent collaboration platform to connect the information systems of various work departments within the enterprise include: Obtain the first attribute information of the existing information system of each work department and the second attribute information of the work process; Determine an integration target and an integration scope according to the first attribute information and the second attribute information; Develop an integration strategy and technology roadmap based on the integration goals and integration scope; Designing a distributed system architecture for the intelligent collaboration platform according to the integration strategy and the technical roadmap, planning a data storage and processing solution for the intelligent collaboration platform, and designing a security architecture and access control mechanism for the intelligent collaboration platform; Formulate unified data standards for the intelligent collaboration platform, establish data mapping relationships between various information systems, and design data conversion and cleaning processes; Develop standardized APIs for the information systems of various work departments, implement real-time data synchronization mechanisms, and develop middleware to handle communications between different information systems; Build a unified user interface for the intelligent collaboration platform, develop a workflow engine, implement unified identity authentication and permission management, and develop data visualization and reporting functions; Integrate natural language processing modules for intelligent search and automatic classification into the intelligent collaboration platform; Embed machine learning algorithms for data analysis and prediction in the intelligent collaboration platform, develop intelligent recommendation systems, and integrate knowledge graph construction and management functions; Conduct unit testing, integration testing, system testing, performance testing, and load testing on the intelligent collaboration platform to optimize system response speed and stability; Develop data migration strategies for the intelligent collaboration platform, perform data cleansing and conversion, and migrate data in batches; Generate user manuals and technical documentation for the intelligent collaboration platform, train users from various work departments, develop and execute a phased deployment plan; Collect user feedback, monitor the system performance of the intelligent collaboration platform, and regularly update and optimize system functions.

4. The office collaboration method based on artificial intelligence according to claim 3 is characterized in that: The step of using natural language processing technology to perform real-time analysis on documents, emails and communication records in the intelligent collaboration platform to extract key information includes: Establish a data interface to obtain documents, emails and communication records in the intelligent collaboration platform in real time and integrate them into the first text; Performing text cleaning on the first text, removing irrelevant characters and punctuation marks, and then performing text standardization to obtain a second text; Using a word segmentation algorithm to segment the second text, and performing part-of-speech tagging to obtain a third text; Identifying and annotating entities and professional terms and keywords in a specific field in the third text; Using a preset text classification model to perform topic classification on the third text; extracting a text summary of the classified third text, and identifying key sentences and keywords as key information from the text summary; Constructing a sentiment analysis model to identify the sentiment tendency of the third text to perform sentiment evaluation and obtain a sentiment evaluation result; Develop an intent recognition model to understand the purpose or demand behind the third text and identify user intentions such as task assignment, problem feedback, and decision requests; Identify the relationship between entities, build the corresponding knowledge graph, and display the logical association between entities; Annotate sentiment evaluation results, user intentions, and logical associations between entities as key information; The annotated key information is stored in a structured database, and a full-text search engine is established for rapid retrieval; Develop data visualization interfaces to display analysis results and implement automatic report generation to summarize key findings; Desensitize data, protect sensitive information, establish access control mechanisms, and ensure information security.

5. The office collaboration method based on artificial intelligence according to claim 4, characterized in that: The step of classifying, associating and evaluating the importance of the extracted key information based on a machine learning algorithm to obtain key information analysis data includes: After data cleaning and standardization of the key information, text features are extracted, metadata features are generated, and context features are created; Select the corresponding classification algorithm, train the multi-label classification model, and classify the key information by topic; Use content-based recommendation algorithms to build association models, and apply graph neural networks to analyze the associations between key information; Develop a multi-factor scoring model that considers the novelty, relevance, and influence of key information to evaluate the importance of key information in the overall text network.

6. The office collaboration method based on artificial intelligence according to claim 5, characterized in that: The step of using the predictive analysis model to predict potential project risks, resource requirements and schedule delays based on the historical work data and the current work data to obtain a project prediction result includes: Integrate the historical working data and the current working data, perform data cleaning to handle missing values ​​and outliers, and perform feature engineering; Use correlation analysis to identify important features from the output of the previous step, apply principal component analysis to reduce the dimensionality of the data, and use domain knowledge to select key predictors; According to the key prediction factors, select corresponding prediction algorithms for different prediction tasks; In combination with the prediction algorithm, develop a deep learning model for complex pattern recognition; Training and validating the deep learning model to obtain a predictive analysis model; Develop automated data input and preprocessing pipelines and implement model inference engines to support batch processing and real-time prediction; Design a visual interface to display prediction results; Develop interactive tools for model interpretation exploration, explain model decisions, and generate feature importance reports; Design a risk scoring system to determine overall project risk and set dynamic thresholds to determine different risk levels; The historical work data and the current work data are input into a predictive analysis model to predict potential project risks, resource requirements and schedule delays to obtain a project prediction result.

7. The office collaboration method based on artificial intelligence according to claim 6, characterized in that: The steps of using augmented reality technology to display project progress, team collaboration status and expected project progress in a virtual environment based on the current work data and the project forecast results to support remote collaboration and decision-making include: Select an appropriate AR development platform, configure the development environment, set up version control and collaborative development tools, and build an AR system; Developing a data API for ensuring that the AR system can access the current working data and the project forecast results in real time; Create 3D visualization models that represent project progress, virtual models that represent team members and resources, and develop interactive charts and data visualization components in the AR system; Create a virtual project manager in the AR system and design the user interface and interaction method; Generate project progress data, team collaboration data and project expected progress data based on the current work data and the project forecast results; In combination with the project progress data, the team collaboration data and the project expected progress data, a dynamic Gantt chart for displaying the project progress, a 3D network diagram for displaying the team collaboration situation, a visual model for displaying resource allocation and utilization, a risk heat map or early warning indicator for indicating risks and a visual expected progress comparison chart for displaying the expected progress of the project are created in the AR system; Configure gesture and voice control functions, virtual sticky notes and annotation functions in AR systems, and create an operational 3D control panel; Create multi-user synchronization modules and virtual meeting room modules in AR systems to support real-time collaboration, and create shared whiteboards and drawing tools to support remote discussions; Develop interactive analytical tools in AR systems to create resource allocation optimizers, risk assessments, and mitigation strategy generators; Develop scenario simulation capabilities in AR systems to display the potential impact of different decisions based on input user decisions.

8. The office collaboration method based on artificial intelligence according to claim 7, characterized in that: The steps of automatically identifying and classifying unstructured information such as meeting minutes and whiteboard contents using deep learning-based image recognition technology include: Collect historical meeting records and whiteboard images, annotate the text, charts, and formulas in the whiteboard images, and annotate the types of content; Using a deep learning convolutional neural network model to perform feature extraction on the whiteboard image, extracting a first feature that can reflect the content of the image; Using the first features and annotations to perform supervised learning on a deep learning model, training a first model that can automatically identify and classify image content; The trained first model is used to automatically identify and classify new meeting records and whiteboard contents to obtain a first recognition result.

9. The office collaboration method based on artificial intelligence according to claim 8, characterized in that: The steps of automatically generating meeting minutes, project reports and work summaries by using natural language generation technology include: Collect historical meeting minutes texts, historical project report texts, and historical work summary texts as the first text sample; Processing the first text sample to obtain a second text sample; Choose the first deep learning model suitable for natural language processing; Using the second text sample to train the first deep learning model to learn the statistical laws and structural features of the language, thereby obtaining a first report generation model; Extracting meeting discussion records, project task records and / or work logs from the first recognition result; The meeting discussion records, project task records and / or work logs are input into the first report generation model to obtain the meeting minutes, project reports and / or work summaries in a standard format.

Citation Information

Cited By

  • Digital human real-time interactive cooperation system and method based on natural language processing

    CN120929039A

  • A digital human real-time interaction and collaboration system and method based on natural language processing

    CN120929039B