Online multi-party management method and system based on enterprise project data

Through an online multi-party management system based on enterprise project data, user project data is collected and processed, feature optimization and cluster analysis is carried out, and correlation evaluation model is established, which solves the problems of enterprise project management process optimization and change adaptation, achieving more efficient project management and success rate improvement.

CN119067364AInactive Publication Date: 2024-12-03HEBEI XIONGAN QISHENGDA TECH SERVICE CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411080571.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-07
Publication Date
2024-12-03
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

It is difficult for enterprises to optimize processes during project management, resulting in a decrease in development speed, and the original management experience cannot be generalized after the project changes, which increases the time and risk of completing new projects.

Method used

Adopting an online multi-party management system based on enterprise project data, through the data management acquisition module, information association module and dynamic management analysis module, collect and process user project data, perform feature optimization, similarity calculation and cluster analysis, and establish correlation evaluation Model, optimize project management.

Benefits of technology

Through data processing of enterprise project information, accurately identify related projects, analyze future development directions, reduce management costs, and improve project completion success rate and efficiency, it is suitable for project management assistance for different enterprises.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119067364A_ABST
    Figure CN119067364A_ABST
Patent Text Reader

Abstract

The invention provides an online multi-party management method and system based on enterprise project data, and the system comprises a data management collection module which is used for registering and logging in a user, collecting and processing user project data, and collecting non-user project information through a programming interface and / or a web crawler; the information association module is used for carrying out characteristic optimization and comprehensive similarity measurement and calculation based on the collected non-user project information and user project information, the future development direction and trend of enterprises are analyzed and judged by analyzing periodic behaviors of the enterprises, and on the basis of the development direction and trend of enterprise users, the enterprise users are subjected to characteristic optimization and comprehensive similarity measurement and calculation. The project related to future development of the user is determined again, and the user can know the project information in advance by updating the project information and prepare for the project needing to be done in the future, so that the user can effectively reduce the management cost in subsequent project management and can improve the success rate and efficiency of project completion, and the user experience is improved. And project management assistance can be performed for different enterprises in a targeted manner.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of enterprise service technology, and in particular to an online multi-party management method and system based on enterprise project data. Background Art

[0002] As the scale of the enterprise continues to expand and the development direction changes, the enterprise needs to complete different projects. There will be great changes in the development process of the enterprise. After the development direction of the enterprise changes, the projects that need to be completed will also change accordingly.

[0003] Currently, when enterprises conduct project management, they basically operate based on the management experience of previous projects. This makes it difficult for enterprises to optimize processes during project management, thereby reducing the speed of enterprise development. Moreover, when the project changes, the original project management experience cannot be universal, which requires enterprises to spend more time to complete new projects, and there is also the risk that the project cannot be completed effectively. Summary of the invention

[0004] 1. Technical issues to be resolved

[0005] In view of the deficiencies of the prior art, the present invention provides an online multi-party management method and system based on enterprise project data to solve the problems raised in the above background technology.

[0006] (II) Technical solution

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: an online multi-party management system based on enterprise project data, comprising:

[0008] Data management and collection module: used to register and log in users, collect and process user project data, and collect non-user project information through programming interfaces and / or web crawlers;

[0009] The information association module performs feature optimization and comprehensive similarity measurement based on the collected non-user project information and user project information, obtains the optimal similarity features and associated projects, and establishes a relevance evaluation model for evaluating projects based on the optimal similarity features;

[0010] Dynamic management analysis module, based on the similarity between user project basic information, project process bias information and related projects, to carry out comprehensive management applications;

[0011] The memory is used to store user information and also to store non-user item information of data collection.

[0012] As a preference of this embodiment, the users and non-users are specifically enterprises, and user project information is collected based on a timeline, wherein the non-user project information includes but is not limited to project information of registered non-user enterprises.

[0013] As a preferred embodiment of this invention, the steps of registering and logging in a user, and collecting and processing user project data include:

[0014] Collect user project information based on user registration information and timeline;

[0015] The user project information is used to construct a user project information feature matrix in units of months. The user project information feature matrix is:

[0016]

[0017] P represents the user-item feature matrix; P 11 The feature data vector representing the first day of the first month of the user project; P M1 The feature data vector representing the first day of the Mth month of the user project; P N1 The feature data vector representing the first day of the Nth month of the user project; P Mj Represents the feature data vector of the jth day of the Mth month of the user project; k represents the total number of days collected in the month.

[0018] As a preferred embodiment of this embodiment, the feature optimization and comprehensive similarity measurement based on the collected non-user project information and user project information to obtain the optimal similarity feature and related projects includes:

[0019] Extract and optimize features based on the feature matrix data of the user project to obtain evaluation features;

[0020] Extracting a number of high-quality process data and / or features from the collected non-user project information as positive samples, and low-quality process data and / or features as negative samples, mixing the positive samples and negative samples as a comprehensive feature sample set, and performing feature optimization on the comprehensive feature sample set to obtain the optimal feature;

[0021] Quantifying the features of the collected non-user project information and collecting a first feature, performing feature dimension reduction on the first feature by using correlation analysis, and obtaining a second feature;

[0022] Based on the optimal feature and the second feature, a mapping update is performed to obtain the optimal similarity feature;

[0023] Similarity measurement and clustering effect evaluation are used to perform cluster analysis on the feature data in the user project and the collected project information feature matrix to obtain the non-user projects most recently associated with the user project.

[0024] As a preferred embodiment of this embodiment, the similarity measurement and clustering effect evaluation are used to perform cluster analysis on the user project and the feature data in the collected project information feature matrix to obtain the user project's most recently associated non-user project, including:

[0025] Similarity measurement and clustering effect evaluation are used to perform cluster analysis on the feature data in the user project and the collected project information feature matrix to obtain the non-user projects most recently associated with the user project, including:

[0026] Select K cluster centers based on the optimal features of the user project and use the cluster center number as the initial cluster vector;

[0027] Calculate the similarity between the data vector in the optimal similarity feature of the non-user item and the initial clustering vector, and assign the data vector in the optimal similarity feature to the cluster with the greatest similarity;

[0028] When all the best similar features of non-user items are clustered, the distance between the data vector of the best similar features of non-user items and the cluster center is calculated;

[0029] Based on the calculation result, the relevance between the non-user item and the user item is determined.

[0030] As a preferred embodiment of this embodiment, the establishment of a correlation evaluation model for evaluating items based on the optimal similarity characteristic includes:

[0031] Taking all the projects of the user as samples, extracting the first feature, extracting the second feature of the user information based on the user project process bias, and using the multi-target retrieval strategy based on the second feature to filter out the associated feature set;

[0032] Based on the acquired associated features and the optimal similarity update mapping, a calculation feature sample set is obtained;

[0033] Extracting evaluation feature sample sets based on the associated project information of user projects obtained through analysis;

[0034] Based on the calculation feature sample set, multiple calculation features are obtained, and a new mapping relationship is established between the calculation features and the evaluation feature sample set to obtain a sample set of evaluation similarity;

[0035] Several sample sets of evaluation similarity data are randomly selected and applied to the machine learning algorithm to obtain a correlation evaluation model for evaluating user projects.

[0036] As a preferred embodiment of this embodiment, the comprehensive management application based on the similarity between the basic information of the user project and the project process bias information and the related projects includes:

[0037] Update related projects based on the completion data and type of user projects, and analyze the changes in user projects;

[0038] In-depth analysis of the types of user projects and the scope and trend of completed data changes, further update the user's related projects, and apply based on the updated related projects.

[0039] As a preferred embodiment of this embodiment, the in-depth analysis of the type of user project and the range and trend of completed data changes, and updating the user's associated projects again includes:

[0040] By extracting interest feature data from the historical process data of user projects and updating it with optimal similarity;

[0041] Extract several independent features based on the latest optimal similarity and adjust the independent features as needed;

[0042] Based on the adjusted independent features and the correlation evaluation model, the correlation between user items and non-user items is obtained.

[0043] As a preferred embodiment of this embodiment, the application based on the updated associated items includes:

[0044] For enterprises with similar project types and stable completion status, relevant information about user project completion will be pushed to users to further improve the completion efficiency of user projects;

[0045] For user-status project types that are completely different, the same type of project information will be pushed to users to help them find project completion methods and processes more quickly. The detailed processes and difficult points of the same type of projects will be pushed to companies that are currently undertaking projects to help users complete projects better.

[0046] As a preferred embodiment of this invention, an online multi-party management method based on enterprise project data includes:

[0047] Collect user information and user project information, and collect project information of non-users;

[0048] Based on the collected non-user project information and user project information, feature optimization and comprehensive similarity measurement are performed to obtain the optimal similarity features and related projects, and a relevance evaluation model for evaluating projects is established based on the optimal similarity features;

[0049] Based on the similarity between the basic information of the user's project, the project process bias information and the related projects, project management is performed according to the user's project status.

[0050] (III) Beneficial effects

[0051] The present invention provides an online multi-party management method and system based on enterprise project data, which has the following beneficial effects: by digitizing the project information of enterprise users, and performing similarity analysis on these data and projects done by other enterprises, the correlation between the user enterprise project and other projects is obtained; by extracting the optimal features of the user project data, and performing similarity measurement and clustering effect evaluation on the characteristics of non-user projects, clustering analysis is performed on the feature data in the feature matrix of user project and non-user project information, and the related projects of the user project can be accurately obtained; at the same time, by analyzing the cyclical behavior of the enterprise, its future development direction and trend are analyzed and determined, and based on the development direction and trend of the enterprise users, the projects related to the user's future development are re-determined; by updating the project information, the user can understand the project information in advance and prepare for the projects to be done in the future, so that the user can effectively reduce the management cost in subsequent project management and improve the success rate and efficiency of project completion, and targeted project management assistance can be provided for different enterprises. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 A connection diagram for big data-based enterprise relationship service management of the present invention;

[0053] Figure 2 A method for obtaining optimal similarity characteristics and associated items for an embodiment of the present invention;

[0054] Figure 3 The present invention provides a method for obtaining the most recently associated items of a user item. DETAILED DESCRIPTION

[0055] The embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be understood as limiting the present invention.

[0056] The disclosure below provides many different embodiments or examples to realize different structures of the present invention. In order to simplify the disclosure of the present invention, the parts and settings of specific examples are described below. Of course, they are only examples, and the purpose is not to limit the present invention. In addition, the present invention can repeat reference numbers and / or reference letters in different examples, and this repetition is for the purpose of simplicity and clarity, which itself does not indicate the relationship between the various embodiments and / or settings discussed. In addition, the present invention provides various specific examples of processes and materials, but those of ordinary skill in the art can appreciate the application of other processes and / or the use of other materials.

[0057] like Figure 1As shown, an embodiment of the present invention provides an online multi-party management system based on enterprise project data, comprising

[0058] Data management and collection module: used to register and log in users, collect and process user project data, and collect non-user project information through programming interfaces and / or web crawlers;

[0059] The information association module performs feature optimization and comprehensive similarity measurement based on the collected non-user project information and user project information, obtains the optimal similarity features and associated projects, and establishes a relevance evaluation model for evaluating projects based on the optimal similarity features;

[0060] Dynamic management analysis module, based on the similarity between user project basic information, project process bias information and related projects, to carry out comprehensive management applications;

[0061] The memory is used to store user information and also to store non-user item information of data collection.

[0062] Specifically, by analyzing user project data and comparing the collected user project information with non-user project information, users can understand the completion process of the same type of projects and the conditions that may be encountered during completion. In addition, through the changes in user projects and the problems encountered during the process, the company pushes information on changed projects and related projects that encounter problems, as well as solutions to changed projects and problems encountered, to provide auxiliary reference for the management of user projects, so that the user project process can proceed smoothly until it is implemented.

[0063] It can be understood that project data includes basic information of the project, full-process data, and problems encountered during the project completion process.

[0064] It should be noted that the data in the memory will be updated periodically.

[0065] Furthermore, the users and non-users are specifically enterprises, and user project information is collected based on the timeline. The non-user project information includes but is not limited to project information of registered non-user enterprises.

[0066] Further, registering and logging in users, and collecting and processing user project data include:

[0067] Collect user project information based on user registration information and timeline;

[0068] The user project information is used to construct a user project information feature matrix in units of months. The user project information feature matrix is:

[0069]

[0070] P represents the user-item feature matrix; P11 The feature data vector representing the first day of the first month of the user project; P M1 The feature data vector representing the first day of the Mth month of the user project; P N1 The feature data vector representing the first day of the Nth month of the user project; P Mj Represents the feature data vector of the jth day of the Mth month of the user project; k represents the total number of days collected in the month.

[0071] It should be noted that the characteristic data may be basic information of the project, the process of the project, important nodes of the project, and other data related to the completion of the project.

[0072] Among them, based on the collected non-user project information and user project information, feature optimization and comprehensive similarity measurement are performed to obtain the optimal similarity features and related projects including:

[0073] Extract and optimize features based on the feature matrix data of the user project to obtain evaluation features;

[0074] Extracting a number of high-quality process data and / or features from the collected non-user project information as positive samples, and low-quality process data and / or features as negative samples, mixing the positive samples and negative samples as a comprehensive feature sample set, and performing feature optimization on the comprehensive feature sample set to obtain the optimal feature;

[0075] The collected non-user project information is feature quantified, and a first feature is collected, and the first feature is reduced in dimension by using correlation analysis to obtain a second feature;

[0076] Based on the optimal feature and the second feature, a mapping update is performed to obtain the optimal similarity feature;

[0077] Similarity measurement and clustering effect evaluation are used to perform cluster analysis on the feature data in the user project and the collected project information feature matrix to obtain the non-user projects most recently associated with the user project.

[0078] It is understandable that the good / bad data of the project process has corresponding data, and the same type of projects will also have good / bad data. By matching these data, the project corresponding to the user project can be matched more accurately.

[0079] Furthermore, similarity measurement and clustering effect evaluation are used to perform cluster analysis on the feature data in the user project and the collected project information feature matrix, and the non-user projects most recently associated with the user project include:

[0080] Similarity measurement and clustering effect evaluation are used to perform cluster analysis on the feature data in the user project and the collected project information feature matrix to obtain the non-user projects most recently associated with the user project, including:

[0081] Select K cluster centers based on the optimal features of the user project and use the cluster center number as the initial cluster vector;

[0082] Calculate the similarity between the data vector in the optimal similarity feature of the non-user item and the initial clustering vector, and assign the data vector in the optimal similarity feature to the cluster with the greatest similarity;

[0083] When all the best similar features of non-user items are clustered, the distance between the data vector of the best similar features of non-user items and the cluster center is calculated;

[0084] Based on the calculation result, the relevance between the non-user item and the user item is determined.

[0085] After clustering, the data will be written into several discrete styles with cluster centers as clusters. Each vector in each cluster represents the distance from the optimal similarity feature of a non-user item to the cluster center. By calculating the length of the distance, the correlation between the non-user item and the user item can be inferred.

[0086] It is understandable that projects related to user projects include, but are not limited to, projects undertaken by companies in the same industry as the product, companies in the upstream and downstream supply chains of the product, and peer companies and suppliers in the company's subsequent development direction.

[0087] Furthermore, a correlation evaluation model for evaluating projects based on the optimal similarity characteristics is established, including:

[0088] Taking all the projects of the user as samples, extract the first feature, extract the second feature of the user information based on the user project process bias, and use the multi-target retrieval strategy based on the second feature to screen out the related feature set;

[0089] Based on the acquired associated features and the optimal similarity update mapping, a calculation feature sample set is obtained;

[0090] Extracting evaluation feature sample sets based on the associated project information of user projects obtained through analysis;

[0091] Based on the calculation feature sample set, multiple calculation features are obtained, and a new mapping relationship is established between the calculation features and the evaluation feature sample set to obtain a sample set of evaluation similarity;

[0092] Several sample sets of evaluation similarity data are randomly selected and applied to the machine learning algorithm to obtain a correlation evaluation model for evaluating user projects.

[0093] Furthermore, based on the similarity between the basic information of the user's project and the project process bias information and the related projects, comprehensive management applications are carried out, including:

[0094] Update related projects based on the completion data and type of user projects, and analyze the changes in user projects;

[0095] In-depth analysis of the types of user projects and the scope and trend of completed data changes, further update the user's related projects, and apply based on the updated related projects.

[0096] Specifically, during the development of user enterprises, managers will constantly adjust the business development direction according to the development status of the enterprise. In this process, different projects need to be completed.

[0097] Among them, when analyzing the development status of the user enterprise, the projects that the enterprise may need to do can be obtained at the same time. In the analysis process, several directions of the development of the user enterprise are first obtained, and the bias and changes of the user enterprise's products or R&D content in a time T are observed. The user enterprise's products or R&D content are marked and marked within this time T. Finally, based on this time T, the company's development status is comprehensively analyzed based on the frequency of the marked content and the proximity to the current time. At this time, the scope and trend of changes in the user enterprise's projects within time T are determined.

[0098] Further, we further analyze the types of user projects and the scope and trend of completed data changes, and update the user's related projects again, including:

[0099] By extracting interest feature data from the historical process data of user projects and updating it with optimal similarity;

[0100] Extract several independent features based on the latest optimal similarity and adjust the independent features as needed;

[0101] Based on the adjusted independent features and the correlation evaluation model, the correlation between user items and non-user items is obtained.

[0102] Further, the application based on the updated associated project includes:

[0103] For enterprises with similar project types and stable completion status, relevant information about user project completion will be pushed to users to further improve the completion efficiency of user projects;

[0104] For user-status project types that are completely different, the same type of project information will be pushed to users to help them find project completion methods and processes more quickly. The detailed processes and difficult points of the same type of projects will be pushed to companies that are currently undertaking projects to help users complete projects better.

[0105] Specifically, by accurately comparing user projects with non-user projects, users can be assisted in managing and optimizing processes for projects that are currently being or will be needed based on completed project data, allowing users to understand project information in advance and prepare for future projects. This will effectively reduce management costs in subsequent project management and improve the success rate and efficiency of project completion.

[0106] This embodiment also provides an online multi-party management method based on enterprise project data, comprising:

[0107] Collect user information and user project information, and collect project information of non-users;

[0108] Based on the collected non-user project information and user project information, feature optimization and comprehensive similarity measurement are performed to obtain the optimal similarity features and related projects, and a relevance evaluation model for evaluating projects is established based on the optimal similarity features;

[0109] Based on the similarity between the basic information of the user's project, the project process bias information and the related projects, project management is performed according to the user's project status.

[0110] The online multi-party management method and system based on enterprise project data provided by the present invention digitizes the project information of enterprise users, and performs similarity analysis on these data and projects done by other enterprises to obtain the correlation between the user enterprise project and other projects. By extracting the optimal features of the user project data and performing similarity measurement and clustering effect evaluation on the characteristics of non-user projects, cluster analysis is performed on the feature data in the feature matrix of user project and non-user project information, and the related projects of the user project can be accurately obtained. At the same time, by analyzing the cyclical behavior of the enterprise, its future development direction and trend are analyzed and determined, and based on the development direction and trend of the enterprise users, the projects related to the user's future development are re-determined. By updating the project information, the user can understand the project information in advance and prepare for the projects to be done in the future, so that the user can effectively reduce the management cost in subsequent project management and improve the success rate and efficiency of project completion.

[0111] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An online multi-party management system based on enterprise project data, characterized by: include Data management and collection module: used to register and log in users, collect and process user project data, and collect non-user project information through programming interfaces and / or web crawlers; The information association module performs feature optimization and comprehensive similarity measurement based on the collected non-user project information and user project information, obtains the optimal similarity features and associated projects, and establishes a relevance evaluation model for evaluating projects based on the optimal similarity features; Dynamic management analysis module, based on the similarity between user project basic information, project process bias information and related projects, to carry out comprehensive management applications; The memory is used to store user information and also to store non-user item information of data collection.

2. The online multi-party management system based on enterprise project data according to claim 1, characterized in that: The users and non-users are specifically enterprises, and user project information is collected based on the timeline. The non-user project information includes but is not limited to project information of registered non-user enterprises.

3. The online multi-party management system based on enterprise project data according to claim 2, characterized in that: The registering and logging in of users, and collecting and processing user project data include: Collect user project information based on user registration information and timeline; The user project information is used to construct a user project information feature matrix in units of months. The user project information feature matrix is: P represents the user-item feature matrix; P 11 The feature data vector representing the first day of the first month of the user project; P M1 The feature data vector representing the first day of the Mth month of the user project; P N1 The feature data vector representing the first day of the Nth month of the user project; P Mj Represents the feature data vector of the jth day of the Mth month of the user project; k represents the total number of days collected in the month.

4. The online multi-party management system based on enterprise project data according to claim 3, characterized in that: The feature optimization and comprehensive similarity measurement based on the collected non-user project information and user project information to obtain the optimal similarity features and related projects include: Extract and optimize features based on the feature matrix data of the user project to obtain evaluation features; Extracting a number of high-quality process data and / or features from the collected non-user project information as positive samples, and low-quality process data and / or features as negative samples, mixing the positive samples and negative samples as a comprehensive feature sample set, and performing feature optimization on the comprehensive feature sample set to obtain the optimal feature; Quantifying the collected non-user project information, collecting a first characteristic, and performing a dimensionality reduction on the first characteristic by using a correlation analysis to obtain a second characteristic; Based on the optimal feature and the second feature, a mapping update is performed to obtain the optimal similarity feature; Similarity measurement and clustering effect evaluation are used to perform cluster analysis on the feature data in the user project and the collected project information feature matrix to obtain the non-user projects most recently associated with the user project.

5. The online multi-party management system based on enterprise project data according to claim 4, characterized in that: The similarity measurement and clustering effect evaluation are used to perform cluster analysis on the user project and the feature data in the collected project information feature matrix, and the non-user projects most recently associated with the user project include: Similarity measurement and clustering effect evaluation are used to perform cluster analysis on the feature data in the user project and the collected project information feature matrix to obtain the user project's most recently associated non-user project, including: Select K cluster centers based on the optimal features of the user project and use the cluster center number as the initial cluster vector; Calculate the similarity between the data vector in the optimal similarity feature of the non-user item and the initial clustering vector, and assign the data vector in the optimal similarity feature to the cluster with the greatest similarity; When all the best similar features of non-user items are clustered, the distance between the data vector of the best similar features of non-user items and the cluster center is calculated; Based on the calculation result, the relevance between the non-user item and the user item is determined.

6. The online multi-party management system based on enterprise project data according to claim 4, characterized in that: The establishment of a correlation evaluation model for evaluating projects based on the optimal similarity characteristic includes: Taking all the projects of the user as samples, extracting the first feature, extracting the second feature of the user information based on the user project process bias, and using the multi-target retrieval strategy based on the second feature to filter out the associated feature set; Based on the acquired associated features and the optimal similarity update mapping, a calculation feature sample set is obtained; Extracting evaluation feature sample sets based on the associated project information of user projects obtained through analysis; Based on the calculation feature sample set, multiple calculation features are obtained, and a new mapping relationship is established between the calculation features and the evaluation feature sample set to obtain a sample set of evaluation similarity; Several sample sets of evaluation similarity data are randomly selected and applied to the machine learning algorithm to obtain a correlation evaluation model for evaluating user projects.

7. The online multi-party management system based on enterprise project data according to claim 4, characterized in that: The integrated management application based on the similarity between the basic information of the user project and the project process bias information and the related projects includes: Update related projects based on the completion data and type of user projects, and analyze the changes in user projects; In-depth analysis of the types of user projects and the scope and trend of completed data changes, further update the user's related projects, and apply based on the updated related projects.

8. The online multi-party management system based on enterprise project data according to claim 7, characterized in that: The in-depth analysis of the type of user project and the range and trend of completed data changes, and updating the user's related projects again include: By extracting interest feature data from the historical process data of user projects and updating it with optimal similarity; Extract several independent features based on the latest optimal similarity and adjust the independent features as needed; Based on the adjusted independent features and the correlation evaluation model, the correlation between user items and non-user items is obtained.

9. The online multi-party management system based on enterprise project data according to claim 8, characterized in that: The application based on the updated associated items includes: For enterprises with similar project types and stable completion status, relevant information about user project completion will be pushed to users to further improve the completion efficiency of user projects; For user-status project types that are completely different, the same type of project information will be pushed to users to help them find project completion methods and processes more quickly. The detailed processes and difficult points of the same type of projects will be pushed to companies that are currently undertaking projects to help users complete projects better.

10. An online multi-party management method based on enterprise project data, characterized by: include Collect user information and user project information, and collect project information of non-users; Based on the collected non-user project information and user project information, feature optimization and comprehensive similarity measurement are performed to obtain the optimal similarity features and related projects, and a relevance evaluation model for evaluating projects is established based on the optimal similarity features; Based on the similarity between the basic information of the user's project, the project process bias information and the related projects, project management is performed according to the user's project status.