Investment and research information sharing and distributing method for group business management and control

Abstract

The invention discloses an investment and research information sharing and distributing method for group business management and control, and relates to the technical field of data sharing, and the method comprises the following specific steps: information collection and integration: collecting an external authority information source and capturing related investment and research information by using a network robot, according to the method, multiple authoritative external information sources are regularly monitored and sorted, investment and research data generated by the internal business departments of the group are collected, the universality and accuracy of the information sources are ensured, meanwhile, intelligent screening is carried out by utilizing a rule-based expert system, and the accuracy of the information sources is improved. According to the method and the system, the pertinence and availability of the information are further improved by combining the social relation data with the user portraits and the project requirements, and the personalized and accurate pushing of the information is realized, so that the group enterprises can quickly obtain the information closely related to the business decision.

Description

Technical Field

[0001] The present invention relates to the field of information sharing technology, and specifically to a method for sharing and distributing investment research information for group business management and control. Background Art

[0002] In the operation and management of modern large-scale group enterprises, investment research information plays a vital role. It can not only help the top management and business departments of the enterprise to make more accurate and scientific decisions, but also optimize business processes and improve overall operational efficiency. With the continuous development of information technology, the group enterprises' demand for investment research information is becoming increasingly diversified and complex, covering macroeconomics, policies and regulations, industry trends, competitor analysis and other aspects. Therefore, how to efficiently collect, integrate, screen and distribute these investment research information so that it can accurately match the actual needs of different positions and business projects within the group has become an important issue that needs to be solved urgently.

[0003] Traditionally, group companies often face many challenges when dealing with investment research information. On the one hand, information sources are scattered and lack a unified and effective integration mechanism, resulting in serious information islands and making it difficult for employees to obtain comprehensive and accurate information. On the other hand, the information distribution process lacks intelligence and personalization, and often adopts a one-size-fits-all approach, which is unable to accurately push information based on factors such as employee job requirements, project characteristics, and social relationships. In addition, traditional recommendation methods rely more on content or user portraits, ignoring the impact of internal employee social relationships on information interest, resulting in low information utilization and failure to meet the actual needs of the group's business management. These problems not only affect the communication efficiency within the group, but also restrict the overall development of the company.

[0004] In response to the above problems, it is necessary to optimize the existing investment research information sharing and distribution method used for group business management. By building a complete information collection and integration mechanism, the effective integration of external and internal information is achieved. By formulating scientific screening rules and labeling methods, and introducing integrated recommendation and distribution strategies based on social relationships, accurate information push and personalized services are achieved. Therefore, it is of great significance to develop a method for sharing and distributing investment research information for group business management that can comprehensively realize the above characteristics. Summary of the invention

[0005] The purpose of the present invention is to make up for the shortcomings of the prior art and to provide a method for sharing and distributing investment and research information for group business management. It can realize accurate and efficient sharing and distribution of investment and research information by constructing a complete set of information collection and integration, screening and labeling, fusion recommendation and distribution, and feedback optimization mechanisms. In the information collection and integration stage, the present invention adopts a combination of external monitoring and internal reporting to ensure the breadth and accuracy of information sources. In the screening and labeling stage, multiple sets of screening rules are formulated in combination with the group's strategic goals and job requirements, and the pertinence and availability of information are improved by combining intelligent screening with manual labeling. In the fusion recommendation and distribution stage, full use is made of the social relationship data of employees within the group, combined with user portraits and project requirements, to realize personalized and accurate push of information. In the feedback optimization stage, by collecting and analyzing employee feedback, the entire information sharing and distribution process is continuously optimized to ensure continuous optimization and applicability of the method.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a method for sharing and distributing investment research information for group business management, the method comprising the following specific steps:

[0007] Information collection and integration: Collect external authoritative information sources and use network robots to crawl relevant investment research information, and collect internal information reported by each business department according to the template and reviewed, and summarize them in a temporary information database to be processed. After regular integration, use the text similarity algorithm to remove duplicate information, and then subdivide it into different business management and control areas according to content relevance to form a total investment research information database;

[0008] Information screening and annotation: The group's business and information management expert team formulates multiple sets of information screening rules based on the group's strategy, business sector direction and job responsibilities, embeds an expert system to screen out information sets to be annotated that meet the rules of each position and project, and uses a weighted evaluation formula based on the importance of information combined with other dimensions. Professionals add annotation information to the information to form an information resource pool, and conduct regular spot checks to ensure accurate annotations.

[0009] Information fusion recommendation and distribution: Collect social relationship data on employee collaboration within departments, project cooperation, mentor-apprentice relationships, and information interaction history from the group's internal office system, use the social relationship weight comprehensive evaluation formula to analyze the potential correlation of employees' information interests in the social relationship network, and build employee user portraits. Formulate comprehensive distribution strategies based on positions, projects, and social relationships, and push information to employees through multiple channels such as office software and SMS reminders;

[0010] Feedback optimization: A feedback portal is set up on the group's internal information platform to collect employees' text feedback on information quality and push rationality, organize it into a data set, analyze the feedback content using the feedback question importance ranking formula, extract common problems and personalized needs, and then adjust the information screening, labeling, integrated recommendation and distribution content accordingly according to the score ranking results to achieve continuous optimization.

[0011] Furthermore, the information collection and integration utilizes a text similarity algorithm to remove duplicate information, and the algorithm formula is: ,in, Display information and Information The similarity value of Represents the total number of feature words in the information text. It is The weight of the feature words, and Characteristic words In Information and Information The frequency of words in .

[0012] Furthermore, the information screening and annotation is combined with other dimensions through a weighted evaluation formula of information importance, and professionals add annotation information to the information to form an information resource pool. The formula is: ,in, Represents the comprehensive score of the importance of the information. , , are the weight coefficients corresponding to different influencing factors, Indicates the degree of relevance between the information and the group's current strategic goals. The score representing the impact of information on the implementation of business projects. It is the attention score of the information in the business field.

[0013] Furthermore, in the information screening and marking step, the information marking information includes importance, timeliness, and related business sectors, wherein the importance is divided into three levels: high, medium, and low. The comprehensive importance score of the information is 8-10 for high importance, 4-7 for medium importance, and 0-3 for low importance. The timeliness is divided into emergency, short-term validity, and long-term reference.

[0014] Furthermore, in the information fusion recommendation and distribution step, a social relationship weight comprehensive evaluation formula is used to analyze the potential association of employees' information interests in the social relationship network, and the formula is: ,in, Represents the comprehensive association weight value between two employees based on multiple social relationships. represents the total number of social relationship types considered, It is The weight coefficient corresponding to the social relationship type, Indicates the relationship between two employees. The closeness score of the social relationship.

[0015] Furthermore, in the information fusion recommendation and distribution step, a comprehensive distribution strategy is formulated based on positions, projects and social relationships, and the strategies include: general distribution based on positions, special distribution based on projects and supplementary distribution based on social relationships. The general distribution based on positions selects information with high importance, strong timeliness and close relevance to positions from the labeled information resource pool according to the user portrait characteristics of different positions and the frequency of position information demand, and formulates a list of investment research information to be pushed to each position. The special distribution based on projects is based on the user portraits of the project team members for the ongoing business projects, and is combined with the project stage and special screening rules to screen out the information. Information that meets project needs will be pushed to project team members at project nodes and on a daily basis, and supplemented and distributed based on social relationships. For each employee, other employees with strong connections with him / her are found in the social relationship network with him / her as the center, and a list of information with frequent interactions and high attention on the investment research information platform is obtained. The list is deduplicated and screened, and information that has been pushed through job or project recommendations is removed. The information is then screened again using the information recommendation relevance screening formula based on the relevance of the information to the employee's job and the projects he / she is involved in. The screened information is integrated into the regular job and project information push at 20%-30% of the total amount of information pushed each time.

[0016] Furthermore, in the information fusion recommendation and distribution step, secondary screening is performed using an information recommendation relevance screening formula based on the relevance of the information to the employee's position and the project he / she is involved in. The formula is: ,in, The comprehensive relevance score of the information recommended by an employee based on social relationships. is the weight coefficient that balances the influence of social relationship interests and the influence of employee business associations. Indicates the relevance score of information based on social relationship interests. It is the score of the business relevance between the information and the employee’s position and the projects he / she is involved in.

[0017] Furthermore, in the feedback optimization step, the feedback content is analyzed using the feedback question importance ranking formula, which is: ,in, Indicates The overall importance score of each feedback question. , , are the weight coefficients corresponding to different influencing factors, Represents the frequency score of the feedback question. It is the score of the impact of feedback questions on the effect of information use. A score indicating how closely the feedback issue is related to the group's core business.

[0018] Compared with the prior art, this method for sharing and distributing investment research information for group business management has the following beneficial effects:

[0019] 1. The present invention ensures the breadth and accuracy of information sources by regularly monitoring and sorting out multiple authoritative external information sources and collecting investment research materials generated by the group's internal business departments. At the same time, it uses a rule-based expert system for intelligent screening and manual labeling to further improve the pertinence and availability of information. In addition, by integrating social relationship data, combining user portraits and project requirements, personalized and accurate information push is achieved, which helps group companies quickly obtain information closely related to business decisions, improve business decision-making efficiency, optimize business processes, and thus promote the overall development of group companies.

[0020] 2. The present invention sets up a convenient feedback portal on the group's internal information platform, so that employees can evaluate and provide feedback on the quality of information and the rationality of the push at any time. These feedback information are regularly collected and organized, and analyzed using natural language processing technology to extract common problems and personalized needs reported by employees. Based on the analysis results, the entire investment research information sharing and distribution method is continuously optimized, which helps the present invention better adapt to the group's changing business needs, improve the accuracy of information recommendations and employee acceptance, and further enhance the group's business management and control effects.

[0021] Other advantages, objectives and features of the present invention will be set forth in part in the following description and, in part, will be apparent to those skilled in the art based on an examination of the following or may be taught from the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0023] Figure 1 This is a process operation diagram of a method for sharing and distributing investment research information for group business management;

[0024] Figure 2 The present invention is a flowchart of a method for sharing and distributing investment research information for group business management. DETAILED DESCRIPTION

[0025] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation mode, structure, characteristics and effects of the present invention are described in detail below in combination with the accompanying drawings and preferred embodiments.

[0026] Embodiment 1

[0027] A large manufacturing group, whose business covers multiple sub-sectors such as automobile manufacturing, mechanical parts production and electronic product manufacturing, has many subsidiaries and departments distributed in different regions. It is carrying out multiple major projects at the same time, such as new automobile research and development projects, high-end mechanical parts production line upgrade projects, etc. There are many employees in various positions in the group, including R&D engineers, marketing personnel, production management personnel, etc. They all need a large amount of timely and accurate investment research information support in the process of business decision-making and project promotion, but the efficiency of information acquisition and distribution has been low and the accuracy is insufficient.

[0028] By arranging a professional information collection team, we regularly sort out external information sources such as the website of the Automobile Industry Research Association, authoritative databases in the field of mechanical engineering, and manufacturing industry policy reports issued by local governments. We use network robots to capture relevant information based on set keywords (such as "new energy vehicle technology trends" and "high-precision parts processing technology", etc.), and manually screen investment research materials on manufacturing industry analysis from professional financial institutions, and summarize them in a temporary information database to be processed. We collect about 2,500 external information each month, and issue a unified investment research information template to each subsidiary and department. We require each R&D department to report on the progress of new product development, technical difficulty analysis and other information every month, the production department to report on production line efficiency data and quality control reports, and the marketing department to provide market research reports, competitive product analysis, etc. After the information review team reviews the authenticity, accuracy and group confidentiality requirements, an average of 2,000 internal information is stored in the temporary information database each month.

[0029] Regular integration, for example, every half month to integrate the temporary pending information database, using text similarity algorithm (For example, setting the weight of feature words When giving core technology related terms a higher weight, the frequency The information is obtained through word segmentation and statistics of the text. After removing duplicate information, it is subdivided into 10 categories according to the business control field, such as automobile R&D, parts production, marketing, etc., to form an integrated investment research information database. The number of information in the database remains stable at around 4,000.

[0030] The Group has organized a team of automotive manufacturing experts, mechanical engineering experts and information management professionals to formulate corresponding information screening rules based on the Group's overall strategy (such as focusing on the development of new energy vehicle manufacturing, improving the self-sufficiency rate of high-end parts, etc.), the key directions of each business segment (automobile R&D focuses on autonomous driving technology, etc.) and the core responsibilities of different positions (R&D positions focus on new technology applications, and marketing positions focus on market trends and customer needs). These rules cover more than 50 positions and 10 key business projects. These rules are embedded in a rule-based expert system to screen the integrated investment research information database. For example, information on the latest R&D results of new energy batteries and breakthroughs in intelligent driving sensor technology for automotive R&D engineer positions is screened to form a collection of information to be annotated. About 2,500 pieces of information to be annotated are generated each month, and 10 professional information annotators are assigned to evaluate the information according to the weighted evaluation formula based on the degree of importance. (Among them, for information related to the core strategic direction of the group, The value is relatively high, and the degree of correlation between information and strategy is manually evaluated. , Impact on project implementation And internal platform attention The information is marked with importance (high, medium and low), timeliness (such as urgent policy adjustment information is marked as urgent), and related business sectors (clearly corresponding to the automobile manufacturing sector, etc.), to form a preliminary available information resource pool. The marked information is randomly checked and reviewed every month, and the marking accuracy rate is maintained at above 92%.

[0031] We extracted data on social relationships from the group’s internal office system, including departmental collaboration records (such as records of different teams in the automotive R&D department collaborating to overcome technical difficulties), project cooperation records (employees’ collaboration and task details in various business projects), mentor-apprentice relationship information (old engineers’ guidance records for new employees), and employee interaction history on the group’s internal investment and research information platform (likes, comments, and forwarding of information). We collected more than 8,000 departmental collaboration records, project cooperation records involving more than 300 projects, mentor-apprentice relationship information covering more than 500 pairs of mentors and apprentices, and more than 150,000 interaction history records. We used the social relationship weight comprehensive evaluation formula to comprehensively evaluate the data. (Set the weight of the master-apprentice relationship The weight of the cooperation relationship in important projects is 0.8, and the weight of cooperation relationship in general projects is 0.4-0.6. The weight of frequent cooperation relationship in departments is 0.3-0.5. The weight is quantified by the number of projects and the frequency of cooperation. ) Analyze the potential connections between employees in terms of investment research information interests.

[0032] Obtain basic employee information from the group's human resources management system, and build user portraits based on the employees' roles in past projects and their preferences for browsing and downloading information. For ongoing business projects (such as members of the new car R&D project team), add information such as project responsibilities and key information needs (such as the need for new car body material application information) to their user portraits. Formulate a comprehensive distribution strategy based on positions, projects and social relationships. For general distribution based on positions, select information from the labeled information resource pool that is highly important, timely and closely related to the position based on the user portraits of different positions every day. For example, push 8-10 pieces of information covering new technologies and new material applications to the car R&D position every day. For special project-based distribution, push information that meets project needs, such as the latest configuration of competing models and changes in potential customers' demand for car functions, to project team members at key nodes of new car R&D projects (such as concept design and prototype trial production stages) and on a weekly basis. Push 15-20 pieces of information each time. For supplementary distribution based on social relationships, use the information recommendation relevance screening formula (set up 0.3, determined based on the interaction between related employees , determined based on the degree of match between information, positions and projects ) Filter information and incorporate about 25% of it into regular push notifications. For example, if peers in a certain R&D personnel's social network frequently interact with information on a new manufacturing process and the information is related to his position and project, the information will be recommended to him. The information will be pushed to employees through the unified office software used within the group (such as enterprise WeChat). The push message contains the title, summary, importance and timeliness of the information. For important and urgent information such as the government's sudden introduction of major manufacturing industry support policies, SMS reminders are used to ensure that relevant personnel are aware of and view it in a timely manner, which effectively improves the efficiency of information transmission and employees' attention to information.

[0033] A convenient feedback portal is set up on the group's internal information platform. After using investment research information, employees can provide feedback at any time on information quality (such as accuracy and practicality of information content), rationality of push (whether it meets job or project requirements, and whether the push frequency is appropriate), etc. About 400 employee feedback information is collected each month, and the feedback question importance ranking formula is used. (Set according to the group's emphasis on feedback factors , , Weight, analysis feedback problem frequency , the degree of influence on the effect of information use and the closeness of the relationship with the core business of the Group ) analyzed the feedback information and extracted common problems such as inaccurate timeliness of some information markings, slightly redundant content of information push for certain positions, and personalized needs of individual R&D personnel who wanted to increase international cutting-edge manufacturing process information. According to the analysis results, timely adjustments were made to information screening and marking rules (such as optimizing the timeliness marking judgment method), integrated recommendation and distribution strategies (such as adjusting the push cycle of some positions, optimizing the weight parameters of social relationship recommendations), and other links. After continuous optimization, employees' satisfaction with investment research information increased from the initial 75% to 88%, which effectively promoted the smooth development of various business projects of the group and the improvement of the overall business control level.

[0034] Embodiment 2

[0035] A comprehensive financial services group, whose businesses cover banking, securities, insurance, asset management and other fields, has branches in many cities across the country. Its employees include financial analysts, financial advisors, risk control specialists, insurance actuaries, etc. The group is promoting a number of important business initiatives, such as developing new financial products, expanding the market share of insurance business and optimizing asset allocation strategies. Employees in various positions need accurate and timely investment research information to assist in business decision-making and business development. However, the previous information distribution lacked specificity and could not meet diverse needs.

[0036] A dedicated person is responsible for regularly monitoring authoritative external information sources such as research reports of major stock exchanges and websites of well-known financial research institutions. Based on the set keywords (such as "monetary policy adjustment trends", "analysis of hot sectors in the securities market", "insurance industry innovative product cases", etc.), network robot technology is used to crawl information. After preliminary screening to remove duplicate and incorrect format content, it is summarized into a temporary information database to be processed. About 3,000 external information are collected each month. Each business department (banking department reports deposit and loan business data and market analysis, securities department provides transaction data and investment strategy research, insurance department uploads new product development progress and market feedback, asset management department reports asset allocation and return analysis, etc.) regularly uploads internal information according to a unified investment research information template. After strict review by the information review team (to ensure data authenticity and compliance with financial regulatory requirements, etc.), an average of 2,500 internal information is stored in the temporary information database each month. The information in the temporary information database is integrated every half a month and the text similarity algorithm is used to collect information. (Set appropriate feature word weights for key terms in the financial field , by text processing word frequency statistics ) removes duplicate information, and then subdivides it into eight categories according to business control areas, including banking business, securities business, insurance business, and asset management, to form an integrated investment research information database. The amount of information in the database is stable at around 5,000.

[0037] The Group organizes financial industry experts and information management teams to formulate information screening rules based on the Group's overall strategy (such as strengthening comprehensive financial service capabilities, improving the market competitiveness of each business segment, etc.), the key development directions of each business segment (securities business focuses on investment opportunities in emerging industries, insurance business focuses on health insurance product innovation, etc.) and the core responsibilities of different positions (financial analysts focus on macroeconomic and market trend analysis, financial advisors focus on the applicability of financial products, etc.), covering more than 60 positions and 8 key business projects. The screening rules are embedded in the expert system to screen the integrated investment research information database, such as screening out macroeconomic data interpretation for financial analyst positions, industry sector trend forecasts and other information to form a collection of information to be labeled. About 3,000 pieces of information to be labeled are generated each month, and 8 professional information labelers are arranged to use a weighted evaluation formula according to the degree of importance. (Combined with the group's strategic considerations, appropriately improve the information that affects strategic decisions Value, through manual evaluation of information and group strategy 、Influence on business project implementation And the attention within the group The score is determined by the importance of information annotation (high, medium, or low), as well as the timeliness (urgent changes in regulatory policies are marked as urgent) and related business segments (clearly corresponding to specific financial business segments), to form a preliminary available information resource pool. The labeled information is randomly inspected and reviewed every month, and the labeling accuracy is maintained at above 90%.

[0038] From the group's internal office system, we collected records of intra-departmental collaboration between employees (such as records of collaborative processing of large loan business within the bank's credit department), project cooperation records (employee cooperation in various financial product development projects), information on mentor-apprentice relationships (business guidance records for newcomers from senior financial experts), and employee interaction history on the group's internal investment research information platform (comments on and forwarding of financial information, etc.). We collected more than 6,000 departmental collaboration records, project cooperation records involving more than 200 projects, mentor-apprentice relationships covering more than 400 pairs of mentors and apprentices, and 120,000 interaction history records. We used the social relationship weight comprehensive evaluation formula (Set the weight of the master-apprentice relationship The weight of the important project cooperation relationship is 0.6-0.8 according to the importance of the project and the duration of cooperation. The weight of the general project cooperation relationship is set at 0.4-0.6, and the weight of the frequent cooperation relationship within the department is set at 0.3-0.5. The relevant cooperation indicators are determined by quantifying ) Analyze the potential connections between employees in terms of investment research information interests.

[0039] Obtain basic employee information from the group's human resources management system, build user portraits based on their participation in past business projects and information browsing preferences, and add specific project-related information (such as key information needs such as the characteristics of the project's target customer group and product return expectations) to the user portraits of ongoing business projects (such as members of a new wealth management product development project team). Formulate a comprehensive distribution strategy based on positions, projects, and social relationships. For general distribution based on positions, select information that is highly important, timely, and closely related to the position from the labeled information resource pool based on user portraits of different positions every week. For example, push 10-15 pieces of information covering macroeconomic analysis and in-depth industry research to financial analysts every week. For special project-based distribution, push information that meets project needs such as the characteristics of competing wealth management products and changes in target customer investment preferences to project team members at key nodes of new wealth management product development projects (such as product design and market testing stages) and regularly every two weeks. Push 12-18 pieces of information each time. For supplementary distribution based on social relationships, use the information recommendation correlation screening formula (set up 0.25, determined based on the interaction between related employees , determined based on the degree of match between information, positions and projects Information is filtered and incorporated into regular push notifications at a ratio of about 20%. For example, if peers in a financial advisor's social network frequently interact with information about a certain type of innovative financial product and the information is highly matched with his position and project, it will be recommended to him. Information is pushed to employees through the commonly used office software within the group. The push message contains the title, summary, importance and timeliness of the information. For important and urgent information such as the sudden adjustment of monetary policy by the central bank, SMS reminders are also used to ensure that relevant personnel can know and view it at the first time, which enhances the timeliness and effectiveness of information transmission and enables employees to adjust their business strategies in time according to the information.

[0040] A feedback portal is set up on the group's internal information platform, where employees can provide feedback at any time on the quality of investment research information (such as the accuracy of the information, the degree of practicality and help to the business), the rationality of push (whether it meets the job and project requirements, and whether the push frequency is appropriate). About 350 employee feedback information is collected each month, and the feedback question importance ranking formula is used to rank the feedback questions. (Reasonably set according to the group's emphasis on each feedback factor , , Weight, analysis feedback problem frequency , the degree of influence on the effect of information use and the closeness of the relationship with the core business of the Group ) analyzed the feedback data and extracted common problems such as the importance of some information being marked inconsistent with the actual situation, the timeliness of some job information being pushed in an untimely manner, and the personalized needs of individual insurance actuaries who wanted to increase information on risk assessment cases for specific types of insurance. Based on the analysis results, the company made targeted adjustments to the information screening and marking rules (such as recalibrating the importance scoring standards), integrated recommendation and distribution strategies (such as optimizing the timing of some job information push, and adjusting relevant parameters for social relationship recommendations), and other links. After continuous optimization, employees' satisfaction with investment research information increased from the initial 72% to 86%, significantly improving the utilization efficiency of investment research information within the group, and providing strong support for the steady development of each business segment and the overall business management of the group.

[0041] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technical personnel in this field can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any brief modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. A method for sharing and distributing investment research information for group business management, characterized in that: The system consists of the following components: Information collection and integration: Collect external authoritative information sources and use network robots to crawl relevant investment research information, and collect internal information reported by each business department according to the template and reviewed, and summarize them in a temporary information database to be processed. After regular integration, use the text similarity algorithm to remove duplicate information, and then subdivide it into different business management and control areas according to content relevance to form a total investment research information database; Information screening and annotation: The group's business and information management expert team formulates multiple sets of information screening rules based on the group's strategy, business sector direction and job responsibilities, embeds an expert system to screen out information sets to be annotated that meet the rules of each position and project, and uses a weighted evaluation formula based on the importance of information combined with other dimensions. Professionals add annotation information to the information to form an information resource pool, and conduct regular spot checks to ensure accurate annotations. Information fusion recommendation and distribution: Collect social relationship data on employee collaboration within departments, project cooperation, mentor-apprentice relationships, and information interaction history from the group's internal office system, use the social relationship weight comprehensive evaluation formula to analyze the potential correlation of employees' information interests in the social relationship network, and build employee user portraits. Formulate comprehensive distribution strategies based on positions, projects, and social relationships, and push information to employees through multiple channels such as office software and SMS reminders; Feedback optimization: A feedback portal is set up on the group's internal information platform to collect employees' text feedback on information quality and push rationality, organize it into a data set, analyze the feedback content using the feedback question importance ranking formula, extract common problems and personalized needs, and then adjust the information screening, labeling, integrated recommendation and distribution content accordingly according to the score ranking results to achieve continuous optimization.

2. According to claim 1, a method for sharing and distributing investment research information for group business management and control, characterized in that: The information collection and integration uses a text similarity algorithm to remove duplicate information, and the algorithm formula is: ,in, Display information and Information The similarity value of Represents the total number of feature words in the information text. It is The weight of the feature words, and Characteristic words In Information and Information The frequency of words in .

3. According to claim 1, a method for sharing and distributing investment research information for group business management and control, characterized in that: The information screening and annotation is combined with other dimensions through a weighted evaluation formula of information importance, and professionals add annotation information to the information to form an information resource pool, and the formula is: ,in, Represents the comprehensive score of the importance of the information. , , are the weight coefficients corresponding to different influencing factors, Indicates the degree of relevance between the information and the group's current strategic goals. The score representing the impact of information on the implementation of business projects. It is the attention score of the information in the business field.

4. A method for sharing and distributing investment research information for group business management and control according to claim 3, characterized in that: In the information screening and marking step, the information marking information includes importance, timeliness, and related business sectors, wherein importance is divided into three levels: high, medium, and low. The comprehensive importance score of information is 8-10 for high importance, 4-7 for medium importance, and 0-3 for low importance. Timeliness is divided into emergency, short-term validity, and long-term reference.

5. The method for sharing and distributing investment research information for group business management according to claim 1, characterized in that: In the information fusion recommendation and distribution step, the social relationship weight comprehensive evaluation formula is used to analyze the potential association of employees' information interests in the social relationship network, and the formula is: ,in, Represents the comprehensive association weight value between two employees based on multiple social relationships. represents the total number of social relationship types considered, It is The weight coefficient corresponding to the social relationship type, Indicates the relationship between two employees. The closeness score of the social relationship.

6. The method for sharing and distributing investment research information for group business management according to claim 1, characterized in that: In the information fusion recommendation and distribution step, a comprehensive distribution strategy is formulated based on positions, projects and social relationships, and the strategies include: general distribution based on positions, special distribution based on projects and supplementary distribution based on social relationships. The general distribution based on positions selects information with high importance, strong timeliness and close relevance to positions from the marked information resource pool according to the user portrait characteristics of different positions and the frequency of position information demand, and formulates a list of investment research information to be pushed to each position. The special distribution based on projects is targeted at ongoing business projects, and based on the user portraits of project team members, the project stage and special screening rules are combined to screen out matching projects. Information on project needs is pushed to project team members at project nodes and on a daily basis, and supplemented and distributed based on social relationships. For each employee, other employees with strong connections with him / her are found in the social relationship network with him / her as the center, and a list of information with frequent interactions and high attention on the investment research information platform is obtained. The list is deduplicated and screened, and information that has been pushed through job or project recommendations is removed. The information is then screened again using the information recommendation relevance screening formula based on the relevance of the information to the employee's job and the projects he / she is involved in. The screened information is integrated into the regular job and project information push at 20%-30% of the total amount of information pushed each time.

7. A method for sharing and distributing investment research information for group business management and control according to claim 6, characterized in that: In the information fusion recommendation and distribution step, secondary screening is performed using the information recommendation relevance screening formula based on the relevance of the information to the employee's position and the project he / she is involved in. The formula is: ,in, The comprehensive relevance score of the information recommended by an employee based on social relationships. is the weight coefficient that balances the influence of social relationship interests and the influence of employee business associations. Indicates the relevance score of information based on social relationship interests. It is the score of the business relevance between the information and the employee’s position and the projects he / she is involved in.

8. The method for sharing and distributing investment research information for group business management according to claim 1, characterized in that: In the feedback optimization step, the feedback content is analyzed using the feedback question importance ranking formula, which is: ,in, Indicates The overall importance score of each feedback question. , , are the weight coefficients corresponding to different influencing factors, Represents the frequency score of the feedback question. It is the score of the impact of feedback questions on the effect of information use. A score indicating how closely the feedback issue is related to the group's core business.

Images