Scientific research project management system and method based on big data processing technology

Abstract

The invention discloses a scientific research project management system based on big data processing, and relates to the technical field of project management. The data acquisition module comprises a plurality of key modules, the data acquisition module is connected with multiple sources, acquires project data greater than or equal to 10Hz, cleans the project data, transmits the cleaned project data to a storage center adopting a Hadoop distributed architecture through ETL (Extract Transform Load), and establishes an index and compresses and backs up Snappy according to multiple dimensions; the data analysis module performs clustering and mining association based on a Spark framework; the project progress monitoring module is used for capturing key nodes, displaying the key nodes by using a Gantt chart, and lagging 20% early warning pushing; the personnel management module records personnel information, intelligently recommends and evaluates performance; the fund management module tracks the fund and gives out a report early warning when the fund deviates from the budget by 15%; and the achievement management module collects and arranges achievements, mines and compares to promote transformation. According to the system, scientific research projects are comprehensively and intelligently managed, data are accurately collected and efficiently analyzed, the progress is monitored in real time, personnel funds are reasonably managed and controlled, achievement transformation is assisted, and scientific research benefits are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of project management systems, and particularly to a scientific research project management system and method based on big data processing technology. Background Art

[0002] In the current era of rapid technological development, scientific research projects have shown an explosive growth, with their scale becoming increasingly large and complexity continuously rising. The traditional scientific research project management methods have been difficult to adapt to this new situation and have exposed many drawbacks.

[0003] From the perspective of data management, scientific research projects involve a large amount of data, including basic project information, research progress, personnel input, fund usage, etc. However, these data are often scattered in various isolated systems. For example, academic databases act independently, the internal systems of scientific research institutions lack integration, and the data on project application platforms are updated laggingly. This not only makes data collection difficult and consumes a large amount of manpower for summarization and collation, but also greatly reduces the data accuracy, with frequent occurrence of duplicate and incorrect data, seriously affecting subsequent analysis and decision-making.

[0004] In the project monitoring link, it mostly relies on manual regular reports and rough inspections, lacking real-time and accuracy. Project leaders are difficult to intuitively and timely grasp the project progress and often only know after problems occur, delaying the opportunity for remedy. For example, when the progress of a key experimental node lags behind, due to the failure to detect it in time, the subsequent links are not smoothly connected, and the entire project cycle is extended.

[0005] There are also dilemmas in personnel management. The team is formed only based on experience or personal relationships, and it is impossible to accurately match professional talents according to project needs, resulting in waste of human resources or obstacles to project progress. The performance evaluation has mostly subjective factors and is difficult to objectively measure the contributions of scientific research personnel, hitting the enthusiasm for scientific research.

[0006] The fund management is extensive, the financial records are chaotic, it is difficult to track the flow of each expenditure, and it is impossible to accurately calculate the input-output ratio. The phenomenon of cost overrun is common, but it is difficult to give timely warnings and control.

[0007] In terms of achievement management, after the scientific research achievements are produced, there is a lack of effective sorting and promotion mechanisms. A large number of excellent achievements are buried, the knowledge exchange is not smooth, and the transformation efficiency is extremely low, unable to maximize the scientific research value.

[0008] In summary, there is an urgent need for an innovative scientific research project management system, method and medium relying on big data processing technology to break the existing deadlock, improve the efficiency of scientific research project management, and promote the vigorous development of scientific research undertakings. Summary of the Invention

[0009] A scientific research project management system and method based on big data processing technology proposed by the present invention are used to solve the problems mentioned in the above existing technologies.

[0010] To achieve the above object, the present invention adopts the following technical solutions: A scientific research project management system based on big data processing technology, comprising:

[0011] Data acquisition module: Connect to scientific research data sources, academic databases, internal systems of scientific research institutions, and project application platforms. Set the data acquisition frequency as f and the acquisition cycle adjustment coefficient as k. When the data source update rate accelerates and the system load L increases and satisfies L > L0, where L0 is a preset load reference value, the acquisition frequency is increased according to the formula f = f0 + k×(L - L0), where f0 is the initial acquisition frequency; use a rule-based screening algorithm to remove error data;

[0012] Data storage center: Adopt the Hadoop Distributed File System HDFS to store data in categories and establish indexes according to the dimensions of projects, time, and disciplines; set the number of storage nodes as n and the data redundancy as r. According to the formula where m is the number of data backup copies, reasonably set the redundancy. When storing, use the Snappy compression algorithm, and set the compression ratio as c. After testing and optimization, it has the functions of data backup and recovery;

[0013] Data analysis module: Based on the Spark big data analysis framework, use the clustering analysis algorithm to classify according to research directions and technical difficulty characteristics, introduce a dynamic weight adjustment mechanism. Set the scientific research hotspot change rate as Δh and the feature weight adjustment step as s. When Δh > Δh0, where Δh0 is the hotspot change threshold, according to the formula w i = w i0 + s×Δh, w i is the implementability of the adjusted feature;

[0014] Project progress monitoring module: By collecting project node data in real time, combining with the project plan schedule, visually display the project progress using a Gantt chart, set the progress warning threshold, and set the warning delay time as T w , to ensure timeliness, T w does not exceed 5 seconds. By optimizing the warning push system architecture and adopting the asynchronous message queue combination technology, reduce the time from warning trigger to delivery;

[0015] Personnel management module: Record the professional skills, project participation experience, and achievement contribution information of scientific research personnel, and use a recommendation algorithm to match the most suitable team members for new projects according to project requirements; set the personnel skill matching degree as m s , the project experience matching degree as m p , and the comprehensive matching degree M = α×m s + β×m p, where α and β are weight coefficients, which are set according to the key requirements of the project. For technology R & D projects, α > β; for application promotion projects, β > α. Members are recommended in descending order; personnel performance evaluation is based on data, including the number of papers published n p , the number of patent applications n a , the conversion rate of project achievements r c and other indicators. The performance score P = γ × n p + δ × n a + ∈ × r c , where γ, δ, and ∈ are the corresponding indicator weights, which are set according to the project type and institutional orientation;

[0016] Fund management module: Track the flow of project funds in real time, connect with the financial system, record every expense, and use the cost-benefit analysis model to calculate the input-output ratio of the project; Let the input funds be I and the output income be O, and the input-output ratio When the fund usage deviates from the budget by 15%, issue a warning and generate a detailed financial statement for review;

[0017] Achievement management module: Automatically collect and organize papers, patents, and software copyright achievements generated by scientific research projects, and extract achievements using text mining technology; Let the innovation score of the achievement be C and the influence score be I m , and the comprehensive evaluation score where are weight coefficients.

[0018] Furthermore, it also includes a risk assessment module: Based on historical data and real-time dynamic information, construct a risk prediction model in the dimensions of technology, market, and personnel, and use Monte Carlo simulation to quantify the risk probability and loss degree; Let the technology risk probability be P t , and the loss degree be L t ; the market risk probability is P m , and the loss degree is L m ; the personnel risk probability is P p , and the loss degree is L p ; the total risk probability P total = ω t × P t + ω m × P m + ω p × P p , where ω t , ω m , ω p are the corresponding risk weights, which are set according to the project stage and industry characteristics.

[0019] Furthermore, it also includes a user interaction module: providing a terminal operation interface for researchers, project leaders, and managers, accessed through mobile terminals or Web terminals, supporting voice commands, gesture operations, the time interval from operating commands to receiving execution feedback, and the total operation feedback delay T op =max, the mobile APP adopts a simple interface design and is adapted to the mobile phone system.

[0020] Furthermore, the data analysis module introduces a dynamic weight adjustment mechanism during analysis. According to the real-time changes in scientific research hotspots, the weights of different features in the cluster are adjusted, and the data is updated every hour. If the hotspot change rate Δh exceeds the preset Δh0, the weight adjustment process is started according to the formula w i =w i0 +s×Δh, where w i is the weight of the adjusted feature, w i0 is the initial weight.

[0021] Furthermore, a method for scientific research project management based on big data processing technology is applied, comprising the following steps:

[0022] Data collection and preprocessing: The data collection module collects and screens scientific research project data, and the processed data is stored in the data storage center;

[0023] Data analysis and decision-making: The algorithm of the data analysis module mines the stored data, formulates project management strategies based on the analysis results, adjusts the project schedule, optimizes staffing, and reasonably allocates funds;

[0024] Project monitoring and feedback: The progress monitoring module, personnel management module, and funding management module monitor the project status in real time, collect feedback information, and make timely adjustments to any problems that arise;

[0025] Results management and transformation: The management module manages scientific research results, promotes the sharing and transformation of results, and enhances the overall value of scientific research projects.

[0026] Compared with the prior art, the present invention has the following beneficial effects:

[0027] Data collection is efficient and accurate, multi-source data is aggregated in real time, and redundancy is removed to ensure data quality, laying a solid foundation for decision-making. The storage center uses a distributed architecture to store massive amounts of data in an orderly manner, with convenient indexing, compression efficiency, and security.

[0028] The analysis module digs deep into data associations, clustering and association rule mining, keeps up with the dynamic adjustment of scientific research hotspots, provides high-value insights, and accurately assists decision-making. Progress monitoring is visualized in real time, with sensitive early warnings and Gantt chart interactive operations, allowing for quick adjustment of tasks to prevent problems before they occur.

[0029] Personnel management is intelligent and fair, precisely matching members, scientifically quantifying performance, and stimulating scientific research vitality. Funding management is meticulous and transparent, with real-time tracking of fund flows, accurate calculation of cost-effectiveness, and timely warnings to control costs.

[0030] Outcome management is innovative and powerful. Blockchain protects copyrights and promotes sharing, with precise push and high conversion rates. The interaction module is convenient and diverse, with smooth operation, enhancing the efficiency and quality of scientific research project management throughout the process. Brief Description of the Drawings

[0031] Figure 1 It is a schematic block diagram of a scientific research project management system based on big data processing technology proposed by the present invention;

[0032] Figure 2 It is a schematic block diagram of a scientific research project management method based on big data processing technology proposed by the present invention. Detailed Embodiments

[0033] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0034] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "transverse", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", etc. indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as a limitation of the present invention.

[0035] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more of the said features. In the description of the present invention, "a plurality of" means two or more unless otherwise specifically defined. In addition, the terms "installed", "connected" and "joined" should be understood in a broad sense. For example, it may be a fixed connection, a detachable connection or an integral connection; it may be a mechanical connection or an electrical connection; it may be directly connected or indirectly connected through an intermediate medium, and it may be the communication inside two components. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances. The present invention will be further described in detail below with reference to the accompanying drawings.

[0036] Referring to Figure 1-2 : A scientific research project management system based on big data processing technology, comprising:

[0037] Data acquisition module: Connect to various scientific research data sources, such as academic databases, internal systems of scientific research institutions, project application platforms, etc., and collect data such as basic project information, research progress, personnel input, and fund usage at a frequency not lower than 10Hz. Let the data acquisition frequency be f and the acquisition cycle adjustment coefficient be k. When the data source update rate accelerates and the system load L increases and satisfies L > L0 (L0 is the preset load reference value), the acquisition frequency is dynamically increased according to the formula f = f0 + k×(L - L0), where f0 is the initial acquisition frequency; use a rule-based cleaning algorithm to remove duplicate and incorrect data. For example, set the duplicate data judgment rule that if the data content similarity exceeds 90%, it is determined as duplicate, and the incorrect data is identified based on the predefined data format and value range, and the cleaned data is transmitted to the data storage center through an ETL tool.

[0038] Data storage center: Adopt the Hadoop Distributed File System (HDFS) to classify and store massive scientific research data, and establish indexes according to dimensions such as projects, time, and disciplines. Let the number of storage nodes be n and the data redundancy be r. To ensure data reliability, set the redundancy reasonably according to the formula (m is the number of data backup copies), usually taking values between 2 and 3; use the Snappy compression algorithm during storage, set the compression ratio to c, and after multiple tests and optimizations, keep it within the range of 3 - 5 to improve the storage efficiency, and at the same time have the functions of data backup and recovery to ensure data security.

[0039] Data analysis module: Based on the Spark big data analysis framework, it conducts multi-dimensional in-depth mining on the stored data. It classifies scientific research projects according to characteristics such as research direction and technical difficulty using clustering analysis algorithms, and introduces a dynamic weight adjustment mechanism. Let the change rate of scientific research hotspots be Δh, and the adjustment step size of feature weights be s. When Δh > Δh0 (Δh0 is the hotspot change threshold), according to the formula w i = w i0 + s×Δh (w i is the implementability of the adjusted feature.

[0040] Project progress monitoring module: By collecting key node data of the project in real time, combined with the project plan schedule, it visually displays the project progress using a Gantt chart, sets a progress warning threshold. For example, if the actual progress of a certain stage lags behind the planned progress by 20%, it automatically triggers a warning and pushes it to the project leader and management personnel. Let the warning delay time be T w , to ensure timeliness, T w does not exceed 5 seconds. By optimizing the warning push system architecture and using an asynchronous message queue combined with multi-threaded technology, it reduces the time from warning trigger to delivery.

[0041] Personnel management module: Records information such as the professional skills, project participation experiences, and achievement contributions of scientific research personnel, and uses intelligent recommendation algorithms to match the most suitable team members for new projects according to project requirements. Let the personnel skill matching degree be m s , the project experience matching degree be m p , and the comprehensive matching degree M = α×m s + β×m p , where α and β are weight coefficients, set according to the key requirements of the project. For example, in technology R & D projects, α > β, and in application promotion projects, β > α. It recommends members in descending order; Personnel performance evaluation is based on objective data, such as the number of papers published n p , the number of patent applications n a , the project achievement conversion rate r c and other indicators. The performance score P = γ×n p + δ×n a + ∈×r c (γ, δ, ∈ are the corresponding indicator weights, set according to the project type and institutional orientation), to achieve fair and just evaluation.

[0042] Fund management module: Real-time tracks the flow of project funds, interfaces with the financial system, classifies and records each expense, and uses a cost-benefit analysis model to calculate the input-output ratio of the project. Let the input funds be I and the output benefits be O, and the input-output ratio When the fund usage deviates from the budget by more than 15%, it issues a warning and generates a detailed financial statement for review. Let the statement generation time be T r, by optimizing the report generation algorithm and adopting data pre-aggregation and templated generation technologies, make T r control.

[0043] Achievement management module: Automatically collect and organize achievements such as papers, patents, and software copyrights generated by scientific research projects, use text mining technology to extract achievement keywords, compare with the forefront research at home and abroad, and evaluate the innovation and influence of the achievements. Set the innovation score of the achievement as C m , and the comprehensive evaluation score Among them is the weight coefficient, set according to the characteristics of the field), establish an achievement sharing platform to promote knowledge exchange and transformation.

[0044] In the present invention, there is also a risk assessment module: Based on historical data and real-time dynamic information, construct a risk prediction model in risk dimensions such as technology, market, and personnel, and use methods such as Monte Carlo simulation to quantify the risk probability and loss degree. Set the technology risk probability as P t , and the loss degree as L t ; the market risk probability is P m , and the loss degree is L m ; the personnel risk probability is P p , and the loss degree is L p ; the total risk probability P total =ω t ×P t +ω m ×P m +ω p ×P p (ω t , ω m , ω p are the corresponding risk weights, set according to the stage of the project and the characteristics of the industry), formulate countermeasures in advance to reduce project risks.

[0045] In the present invention, there is also a user interaction module: Provide personalized operation interfaces for scientific research personnel, project leaders, and management personnel, access through mobile terminals or Web terminals, support voice commands, gesture operations, etc. The time interval from the operation command to receiving the execution feedback, the total operation feedback delay T op =max, the mobile APP adopts a simple interface design, adapts to mainstream mobile phone systems, is easy to operate, and network administrators can control the network status anytime and anywhere.

[0046] In the present invention, when the data acquisition module acquires data, it adopts an incremental acquisition strategy. For data with frequent updates, such as the daily progress report of the project, only the newly added or modified parts are acquired, reducing the data transmission volume and improving the acquisition efficiency. At the same time, the AES encryption technology is used, and the encryption key length is set to, to ensure security, not less than 128 bits, to ensure the security of data transmission.

[0047] In the present invention, when the data analysis module performs clustering analysis, a dynamic weight adjustment mechanism is introduced. According to the real-time changes of scientific research hotspots, the weights of different features in clustering are adjusted, so that the classification result is more in line with the current scientific research trend and the analysis accuracy is improved. Specifically, a scientific research hotspot monitoring sub-module is constructed. Based on natural language processing technology, this sub-module captures the latest research trends in major scientific research information platforms, academic forums, and professional databases in real time, quantifies the changes in hotspots in the form of keywords, research topic heat values, etc., and updates the hotspot data every interval (such as 1 hour). When it is detected that the hotspot change rate Δh exceeds the preset Δh0 (such as 0.2), the weight adjustment process is started, and according to the formula w i = w i0 + s×Δh (where w i is the weight of the adjusted feature, w i0 is the initial weight, and s is the adjustment step size set according to experience, such as 0.05), the weights of the clustering features are adjusted to ensure that the clustering analysis can keep up with the forefront of scientific research.

[0048] In the present invention, in the project progress monitoring module, the Gantt chart supports interactive operations. When the user clicks on the project phase in the chart, the detailed task list, responsible person, and completion status can be viewed, which is convenient for in-depth understanding of project details, and tasks can be directly adjusted on the chart. In terms of technical implementation, the Gantt chart visualization interface is built using HTML5-based Canvas technology. When the user clicks on the graphic area corresponding to a certain project phase, the front-end interface sends a data request to the background through Ajax technology. The background quickly queries the database, obtains the detailed task list, responsible person information, and completion progress details of this phase, and then returns them to the front-end in JSON format. The front-end displays them in the form of a pop-up window; if the user adjusts the tasks in the pop-up window, such as modifying the start and end times of tasks, responsible persons, etc., the front-end sends the adjustment instructions to the background in the same way. After the background verifies the legality of the instructions (based on project rules, personnel permissions, etc.), it updates the database and synchronizes it to the operation interfaces of relevant personnel in real time. The whole process ensures data consistency, and the operation response time does not exceed T s (such as 3 seconds).

[0049] The present invention also discloses a scientific research project management method based on big data processing technology, including the following steps:

[0050] Data collection and preprocessing: The scientific research project data is collected and cleaned by the data collection module of the method, and the processed data is stored in the data storage center described in claim 1.

[0051] Data analysis and decision-making: The algorithms of the data analysis module mine the stored data, and formulate project management strategies based on the analysis results, such as adjusting the project schedule, optimizing personnel allocation, and reasonably allocating funds.

[0052] Project Monitoring and Feedback: Progress monitoring module, personnel management module, funding management module, etc. monitor all aspects of the project in real time, collect feedback information, promptly adjust and handle problems that arise, and ensure the smooth progress of the project.

[0053] Outcome Management and Transformation: The management module manages scientific research outcomes, promotes the sharing and transformation of outcomes, and enhances the overall value of scientific research projects.

[0054] In the present invention, in the data collection and preprocessing stage, the data cleaning process is refined. In addition to removing duplicate and incorrect data, the data format is also standardized to ensure data consistency and usability. During specific operations, according to the differences in data formats of different data sources, a format conversion template library is established. For example, for the date format, it is uniformly converted to the "YYYY - MM - DD" format. For numerical data, the appropriate precision is determined according to its meaning. For example, funding data is retained to two decimal places. At the same time, data verification rules are set to verify the converted data, such as checking whether required fields are complete and whether the data range is reasonable. If problems are found, the data source is traced back in time for correction to ensure the reliability of the data quality entering the storage center.

[0055] In the present invention, in the data analysis and decision - making stage, when potential risks are found in the project, the risk assessment module is promptly activated to quantitatively assess the risks. The management strategy is adjusted in combination with the assessment results to reduce the impact of risks. The activation process is as follows: When the data analysis module monitors potential risk signals such as abnormal project progress delays, trends of over - budget funding, and changes in key personnel, it automatically triggers the activation instruction of the risk assessment module. The risk assessment module, based on historical data and real - time dynamic information, quickly uses methods such as Monte Carlo simulation to quantify the risk probability and loss degree, and presents the assessment results in the form of a visual report (such as a risk heat map, where the darker the color, the higher the risk) to project managers. According to the risk assessment results and in combination with the actual situation of the project, project managers adjust management strategies such as the project schedule plan (such as accelerating the execution speed of tasks on the critical path), optimizing personnel allocation (such as urgently deploying professional and technical personnel to solve technical problems), and reasonably allocating funds (such as suspending non - critical expenses and giving priority to ensuring funds for critical tasks) to reduce the impact of risks on the project.

[0056] In the present invention, in the stage of achievement management and transformation, blockchain technology is used to protect the copyright of achievements and trace the dissemination, and at the same time, combined with the user interaction module, achievements are pushed according to user needs and preferences to improve the efficiency of achievement transformation. In terms of the application of blockchain technology, a unique blockchain identifier is created for each scientific research achievement, and metadata such as the creator, creation time, and affiliated project of the achievement are recorded. When the achievement is disseminated on the sharing platform, all operation information such as access, download, and citation is recorded on the blockchain to achieve traceability; in terms of achievement pushing, based on the information such as the user's browsing history, search keywords, and professional field collected by the user interaction module, a user interest model is constructed, and the collaborative filtering algorithm is used to recommend highly relevant scientific research achievements to the user, and the recommendation accuracy reaches (such as more than 80%), increasing the possibility of achievement transformation.

[0057] The above is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, makes equivalent substitutions or changes, and should be covered by the protection scope of the present invention.

Claims

1. A scientific research project management system based on big data processing technology, characterized in that, including: Data acquisition module: Connect to scientific research data sources, academic databases, internal systems of scientific research institutions, and project application platforms. Set the data acquisition frequency as f and the acquisition cycle adjustment coefficient as k. When the data source update rate accelerates and the system load L increases, and L > L0 (where L0 is the preset load reference value), according to the formula f = f0 + k×(L - L0), the acquisition frequency is increased, where f0 is the initial acquisition frequency; Use a rule-based screening algorithm to remove incorrect data; Data storage center: The Hadoop Distributed File System (HDFS) is adopted to classify and store data, and indexes are established according to the dimensions of projects, time, and disciplines. The number of storage nodes is set to n, and the data redundancy is set to r. According to the formula where m is the number of data backup copies, the redundancy is set. When storing, the Snappy compression algorithm is used, and the compression ratio is set to c. After testing and optimization, it has the functions of data backup and recovery; Data analysis module: Based on the Spark big data analysis framework, the clustering analysis algorithm is used to classify according to research directions and technical difficulty characteristics. A dynamic weight adjustment mechanism is introduced. Let the scientific research hot spot change rate be Δh and the characteristic weight adjustment step be s. When Δh > Δh0, where Δh0 is the hot spot change threshold, according to the formula w i = w i0 + s×Δh, w i is the implementability of the adjusted feature; Project Progress Monitoring Module: By collecting project node data in real time, combining with the project plan schedule, using Gantt charts to visually display the project progress, setting a progress warning threshold, and setting the warning delay time as T w , to ensure timeliness, T w shall not exceed 5 seconds. By optimizing the warning push system architecture and adopting asynchronous message queue combination technology, the time from warning trigger to delivery is reduced; Personnel Management Module: Records the professional skills, project participation experiences, and achievement contribution information of scientific research personnel. Using a recommendation algorithm, it matches the most suitable team members for new projects according to project requirements. Let the personnel skill matching degree be m s , the project experience matching degree be m p , and the comprehensive matching degree M = α × m s + β × m p , where α and β are weight coefficients, set according to the key requirements of the project. For technology R & D projects, α > β, and for application promotion projects, β > α. Members are recommended in descending order; Personnel performance evaluation is based on data, including the number of papers published n p , the number of patent applications n a , the project achievement conversion rate r c and other indicators. The performance score P = γ × n p + δ × n a + ∈ × r c , where γ, δ, and ∈ are the corresponding indicator weights, set according to the project type and institutional orientation; Fund management module: Track the flow of project funds in real time, connect to the financial system, classify and record each expense, and use a cost-benefit analysis model to calculate the input-output ratio of the project; Let the investment be I, the output return be O, and the input-output ratio When the funds used deviate from the budget by 15%, a warning is issued and a detailed financial statement is generated for review; Achievement Management Module: Automatically collect and organize papers, patents, and software copyright achievements generated by scientific research projects, and extract achievements using text mining technology; set the innovation score of the achievement as C and the influence score as I m , and the comprehensive evaluation score where θ, is the weight coefficient.

2. The scientific research project management system based on big data processing technology according to claim 1, characterized in that, It also includes a risk assessment module: Based on historical data and real-time dynamic information, construct a risk prediction model in the dimensions of technology, market, and personnel, and use Monte Carlo simulation to quantify the risk probability and loss degree; Set The probability of technical risk is P t , and the loss degree is L t ; the probability of market risk is P m , and the loss degree is L m ; the probability of personnel risk is P p , and the loss degree is L p ; the total risk probability P total = ω t × P t + ω m × P m + ω p × P p , where ω t , ω m , ω p are the corresponding risk weights, which are set according to the project stage and industry characteristics.

3. The scientific research project management system based on big data processing technology according to claim 1, wherein It also includes a user interaction module: providing a terminal operation interface for scientific researchers, project leaders, and managers, accessible through mobile terminals or the Web side, supporting voice commands and gesture operations, and the time interval from the operation command to receiving the execution feedback, with the total operation feedback delay T op = max. The mobile APP adopts a simple interface design and is adapted to the mobile phone system.

4. The scientific research project management system for big data processing technology according to claim 1, characterized in that When the data analysis module conducts analysis, it introduces a dynamic weight adjustment mechanism. According to the real-time changes of scientific research hotspots, it adjusts the weights of different features in clustering, updates the data hourly, and starts the weight adjustment process if it monitors that the hotspot change rate Δh exceeds the preset Δh0. According to the formula w i = w i0 + s×Δh, where w i is the weight of the adjusted feature, and w i0 is the initial weight.

5. A method for implementing a scientific research project management system based on big data processing technology as described in any one of claims 1-4, characterized in that, including the following steps: Data acquisition and preprocessing: Collect and screen scientific research project data by the method of the data acquisition module, and store the processed data in the data storage center; Data analysis and decision-making: Use the algorithm of the data analysis module to mine the stored data, formulate project management strategies according to the analysis results, adjust the project schedule, optimize personnel allocation, and reasonably allocate funds; Project monitoring and feedback: The progress monitoring module, personnel management module, and fund management module monitor the project situation in real time, collect feedback information, and adjust and handle problems in a timely manner; Outcome management and transformation: The management module manages scientific research outcomes, promotes the sharing and transformation of outcomes, and enhances the overall value of scientific research projects.

Images