College scientific research process automation and intelligent decision system and method

Abstract

The invention discloses a college scientific research process automation and intelligent decision-making system and method, and aims to solve the problem of college scientific research management. According to the method, various data such as scientific research achievements of teachers and student participation projects are collected through a multi-channel data interface, and a Bayesian probability model is used for cleaning and standardization processing. Data are stored by means of the block chain in combination with an RSA asymmetric encryption algorithm, and safety is guaranteed. And constructing a knowledge graph by applying a natural language processing technology and a Skip-gram model, and mining a scientific research relationship. And deeply analyzing data through a machine learning algorithm, and identifying project risks and opportunities. And personalized scientific research service recommendation is provided for teachers and students based on user preferences. A scientific research process is automatically generated and monitored, intelligent suggestions are provided for scientific research management decisions according to analysis results, resource allocation is optimized by integrating multiple factors, and intelligent and scientific development of scientific research management of colleges and universities is promoted.

Description

Technical Field

[0001] The present invention relates to the field of efficient scientific research management, and in particular to a system and method for automating scientific research processes and intelligent decision-making in universities. Background Art

[0002] In the field of scientific research management in universities, traditional research processes and decision-making methods face numerous challenges. Currently, university research data is scattered across independent systems, such as academic administration systems and personnel management systems. Furthermore, there is a lack of effective integration with external academic databases and research collaboration platforms. This makes data collection extremely difficult, and the data that is collected is often subject to significant noise, severely impacting data quality and making it difficult to fully and accurately reflect the actual state of university research.

[0003] Conventional data storage methods lack security, making data vulnerable to leakage and tampering, making it difficult to ensure the integrity and confidentiality of scientific research data. Research processes lack automation, and task allocation and timeline setting rely on manual operations, which is inefficient and prone to human error. Research decisions are often made based on experience, lacking scientific data support. This leads to irrational allocation of research resources, difficulty accurately identifying potential risks and opportunities in research projects, and missed opportunities for valuable research collaboration and development.

[0004] In view of this, it is urgent to develop a scientific research process automation and intelligent decision-making system and method for universities, aiming to solve the pain points in scientific research management mentioned above and promote the improvement of scientific research management level in universities. Summary of the Invention

[0005] In order to solve the above technical problems, the technical solution adopted by the method for automating scientific research processes and intelligent decision-making in universities of the present invention includes the following steps: S1: Data collection: Through multi-channel data interfaces, various types of data related to university scientific research are collected, including: data on teachers' scientific research results, data on students' participation in scientific research projects, data on the use of scientific research equipment, and academic literature data; the noise data identification formula of the Bayesian probability model is used: , where A represents real data and B represents observed data. The collected data is preliminarily cleaned to remove noise data and standardized according to the preset format; S2: Blockchain data storage: The preliminarily processed data is stored using blockchain technology, and the RSA asymmetric encryption algorithm is used to encrypt and store the data; in the distributed ledger of the blockchain, each data block contains the hash value of the previous data block. , the hash value of the current data block ,in The data of the current data block; this ensures that the data cannot be tampered with and is authentic, especially ensuring the safe storage of scientific research data and key project information, and providing protection for scientific research integrity; S3: Semantic Understanding and Knowledge Graph Construction: In the intelligent analysis module, natural language processing technology is used to understand the semantics of academic literature data, using the word vector Skip-gram model. The formula is: ,in As the central word, is the context word, and are the corresponding word vectors, V is the vocabulary size, and the text data is converted into vector representation to build a knowledge graph in the field of university scientific research. The relationship between research topics, researchers, and scientific research institutions is mined through the knowledge graph to provide knowledge support; S4: Intelligent Analysis: Using machine learning algorithms and data mining techniques, combined with relational data in the knowledge graph, we conduct in-depth analysis of the integrated data, explore the correlations between the data, analyze the relationship between teachers' scientific research results and the research equipment used and participating students, and identify potential risks and opportunities in scientific research projects; S5: Personalized scientific research service recommendation: Based on the user's scientific research behavior data and preference data, through personalized recommendation algorithm, using formula ,in represents the similarity between items i and j, N(i) and N(j) represent the sets of users who like items i and j respectively. and Represent user u's ratings of items i and j, respectively, calculate the similarity between users, provide users with customized scientific research service recommendations, recommend scientific research project cooperation opportunities and academic conferences suitable for their research directions to teachers; recommend scientific research project participation channels and related courses that match their interests and abilities to students, thereby improving the utilization efficiency of scientific research resources and users' enthusiasm for participating in scientific research; S6: Process Automation: Based on the analysis results, the execution process of the scientific research project is automatically generated, including task allocation and time node setting. The execution of the process is monitored in real time. When anomalies occur, early warnings are automatically issued and the process is adjusted. S7: Intelligent decision-making step: Based on data analysis and process execution, provide intelligent suggestions for scientific research management decisions, including: reasonable allocation decisions of scientific research resources, establishment and termination decisions of scientific research projects. The decision-making model can be continuously optimized and updated according to actual conditions.

[0006] As a further solution of the present invention, in the data collection step, the multi-channel data interface includes: data interfaces of the internal academic affairs management system and personnel management system of the university, as well as data interfaces with external academic databases and scientific research cooperation platforms, and data collection is achieved through these interfaces.

[0007] As a further solution of the present invention, in the data cleaning algorithm, when noise data is identified based on the Bayesian probability model, it is also necessary to dynamically adjust the prior probabilities P(A) and P(B) according to the characteristics of university scientific research data to improve the accuracy of data cleaning.

[0008] As a further solution of the present invention, in the blockchain data storage step, the key length of the RSA algorithm used is 2048 bits or more, thereby improving the security and strength of data encryption.

[0009] As a further solution of the present invention, in the semantic understanding and knowledge graph construction steps, when using the Skip-gram model to construct word vectors, the window size is adaptively adjusted according to the average length of the academic literature data, and the value range is 3-7.

[0010] As a further solution of the present invention, in the intelligent decision-making step, the decision-making model uses the following formula to calculate the resource allocation weight when making scientific research resource allocation decisions: : , in, Indicates the The influence coefficient of each scientific research result on the scientific research project is determined through historical data statistics and expert evaluation; Indicates the The importance weight of each scientific research result is set according to the standards and development trends of the scientific research field; It represents the correlation between the i-th scientific research project and the j-th scientific research result, which is calculated using semantic analysis and knowledge graph; represents the current resource utilization rate of the i-th scientific research project; m is the total number of scientific research projects; n is the number of types of scientific research results; is the adjustment coefficient with a value range of [0.5, 0.8]. It is dynamically adjusted according to the urgency and strategic importance of the scientific research project. This formula comprehensively considers the influence, importance, relevance of scientific research results and the resource utilization factors of the project to achieve the allocation decision of scientific research resources.

[0011] As a further solution of the present invention, a university scientific research process automation and intelligent decision-making system includes the following modules: Data Collection Module: Equipped with multi-channel data interfaces, it connects to the university's internal academic management system, personnel management system, as well as external academic databases and scientific research cooperation platforms. It is responsible for collecting various scientific research-related data, including faculty research results, student participation in scientific research projects, scientific research equipment usage, academic literature, etc., and uses Bayesian probability models to preliminarily clean and standardize the collected data. Data storage module: Uses blockchain technology to store preliminarily processed data, adopts RSA asymmetric encryption algorithm with a key length of 2048 bits or longer to ensure the security and strength of data encryption. Each data block contains the hash value of the previous data block and its own hash value to build a distributed ledger; Semantic Understanding and Knowledge Graph Construction Module: Based on natural language processing technology, the Skip-gram model is used to perform semantic understanding of academic literature data. The window size is adaptively adjusted between 3 and 7 based on the average length of academic documents. Text data is converted into vector representations to construct a knowledge graph in the field of university research and explore the relationships between research topics, personnel, and institutions. Intelligent Analysis Module: Integrates machine learning algorithms and data mining technology, combines relational data in the knowledge graph, conducts in-depth analysis of the integrated data, explores the relationship between teachers' scientific research results, scientific research equipment, and participating students, and identifies potential risks and opportunities in scientific research projects; Personalized recommendation module: Based on user research behavior and preference data, through personalized recommendation algorithms, the module calculates the similarity between users and recommends research project collaboration opportunities and academic conferences to teachers, and recommends research project participation channels and related courses to students; Process automation module: Based on the results of intelligent analysis, it automatically generates the scientific research project execution process, sets task allocation and time nodes, monitors process execution in real time, and automatically warns and adjusts when anomalies occur; Intelligent decision-making module: Based on data analysis and process execution, it provides intelligent suggestions for scientific research management decisions. For example, scientific research resource allocation decisions use specific formulas to calculate resource allocation weights. The decision-making model of this module can be continuously optimized and updated according to actual conditions.

[0012] Description of beneficial effects: The system and method for automating scientific research processes and intelligent decision-making in universities of the present invention have significant advantages. Data is collected from multiple channels and accurately cleaned to ensure data quality. Blockchain is combined with long-key RSA algorithm storage to ensure data security. Knowledge graphs are constructed through natural language processing to explore scientific research relationships. In-depth analysis such as machine learning is used to identify project risks and opportunities. Personalized scientific research service recommendations are provided based on user preferences to improve resource matching. Scientific research processes are automatically generated and monitored to improve execution efficiency. The intelligent decision-making module gives suggestions based on multiple factors, optimizes resource allocation, and comprehensively promotes the scientific and intelligent management of scientific research in universities. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 This is a flowchart of the steps of a method for automating scientific research processes and making intelligent decisions in universities according to the present invention; Figure 2 This is a module diagram of a university scientific research process automation and intelligent decision-making system according to the present invention. DETAILED DESCRIPTION

[0014] The present invention will be further described in detail below with reference to the embodiments.

[0015] See also Figure 1 The following is a flowchart of a method for automating scientific research processes and making intelligent decisions in universities. The specific steps are as follows: S1: Data collection: Through multi-channel data interfaces, various types of data related to university scientific research are collected, including: data on teachers' scientific research results, data on students' participation in scientific research projects, data on the use of scientific research equipment, and academic literature data; the noise data identification formula of the Bayesian probability model is used: , where A represents real data and B represents observed data. The collected data is preliminarily cleaned to remove noise data and standardized according to the preset format; S2: Blockchain data storage: The preliminarily processed data is stored using blockchain technology, and the RSA asymmetric encryption algorithm is used to encrypt and store the data; in the distributed ledger of the blockchain, each data block contains the hash value of the previous data block. , the hash value of the current data block ,in The data of the current data block; this ensures that the data cannot be tampered with and is authentic, especially ensuring the safe storage of scientific research data and key project information, and providing protection for scientific research integrity; S3: Semantic Understanding and Knowledge Graph Construction: In the intelligent analysis module, natural language processing technology is used to understand the semantics of academic literature data, using the word vector Skip-gram model. The formula is: ,in As the central word, is the context word, and are the corresponding word vectors, V is the vocabulary size, and the text data is converted into vector representation to build a knowledge graph in the field of university scientific research. The relationship between research topics, researchers, and scientific research institutions is mined through the knowledge graph to provide knowledge support; S4: Intelligent Analysis: Using machine learning algorithms and data mining techniques, combined with relational data in the knowledge graph, we conduct in-depth analysis of the integrated data, explore the correlations between the data, analyze the relationship between teachers' scientific research results and the research equipment used and participating students, and identify potential risks and opportunities in scientific research projects; S5: Personalized scientific research service recommendation: Based on the user's scientific research behavior data and preference data, through personalized recommendation algorithm, using formula ,in represents the similarity between items i and j, N(i) and N(j) represent the sets of users who like items i and j respectively. and Represent user u's ratings of items i and j, respectively, calculate the similarity between users, provide users with customized scientific research service recommendations, recommend scientific research project cooperation opportunities and academic conferences suitable for their research directions to teachers; recommend scientific research project participation channels and related courses that match their interests and abilities to students, thereby improving the utilization efficiency of scientific research resources and users' enthusiasm for participating in scientific research; S6: Process Automation: Based on the analysis results, the execution process of the scientific research project is automatically generated, including task allocation and time node setting. The execution of the process is monitored in real time. When anomalies occur, early warnings are automatically issued and the process is adjusted. S7: Intelligent decision-making step: Based on data analysis and process execution, provide intelligent suggestions for scientific research management decisions, including: reasonable allocation decisions of scientific research resources, establishment and termination decisions of scientific research projects. The decision-making model can be continuously optimized and updated according to actual conditions.

[0016] Furthermore, in the data collection step, the multi-channel data interface includes: data interfaces of the university's internal academic affairs management system and personnel management system, as well as data interfaces with external academic databases and scientific research cooperation platforms, through which data collection is achieved.

[0017] Furthermore, in the data cleaning algorithm, when noise data is identified based on the Bayesian probability model, it is also necessary to dynamically adjust the prior probabilities P(A) and P(B) according to the characteristics of university scientific research data to improve the accuracy of data cleaning.

[0018] Furthermore, in the blockchain data storage step, the key length of the RSA algorithm used is 2048 bits or more, which improves the security and strength of data encryption.

[0019] Furthermore, in the semantic understanding and knowledge graph construction steps, when using the Skip-gram model to construct word vectors, the window size is adaptively adjusted according to the average length of the academic literature data, and the value range is 3-7.

[0020] Furthermore, in the intelligent decision-making step, the decision-making model uses the following formula to calculate the resource allocation weight when making scientific research resource allocation decisions: :

[0021] in, Indicates the The influence coefficient of each scientific research result on the scientific research project is determined through historical data statistics and expert evaluation; Indicates the The importance weight of each scientific research result is set according to the standards and development trends of the scientific research field; It represents the correlation between the i-th scientific research project and the j-th scientific research result, which is calculated using semantic analysis and knowledge graph; represents the current resource utilization rate of the i-th scientific research project; m is the total number of scientific research projects; n is the number of types of scientific research results; is the adjustment coefficient with a value range of [0.5, 0.8]. It is dynamically adjusted according to the urgency and strategic importance of the scientific research project. This formula comprehensively considers the influence, importance, relevance of scientific research results and the resource utilization factors of the project to achieve the allocation decision of scientific research resources.

[0022] For further information, see Figure 2 The figure below is a module diagram of a university scientific research process automation and intelligent decision-making system. The system includes the following modules: Data Collection Module: Equipped with multi-channel data interfaces, it connects to the university's internal academic management system, personnel management system, as well as external academic databases and scientific research cooperation platforms. It is responsible for collecting various scientific research-related data, including faculty research results, student participation in scientific research projects, scientific research equipment usage, academic literature, etc., and uses Bayesian probability models to preliminarily clean and standardize the collected data. Data storage module: Uses blockchain technology to store preliminarily processed data, adopts RSA asymmetric encryption algorithm with a key length of 2048 bits or longer to ensure the security and strength of data encryption. Each data block contains the hash value of the previous data block and its own hash value to build a distributed ledger; Semantic Understanding and Knowledge Graph Construction Module: Based on natural language processing technology, the Skip-gram model is used to perform semantic understanding of academic literature data. The window size is adaptively adjusted between 3 and 7 based on the average length of academic documents. Text data is converted into vector representations to construct a knowledge graph in the field of university research and explore the relationships between research topics, personnel, and institutions. Intelligent Analysis Module: Integrates machine learning algorithms and data mining technology, combines relational data in the knowledge graph, conducts in-depth analysis of the integrated data, explores the relationship between teachers' scientific research results, scientific research equipment, and participating students, and identifies potential risks and opportunities in scientific research projects; Personalized recommendation module: Based on user research behavior and preference data, through personalized recommendation algorithms, the module calculates the similarity between users and recommends research project collaboration opportunities and academic conferences to teachers, and recommends research project participation channels and related courses to students; Process automation module: Based on the results of intelligent analysis, it automatically generates the scientific research project execution process, sets task allocation and time nodes, monitors process execution in real time, and automatically warns and adjusts when anomalies occur; Intelligent decision-making module: Based on data analysis and process execution, it provides intelligent suggestions for scientific research management decisions. For example, scientific research resource allocation decisions use specific formulas to calculate resource allocation weights. The decision-making model of this module can be continuously optimized and updated according to actual conditions.

[0023] The foregoing is merely an embodiment of the present invention and is not intended to limit the present invention. Various modifications and variations are possible for those skilled in the art. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention are intended to be included within the scope of the claims of this application.

Claims

1. A method for automating scientific research processes and making intelligent decisions in universities, characterized in that: The following steps are involved: S1: Data collection: Through multi-channel data interfaces, various types of data related to university scientific research are collected, including: data on teachers' scientific research results, data on students' participation in scientific research projects, data on the use of scientific research equipment, and academic literature data; the noise data identification formula of the Bayesian probability model is used: , where A represents real data and B represents observed data. The collected data is preliminarily cleaned to remove noise data and standardized according to the preset format; S2: Blockchain data storage: The preliminarily processed data is stored using blockchain technology, and the RSA asymmetric encryption algorithm is used to encrypt and store the data; in the distributed ledger of the blockchain, each data block contains the hash value of the previous data block. , the hash value of the current data block ,in The data of the current data block; S3: Semantic Understanding and Knowledge Graph Construction: In the intelligent analysis module, natural language processing technology is used to understand the semantics of academic literature data, using the word vector Skip-gram model. The formula is: ,in As the central word, is the context word, and are the corresponding word vectors, V is the vocabulary size, and the text data is converted into vector representation to build a knowledge graph in the field of university scientific research. The relationship between research topics, researchers, and scientific research institutions is mined through the knowledge graph to provide knowledge support; S4: Intelligent Analysis: Using machine learning algorithms and data mining techniques, combined with relational data in the knowledge graph, we conduct in-depth analysis of the integrated data, explore the correlations between the data, analyze the relationship between teachers' scientific research results and the research equipment used and participating students, and identify potential risks and opportunities in scientific research projects; S5: Personalized scientific research service recommendation: Based on the user's scientific research behavior data and preference data, through personalized recommendation algorithm, using formula ,in represents the similarity between items i and j, N(i) and N(j) represent the sets of users who like items i and j respectively. and Represent user u's ratings of items i and j, respectively, calculate the similarity between users, provide users with customized scientific research service recommendations, recommend scientific research project cooperation opportunities and academic conferences suitable for their research directions to teachers, and recommend scientific research project participation channels and related courses that suit their interests and abilities to students; S6: Process Automation: Based on the analysis results, the execution process of the scientific research project is automatically generated, including task allocation and time node setting. The execution of the process is monitored in real time. When anomalies occur, early warnings are automatically issued and the process is adjusted. S7: Intelligent decision-making step: Based on data analysis and process execution, provide intelligent suggestions for scientific research management decisions, including: reasonable allocation decisions of scientific research resources, establishment and termination decisions of scientific research projects. The decision-making model can be continuously optimized and updated according to actual conditions.

2. A method for automating scientific research processes and making intelligent decisions in universities according to claim 1, characterized in that: In the data collection step, the multi-channel data interface includes: data interfaces of the internal academic affairs management system and personnel management system of the university, as well as data interfaces with external academic databases and scientific research cooperation platforms, through which data collection is achieved.

3. The method for automating scientific research processes and making intelligent decisions in universities according to claim 1, characterized in that: In the data cleaning algorithm, when identifying noise data based on the Bayesian probability model, it is also necessary to dynamically adjust the prior probabilities P(A) and P(B) according to the characteristics of university scientific research data to improve the accuracy of data cleaning.

4. The method for automating scientific research processes and making intelligent decisions in universities according to claim 1, characterized in that: In the blockchain data storage step, the key length of the RSA algorithm used is 2048 bits or more, which improves the security and strength of data encryption.

5. The method for automating scientific research processes and making intelligent decisions in universities according to claim 1, characterized in that: In the semantic understanding and knowledge graph construction steps, when using the Skip-gram model to construct word vectors, the window size is adaptively adjusted according to the average length of the academic literature data, and the value range is 3-7.

6. The method for automating scientific research processes and making intelligent decisions in universities according to claim 1, characterized in that: In the intelligent decision-making step, the decision-making model uses the following formula to calculate the resource allocation weight when making scientific research resource allocation decisions: : ， in, Indicates the The influence coefficient of each scientific research result on the scientific research project is determined through historical data statistics and expert evaluation; Indicates the The importance weight of each scientific research result is set according to the standards and development trends of the scientific research field; It represents the correlation between the i-th scientific research project and the j-th scientific research result, which is calculated using semantic analysis and knowledge graph; represents the current resource utilization rate of the i-th scientific research project; m is the total number of scientific research projects; n is the number of types of scientific research results; is the adjustment coefficient with a value range of [0.5, 0.8]. It is dynamically adjusted according to the urgency and strategic importance of the scientific research project. This formula comprehensively considers the influence, importance, relevance of scientific research results and the resource utilization factors of the project to achieve the allocation decision of scientific research resources.

7. A university scientific research process automation and intelligent decision-making system, characterized by: The system comprises: Data Collection Module: Equipped with multi-channel data interfaces, it connects to the university's internal academic management system, personnel management system, as well as external academic databases and scientific research cooperation platforms. It is responsible for collecting various scientific research-related data, including faculty research results, student participation in scientific research projects, scientific research equipment usage, academic literature, etc., and uses Bayesian probability models to preliminarily clean and standardize the collected data. Data storage module: Uses blockchain technology to store preliminarily processed data, adopts RSA asymmetric encryption algorithm with a key length of 2048 bits or longer to ensure the security and strength of data encryption. Each data block contains the hash value of the previous data block and its own hash value to build a distributed ledger; Semantic Understanding and Knowledge Graph Construction Module: Based on natural language processing technology, the Skip-gram model is used to perform semantic understanding of academic literature data. The window size is adaptively adjusted between 3 and 7 based on the average length of academic documents. Text data is converted into vector representations to construct a knowledge graph in the field of university research and explore the relationships between research topics, personnel, and institutions. Intelligent Analysis Module: Integrates machine learning algorithms and data mining technology, combines relational data in the knowledge graph, conducts in-depth analysis of the integrated data, explores the relationship between teachers' scientific research results, scientific research equipment, and participating students, and identifies potential risks and opportunities in scientific research projects; Personalized recommendation module: Based on user research behavior and preference data, through personalized recommendation algorithms, the module calculates the similarity between users and recommends research project collaboration opportunities and academic conferences to teachers, and recommends research project participation channels and related courses to students; Process automation module: Based on the results of intelligent analysis, it automatically generates the scientific research project execution process, sets task allocation and time nodes, monitors process execution in real time, and automatically warns and adjusts when anomalies occur; Intelligent decision-making module: Based on data analysis and process execution, it provides intelligent suggestions for scientific research management decisions. For example, scientific research resource allocation decisions use specific formulas to calculate resource allocation weights. The decision-making model of this module can be continuously optimized and updated according to actual conditions.

Images