A high school individual chemistry course service system and method

Abstract

The application belongs to the technical field of discipline service, and discloses a college individual discipline service system, which comprises a user management module, a data management module, a discipline data calculation module, a scholar portrait module, a resource recommendation module and a system parameter setting module. In order to improve the individuality and accuracy of college discipline service, the college data center, the application system and the data of the digital resource access control system are extracted, and the scholar portrait is constructed according to four dimensions of basic information, academic achievements, scientific research behavior and hot behavior. The college discipline service platform is constructed under the support of the scholar portrait, and the discipline service such as resource individualization push, discipline development data analysis, auxiliary discipline team establishment and talent introduction is realized. The discipline service mode is reconstructed, the service ability is improved, and the work efficiency of the discipline librarian is improved.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of subject service, and particularly relates to a college personalized subject service system and method. BACKGROUND

[0002] At present, Lib Guides has many problems in the construction of subject service platform. First, the functional modules of Lib Guides are relatively fixed, and the expansibility is poor, which is difficult to meet the increasingly diversified user needs of university libraries. Due to the limitation of architecture design, it is difficult for Lib Guides to add new functions or carry out customized development. In addition, the resource selection and organization of Lib Guides completely depends on the personal ability of subject librarians, resulting in uneven resource quality, and lacking effective resource indexing mechanism, which is not conducive to the secondary development and organization of resources by librarians. At the same time, the interface style of Lib Guides is fixed, and lacks personalized modification ability, which cannot fully show the characteristics of each library.

[0003] The latitude subject service platform also has limitations in design. In terms of service design, the platform overemphasizes functionality, and there is a lack of organic connection between different types of service content, resulting in insufficient overall service feeling, affecting user experience and participation. Due to the fixed functional modules, it is not conducive to the secondary development of users, and the flexibility of the platform is low. In addition, the resource organization system of the platform is not perfect, lacking unified classification standards and standardized metadata management, affecting the effective integration and utilization of resources. In terms of system stability, the platform also has problems such as system crash, data loss, slow response speed, etc., which further reduces the user's satisfaction.

[0004] The subject service of university library is a new service mode that responds to the changes of the times, and its core goal is to provide accurate personalized service for different users. However, there is still a big gap in the construction degree, service mode, service content, service platform and subject librarian of universities at home and abroad. At present, even the "double first-class" universities, the understanding of subject service is still not in place, the service mode is overlapped, the service content is insufficient in depth, and the attention to the construction of subject librarian team is not enough. Especially in the 37 undergraduate colleges in Hunan Province, only 15 university libraries have formally carried out "subject service", and the situation of higher vocational colleges is more severe, and the lack of a suitable subject service platform has become one of the main obstacles to the promotion of subject service work.

[0005] At present, the subject service platform of university library mainly depends on three types of tools: LibGuides, domestic subject service platform (such as latitude subject service platform), and third-party system (such as blog, WeChat public number, and super-star mobile library learning channel). Although LibGuides has certain customization ability, it is not friendly to Chinese, and the response speed is slow because it is deployed overseas. Although the latitude subject service platform is suitable for domestic universities, its function module is fixed, and it lacks pertinence and flexibility. Although the third-party system can reflect certain subject characteristics, it lacks resource integration capability, has high update and maintenance cost, and lacks automatic information integration function. Overall, the application of subject service platform is not popular enough, and the construction mode is various, but a mature and widely applicable application system has not been formed, and the independent development of subject service platform has become an important solution for universities to improve the quality of subject service.

[0006] The subject service under the environment of smart campus needs to fully utilize big data technology, deeply embed user demand into the subject service system, and construct an embedded subject service system with institutional portrait as the core. At present, the subject service platform still has many problems, such as insufficient personalized service, insufficient comprehensive understanding of the current situation of the subject, lack of unified platform support, passive service mode, and research mainly staying in the theoretical stage. These problems lead to the fact that the quality of subject service depends on the personal ability of subject librarians, campus big data cannot effectively support subject service, scholars often cannot get the information they need, thereby reducing the satisfaction of subject service and hindering the development of subject service.

[0007] In order to optimize the subject service platform and improve user experience and satisfaction, the subject service platform in the future should combine user portrait technology, deeply mine scholar data, and form a multi-dimensional scholar portrait to help subject librarians comprehensively understand the current situation of the subject and improve the ability of personalized subject service. In addition, an integrated subject service information platform should be constructed to strengthen the overall management of subject service, and information push should be the main method to improve the convenience and accuracy of subject service. Through these measures, the deficiencies of the current subject service platform can be effectively solved, and the subject service of university library can be developed in the direction of more intelligent, personalized and accurate. SUMMARY

[0008] In view of the problems existing in the prior art, the present application provides a university personalized subject service system.

[0009] The present application is realized in the following way: a university personalized subject service system comprises:

[0010] a user management module, a data management module, a subject data calculation module, a scholar portrait module, a resource recommendation module, and a system parameter setting module.

[0011] The user management module is connected with the data management module, and is used for connecting a unified identity authentication system of a school. After a user inputs an account and a password, the user management module sends an authentication request to the authentication system. The system verifies the validity of the user, and generates a research interest table, a hot behavior table and the like according to returned identity information after verification.

[0012] The data management module is connected with the user management module, the subject data calculation module, the scholar portrait module, the resource recommendation module and the system parameter setting module, and is used for data collection, data preprocessing, data updating and data export.

[0013] The subject data calculation module is connected with the data management module, and is used for calculating a research interest table, a hot behavior table and the like after periodically updating a scholar basic information table, a subject basic information table, an academic achievement table and a scientific research behavior log.

[0014] The scholar portrait module is connected with the data management module, and is used for generating a scholar portrait, combining and querying the portrait and statistically analyzing the portrait.

[0015] The resource recommendation module is connected with the data management module, and is used for resource acquisition, resource import, resource matching and result sending.

[0016] The system parameter setting module is connected with the data management module, and is used for setting system parameters.

[0017] Further, the data management module comprises:

[0018] The data management module comprises four sub-modules of data collection, data preprocessing, data updating and data export. The data collection and preprocessing are the basis for constructing a scholar portrait. The data collection refers to a process of acquiring required data from various data sources. The data preprocessing refers to cleaning, deduplication and conversion processing of original data, so as to facilitate subsequent analysis and application.

[0019] (1) Data collection: ODI or DATA-X tool software is used in the absence of a data center, or an API interface based on a school data center platform is used. When the API interface service is called, an application identifier and a use key are transmitted to sign the parameters to complete identity authentication.

[0020] (2) Data preprocessing; In order to ensure the integrity, uniformity and accuracy of the data set, data preprocessing is needed before using the data, which is transformed into data that can be used to build the portrait of university teachers' scientific research; The process of data preprocessing is different for different data sets, generally including data cleaning, data conversion and data integration; Data cleaning refers to the filtering, filling and deleting of repeated, redundant, missing and abnormal data in the data to ensure the consistency and integrity of the data, and the specific operations include filling the missing values of key fields, removing duplicates and deleting useless fields; Data conversion refers to converting the original data into a unified format suitable for specific analysis tasks;

[0021] (3) Data update; By integrating the Python interpreter to execute the Python script during data collection and data preprocessing, the administrator is given the right to update the data source;

[0022] (4) Data export; According to the needs, the data in each data table can be exported for use by other applications.

[0023] Further, the subject data calculation module comprises:

[0024] The subject service platform has 9 data tables, among which the scholar basic information table, the subject basic information table, the academic achievement table and the scientific research behavior log are original data, and the research interest table and the hot behavior table need to be recalculated after regular update, which is the task of this module.

[0025] Further, the scholar portrait module comprises:

[0026] It includes three sub-modules: scholar portrait generation, portrait combination query and portrait statistical analysis. The scholar portrait generation is to generate the scholar portrait label table and the subject portrait label table according to the original data and the research interest table and the hot behavior table generated by the subject data calculation module; The portrait combination query is to query the scholar portrait label table, such as querying "non-library" + "subject service" to get the scholars who have "subject service" related research results and do not belong to the library; The portrait statistical analysis is similar, and finally a statistical report can be formed.

[0027] Further, the resource recommendation module comprises:

[0028] It is divided into four sub-modules: resource acquisition, resource import, resource matching and result sending.

[0029] Resource acquisition; academic resources are diverse, with a focus on new library books, project application notices, and academic conferences. The data sources are the library book system, the school science and technology department / sociology department website, and the China Academic Conference Network (conf.cnki.net). The data integration methods are API connection and web crawling. Compared with the data acquisition of new books and project applications, the data obtained by web crawling needs to be processed before it can be applied.

[0030] Resource import; extract relevant fields from resource information and fill in the latest academic resource table. In addition to automatic import, manual import is supported.

[0031] Resource matching; matching, results are stored in the recommended results table, and are sorted by similarity.

[0032] Result sending; call the message center to send the corresponding recommended content to the scholar. The previous recommended results are saved in the recommended results table for the scholar to view and evaluate the accuracy of the recommendation, which can be used as a reference for improving the recommendation algorithm. In the next step, the discipline service platform and subject librarians can communicate with scholars through WeChat, and the recommended results are sent to WeChat at regular intervals.

[0033] Further, the system parameter setting module:

[0034] The scholar can set to turn on or turn off resource recommendation. Since the recommendation function needs to collect the scholar's personal behavior information, the person's consent is required, and it is turned off by default.

[0035] The number of recommended resources each time is 50 by default.

[0036] The recommended information sending method is supported by the message center, WeChat, and SMS, as well as their combination. By default, only email is sent.

[0037] The scholar's personal information maintenance, telephone and email, degree and title changes, and school personnel can be automatically synchronized.

[0038] The scholar can define research interests and hot behaviors, and academic resources are recommended accordingly. The time decay coefficient of the subject research interest, the journal source directory, and the conference website need to be set by the subject librarian.

[0039] After the discipline service platform completes the connection of each data source, it generates tags, calculates weights and time decay regularly every day to ensure that the platform maintains the latest discipline data. Recommended resources are sent regularly every week, and the automatic operation mode reduces the dependence on subject librarians. Subject leaders and management departments can enter the platform to understand the development of the discipline. In addition, the platform records the whole process of discipline service, which is conducive to the work handover of subject librarians. The discipline service platform supported by the construction of scholar portraits can improve service capabilities and reconstruct service modes.

[0040] Another object of the present application is to provide a high school individual chemistry service method comprising:

[0041] Step 1, through the user management module to the unified identity authentication system of the school, the user inputs the account password and sends an authentication request to the authentication system, the system verifies the validity, and the system generates a research interest table, a hot behavior table and the like according to the returned identity information after the verification is passed;

[0042] Step 2, through the data management module data acquisition, data preprocessing, data updating and data export;

[0043] Step 3, through the subject data calculation module, the basic information table of scholars, the basic information table of subjects, the academic achievement table and the scientific research behavior log and the like raw data are regularly updated and need to be recalculated to generate the research interest table and the hot behavior table, which is the task of the module;

[0044] Step 4, through the scholar portrait module, the scholar portrait generation, the portrait combination query and the portrait statistical analysis;

[0045] Step 5, through the resource recommendation module, the resource acquisition, the resource import, the resource matching and the result sending;

[0046] Step 6, through the system parameter setting module, the system parameters are set.

[0047] Another object of the present application is to provide a computer device, comprising a memory and a processor, the memory stores a computer program, the computer program is executed by the processor, so that the processor executes the steps of the high school individual chemistry service method.

[0048] Another object of the present application is to provide a computer readable storage medium, storing a computer program, the computer program is executed by the processor, so that the processor executes the steps of the high school individual chemistry service method.

[0049] Another object of the present application is to provide an information data processing terminal, which is used to realize the high school individual chemistry service system.

[0050] In combination with the above technical solutions and the technical problems solved, the technical solution to be protected by the present application has the following advantages and positive effects:

[0051] The present application extracts the data of the university data center, the application system and the digital resource access control system, and constructs the scholar portrait from four dimensions of basic information, academic achievement, scientific research behavior and hot behavior. The portrait comprehensively reflects the research interest, academic achievement and scientific research dynamics of the scholar, and provides accurate support for subsequent individualized service.

[0052] Based on the accurate characterization of the scholar portrait, the system classifies and identifies the subject resources, and intelligently pushes the related resources through the interest matching algorithm, realizing personalized resource recommendation. This method breaks the traditional static resource search mode, enabling scholars to efficiently obtain academic resources that meet their research needs and improving the academic service experience.

[0053] The subject service platform constructed by the present application not only supports personalized resource pushing, but also has functions such as subject development data analysis, subject team auxiliary establishment, and talent introduction decision support. The platform uses data intelligent analysis technology to accurately identify the development trend of the subject, helping colleges and universities optimize subject construction and talent layout.

[0054] The technical solution solves the key technical problems of automatic collection of subject resources, classification service based on scholar portrait, and active academic service, and breaks through the bottleneck of personalized subject service. Through accurate matching of scholar portrait and resources, the platform improves the pertinence and intelligent level of subject service, and improves the work efficiency of subject librarians, providing a new solution for the development of academic ecology in colleges and universities. BRIEF DESCRIPTION OF DRAWINGS

[0055] Figure 1 is the structure block diagram of the personalized subject service system for colleges and universities provided by the embodiment of the present application.

[0056] Figure 2 is the method flow chart of the personalized subject service for colleges and universities provided by the embodiment of the present application.

[0057] Figure 3 is the scholar portrait construction flow chart provided by the embodiment of the present application.

[0058] Figure 4 is the overall architecture diagram of the subject service platform provided by the embodiment of the present application.

[0059] Figure 5 is the investigation result diagram of book recommendation and user demand matching degree provided by the embodiment of the present application.

[0060] Figure 6 is the user satisfaction survey result diagram of different aspects of the system provided by the embodiment of the present application.

[0061] Figure 7 is the user satisfaction survey result diagram of four different aspects of the system content provided by the embodiment of the present application.

[0062] Figure 1 Medium: 1, user management module; 2, data management module; 3, subject data calculation module; 4, scholar portrait module; 5, resource recommendation module; 6, system parameter setting module. DETAILED DESCRIPTION

[0063] In order to make the objects, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and should not be used to limit the present application.

[0064] As shown in Figure 1 The university personalized discipline service system provided by the embodiment of the present application comprises:

[0065] a user management module 1, a data management module 2, a discipline data calculation module 3, a scholar portrait module 4, a resource recommendation module 5, and a system parameter setting module 6.

[0066] The user management module 1 is connected with the data management module 2, and is used for connecting with a unified identity authentication system of a school. After a user inputs an account and a password, the user management module 1 sends an authentication request to the unified identity authentication system. The system verifies the validity of the user, and after the verification is passed, the system generates a research interest table, a hot behavior table and the like according to the returned identity information.

[0067] The data management module 2 is connected with the user management module 1, the discipline data calculation module 3, the scholar portrait module 4, the resource recommendation module 5, and the system parameter setting module 6, and is used for data collection, data preprocessing, data updating and data export.

[0068] The discipline data calculation module 3 is connected with the data management module 2, and is used for calculating a research interest table, a hot behavior table and the like according to a scholar basic information table, a discipline basic information table, an academic achievement table and a scientific research behavior log.

[0069] The scholar portrait module 4 is connected with the data management module 2, and is used for generating a scholar portrait, combining and querying the portrait, and statistically analyzing the portrait.

[0070] The resource recommendation module 5 is connected with the data management module 2, and is used for resource acquisition, resource import, resource matching and result sending.

[0071] The system parameter setting module 6 is connected with the data management module 2, and is used for setting system parameters.

[0072] The detailed signal and data processing process of the university personalized discipline service system provided by the embodiment of the present application is as follows:

[0073] 1. User identity authentication and management (user management module 1)

[0074] User login request: a user inputs an account and a password through a front-end interface, and the system sends the information to a unified identity authentication system of a school.

[0075] Identity Verification: The authentication system verifies the account password, and after verification, returns the user's identity information (such as scholar, student, administrator, etc.).

[0076] User Permission Assignment: The system assigns permissions to users based on identity information, such as subject resource access permissions, data query permissions, etc.

[0077] User Access Log: User Management Module 1 records user login logs, including login time, access IP, device information, etc., for subsequent analysis of user behavior.

[0078] 2. Data Collection and Management (Data Management Module 2)

[0079] Data Collection:

[0080] Through interfaces, connect with internal and external databases (such as academic paper databases, scientific research project databases, library databases, etc.), regularly or in real-time collect user scientific research data, academic achievements, subject resources, etc.

[0081] The collected raw data includes:

[0082] Scholar basic information (name, title, research direction, etc.)

[0083] Subject basic information (subject code, research field, related institutions, etc.)

[0084] Academic achievements (papers, patents, projects, awards, etc.)

[0085] Scientific research behavior log (paper citation, search record, download record, etc.)

[0086] Data Preprocessing:

[0087] Structured and unstructured data conversion, such as parsing paper abstracts, keywords, author information, etc.

[0088] Data deduplication, clean up redundant data, ensure data consistency and integrity.

[0089] Data standardization, such as unified subject classification standards, journal formats, research fields, etc.

[0090] Data Update:

[0091] Set data synchronization strategy, regularly check data update situation, perform incremental update or full update.

[0092] Monitor database changes, such as newly published papers, newly initiated scientific research projects, etc., and automatically synchronize to the system.

[0093] Data Export:

[0094] Multiple data export formats (CSV, Excel, JSON, etc.) are provided for subject librarians and researchers to use.

[0095] 3. Subject Data Computing and Analysis (Subject Data Computing Module 3)

[0096] Data Computing Tasks:

[0097] Classify, label, and statistically analyze the collected raw data.

[0098] Calculate research hotspots, such as the number of recent papers published in a field, citation rates, and collaboration networks.

[0099] Calculate scholars' research interests, combining their research achievements and citation situations to dynamically generate a research interest table.

[0100] Generate a hot behavior table, including scholars' paper submission, citation, and academic collaboration relationships.

[0101] Calculate the degree of subject intersection and analyze the relevance between different subjects to support interdisciplinary research recommendations.

[0102] Calculation Method:

[0103] Use machine learning algorithms (such as K-means clustering and TF-IDF text analysis) for academic interest mining.

[0104] Combine deep learning models to recognize patterns in research behavior logs and predict future research directions for scholars.

[0105] Use natural language processing (NLP) techniques to analyze the semantics of paper and patent text content, extract keywords, and classify them.

[0106] 4. Scholar Portrait Construction (Scholar Portrait Module 4)

[0107] Portrait Generation:

[0108] Establish a scholar portrait based on their published papers, research projects, and academic conference participation records.

[0109] The portrait includes research areas, research interests, academic influence (H-index, citation count), and research collaboration networks.

[0110] Portrait Query:

[0111] Users can search for scholar portraits by keywords (such as "artificial intelligence + image processing") to query a list of scholars who meet the criteria.

[0112] Can be filtered by region, academic institution, research field, etc.

[0113] Portrait Statistical Analysis:

[0114] Statistical trends in a certain discipline, such as the popular research topics in the field in the past five years, and the ranking of active scholars.

[0115] Monitoring the academic cooperation network within the university, analyzing the frequency and strength of cooperation between scholars.

[0116] 5. Resource recommendation (Resource Recommendation Module 5)

[0117] Resource acquisition:

[0118] Through the database API, open academic resource platform (such as CNKI, Google Scholar), etc., to obtain the latest scientific research papers, journals, patents, etc.

[0119] Resource matching:

[0120] Combined with the results of scholar portrait and discipline data calculation, intelligent recommendation of relevant scientific research resources.

[0121] Using collaborative filtering algorithm, based on the reading / reference / contribution behavior of similar scholars for personalized recommendation.

[0122] Combined with semantic analysis, recommend academic resources that are highly matched with the user's research field.

[0123] Results sending:

[0124] Through email, in-site notification, WeChat / APP push, etc. to send resource recommendation information to users.

[0125] Users can set the recommendation frequency (daily, weekly, monthly), and the system automatically pushes the matching scientific research resources.

[0126] 6. System parameter setting (System Parameter Setting Module 6)

[0127] Parameter configuration:

[0128] Users can adjust system parameters, such as data update frequency, recommendation algorithm weight, user access rights, etc.

[0129] Permission management:

[0130] Different roles (scholars, subject librarians, administrators) have different access and management permissions.

[0131] Allow administrators to add, modify, delete users, and assign role permissions.

[0132] System monitoring:

[0133] Monitor system running status, such as server load, data processing efficiency, etc., to ensure system stable operation.

[0134] Provide log recording, analyze system exception conditions and perform repair.

[0135] System workflow summary

[0136] 1. The user logs in through the identity authentication system, and the system verifies the identity and grants corresponding permissions.

[0137] 2. The data management module collects and preprocesses discipline resources and user behavior data, and updates the database regularly.

[0138] 3. The discipline data calculation module analyzes academic data to generate research interest tables, hot behavior tables, etc.

[0139] 4. The scholar portrait module constructs a user academic portrait to support query and statistical analysis.

[0140] 5. The resource recommendation module intelligently matches academic resources to users based on scholar portraits and research interests.

[0141] 6. The system parameter setting module supports system management, permission allocation, log monitoring, etc., to ensure stable system operation.

[0142] The university personalized discipline service system provided by the present application improves the personalization and intelligence level of discipline service through intelligent data processing, scholar portrait construction, and precise resource recommendation, optimizes academic resource utilization efficiency, and provides more efficient and precise service for university research and teaching.

[0143] The data management module provided by the embodiment of the present application comprises four sub-modules of data collection, data preprocessing, data updating and data export, wherein data collection and preprocessing are the basis for constructing a scholar portrait, data collection refers to the process of obtaining required data from various data sources, and data preprocessing is cleaning, deduplication and conversion processing of original data to facilitate subsequent analysis and application.

[0144] The data management module comprises four sub-modules of data collection, data preprocessing, data updating and data export, wherein data collection and preprocessing are the basis for constructing a scholar portrait, data collection refers to the process of obtaining required data from various data sources, and data preprocessing is cleaning, deduplication and conversion processing of original data to facilitate subsequent analysis and application.

[0145] (1) Data collection; in the absence of a data center, ODI or DATA-X tool software is used to realize, otherwise it is realized through an API interface based on a school data center platform, and when calling the API interface service, the application identifier and the use key are transmitted to the parameter signature to complete identity authentication;

[0146] (2) Data preprocessing; in order to ensure the integrity, uniformity and accuracy of the data set, data preprocessing work is needed before using the data, so that it is converted into data that can be used to build the portrait of university teachers' scientific research; the process of data preprocessing is not the same for different data sets, generally including three steps of data cleaning, data conversion and data integration; data cleaning refers to filtering, filling and deleting operations on repeated, redundant, missing and abnormal data in the data to ensure the consistency and integrity of the data, and the specific operations include filling the missing values of key fields, removing duplicates and deleting useless fields; data conversion refers to converting the original data into a unified format suitable for specific analysis tasks;

[0147] (3) Data update; by integrating the Python interpreter to execute the Python script during data collection and data preprocessing, the administrator is given the permission to update the data source;

[0148] (4) Data export; the data in each data table can be exported as needed for use by other applications.

[0149] The subject data calculation module provided by the embodiment of the application comprises:

[0150] The subject service platform has nine data tables, among which the scholar basic information table, the subject basic information table, the academic achievement table and the scientific research behavior log are original data, and after being updated regularly, the research interest table and the hot behavior table need to be recalculated and generated, which is the task of the module.

[0151] The scholar portrait module provided by the embodiment of the application comprises:

[0152] The scholar portrait module comprises three sub-modules of scholar portrait generation, portrait combination query and portrait statistical analysis; the scholar portrait generation is to generate the scholar portrait label table and the subject portrait label table according to the original data and the research interest table and the hot behavior table generated by the subject data calculation module; the portrait combination query is to query the scholar portrait label table, for example, querying “non-library”+“subject service” can obtain the scholars who have the research results related to “subject service” and do not belong to the library; the portrait statistical analysis is similar, and finally a statistical report can be formed.

[0153] The resource recommendation module provided by the embodiment of the application comprises:

[0154] The resource recommendation module comprises four sub-modules of resource acquisition, resource import, resource matching and result sending.

[0155] Resource acquisition; academic resources are diverse, and the library new book, project application notice and academic conference are recommended, and the data sources are library book system, school science and technology department / social science department website and China academic conference network (conf.cnki.net), and the data integration modes are API connection and network crawler; compared with the data acquisition of new books and project applications, the data acquired by the network crawler method needs to be processed after complex processing before application;

[0156] Resource import; the relevant fields in the resource information are filled into the latest academic resource table, and manual import is supported in addition to automatic import;

[0157] Resource matching; matching, the result is stored in the recommended result table, and is sorted according to similarity;

[0158] Result sending; the corresponding recommended content is sent to the scholar by calling the message center, and the previous recommended results are saved in the recommended result table for the scholar to view and evaluate the recommendation accuracy, and the recommended algorithm is improved, and the next step can be connected with the enterprise WeChat to realize that the subject service platform and the subject librarian communicate with the scholar through WeChat, and the recommended result is sent to WeChat at regular intervals.

[0159] The system parameter setting module provided by the embodiment of the application comprises:

[0160] The scholar can set to open or close resource recommendation, and the recommendation function needs to be opened to collect the personal behavior information of the scholar and needs the consent of the person, and the default is closed;

[0161] The number of recommended resources each time is 50 by default;

[0162] The recommended information sending mode is supported by the message center, WeChat, short message and their combination mode, and the default is to send only the email;

[0163] The scholar personal information maintenance, telephone email, degree and title change, and school personnel can be automatically synchronized;

[0164] The scholar can define research interest and hot behavior, and academic resources are recommended accordingly; and the time decay coefficient of the research interest of the subject, the journal source directory and the conference website need to be set by the subject librarian;

[0165] After the subject service platform completes the connection of each data source, the label generation, weight and time decay calculation are performed at regular intervals every day, so that the platform can keep the latest subject data, and the recommended resources are sent at regular intervals every week, and the automatic operation mode reduces the dependence on the subject librarian, the subject responsible person and the management department enter the platform to understand the subject development situation, in addition, the platform records the whole subject service process, which is beneficial to the work handover of the subject librarian; the subject service platform under the support of the scholar portrait can improve the service ability and reconstruct the service mode.

[0166] AsFigure 2 As shown, the high school personalized subject service method provided by the embodiment of the application comprises:

[0167] S101, a unified identity authentication system of a school is connected through a user management module, a user inputs an account password and sends an authentication request to the authentication system, the system verifies the validity, and after verification, the system generates a research interest table, a hot behavior table and the like according to returned identity information;

[0168] S102, data acquisition, data preprocessing, data updating and data export are performed through a data management module;

[0169] S103, a subject data calculation module is used to calculate a research interest table and a hot behavior table after a scholar basic information table, a subject basic information table, an academic achievement table and a scientific research behavior log and the like are periodically updated;

[0170] S104, a scholar portrait module is used for scholar portrait generation, portrait combination query and portrait statistical analysis;

[0171] S105, a resource recommendation module is used for resource acquisition, resource import, resource matching and result sending;

[0172] S106, a system parameter setting module is used for setting system parameters.

[0173] Another object of the application is to provide a computer device comprising a memory and a processor, wherein the memory stores a computer program, and the computer program is executed by the processor to make the processor execute the steps of the high school personalized subject service method.

[0174] Another object of the application is to provide a computer readable storage medium storing a computer program, and the computer program is executed by a processor to make the processor execute the steps of the high school personalized subject service method.

[0175] Another object of the application is to provide an information data processing terminal for realizing the high school personalized subject service system.

[0176] The application is specifically implemented as follows:

[0177] As shown in the accompanying drawings, Figure 5To verify the effectiveness of the user portrait-based subject service, 194 teachers and students majoring in education technology were selected as the object to recommend the newly listed professional books as subject resources within half a year. Using 110103 borrowing data, 12 education technology reference textbooks, 62 education technology courses, and 162 course reference textbooks as data sources to construct user portraits, 18088 new books and 9 lectures within half a year were selected as recommended resources. The books and lectures suitable for them were recommended to them, and the top 11 or 12 books were sent to the students according to the weight, a total of 8730 data were sent, 94 new books and 5 lectures were recommended to the students, the recommendation was sent once a week, a total of 4 times, the push time points were September 11, September 18, September 25, and October 9, everyone received the push message through WeChat, and the average person received 45 books.

[0178] As Figure 6 shown, after two weeks of pushing, a questionnaire was designed and distributed to users to investigate their satisfaction. A total of 194 questionnaires were distributed, and 168 valid questionnaires were recovered, with a recovery rate of 87%. Among the 168 questionnaires, 46 people indicated that they had not browsed the recommended list, the reasons were that they did not follow the school WeChat enterprise number, did not see the message push or saw the message but had no time to see, had no interest to see, among the 46 people, 33 people expressed interest in further understanding, in the later stage, readers can be informed of this new function through other means, and everyone is encouraged to check the recommended content and give feedback.

[0179] As Figure 7 shown, the remaining 122 people who had browsed described their feelings after seeing the recommended results, almost everyone indicated that the recommended resources were related to their course learning, professional learning, and personal interests to varying degrees, 22.45% of people believed that more than 40% of the books were suitable for them, and 64.29% believed that 50-80% of the books in the recommended list were suitable for them, so more than 86% of people indicated that at least half of the books were completely suitable for them, the accuracy rate was 66.6%, and the overall satisfaction was as high as 82%. The questionnaire also investigated the factors that affected their satisfaction, 95% of people indicated that the attractiveness of the content was an important factor affecting the user experience, and the recommendation interface layout, operation complexity, and push frequency were also considered by users. The most satisfying thing for students in this verification was the update speed of the content, while the coverage and novelty of the content were relatively lacking, the reason may be that the user borrowing data was relatively scarce, and to a large extent, it depended on the static user portrait generated by the major and course, resulting in that the recommended content was mostly learning books, which was relatively monotonous, but everyone had the willingness to continue to accept resource recommendation.

[0180] 1 Scholar Portrait Construction

[0181] The construction of scholar user portrait can mine the concise scholar profile from the complex scholar information, which is the basis of personalized chemical discipline service. The concept of "scientific research scholar portrait" appeared in 2016. According to the view of Wen Qingzhan, the scholar portrait is the various attributes and behaviors of the scholar in the academic field, including the basic information of the scholar portrait, the research interest label of the scholar, the evaluation index of the academic influence of the scholar, the information of the works and co-authored works of the scholar. According to the view of Yuan Sha, the scholar portrait is the extraction of the attribute information of the scholar in each dimension for information mining and analysis application. According to the view of Fan Xiaoyu, the portrait of scientific research personnel is a tagged and formalized user model abstracted from the social attributes, scientific research habits and behaviors of scientific research personnel. In short, the scholar portrait has multiple dimensions of information and highlights the modeling of the academic field. Through the extraction, analysis and mining of the attribute information of multiple dimensions in the academic field, an abstracted scientific research model of the scholar is formed. The role of discipline service is to improve the efficiency and discipline ability of the scholar. When constructing the scholar portrait, the basic information of the scholar, the scientific research dynamic information and the basic information of the discipline should be considered as the portrait dimensions. At the same time, the timeliness should also be considered. The discipline portrait label should be adjusted in time to ensure the accuracy of the service.

[0182] 1.1 Dimensions of scholar portrait

[0183] Natural dimension, interest dimension, social dimension, etc. are the multi-level tag system of the user portrait model of digital library. Combined with the needs of discipline service, the scholar portrait is constructed according to four dimensions of personal information, academic achievement, research interest and hot behavior.

[0184] (1) Personal information dimension. It includes name, campus number, scholar identification number, gender, degree, identity, birth date, country, university, department, research field, academic level index, academic relationship, academic title, journal reviewer, academic group position, social relationship, etc. It is the basic data to describe the characteristics of the scholar.

[0185] (2) Academic achievement dimension. It includes papers, works, patents, teaching materials, scientific research projects, awards, decision-making consulting reports, etc. It also includes the total number of published papers, the total number of cited papers, H index, and the total score of scientific research in the past 5 years, etc.

[0186] (3) Research interest dimension. Research interest is a personalized characteristic of the scholar, which is an important basis for providing discipline service. The research interest of the scholar is extracted from the academic achievements in a certain period of time (10 years or 5 years).

[0187] (4) Hot behavior dimension. In order to solve the lag problem of the research interest extracted from the academic achievement, the hot behavior dimension is added, including the literature read by the scholar in the past year, the projects applied for, etc. The latest hot research behavior is extracted.

[0188] 1.2 Construction process of scholar portrait

[0189] The construction of scholar portrait is divided into five steps: data collection, data preprocessing, data storage, data processing, and label generation. Data collection mainly obtains basic information and academic achievements of scholars from the school data center and various application systems, obtains scientific research behavior data from academic resource access control systems and network behavior logs, and obtains some off-campus scholar data or supplementary data of on-campus scholars from the Internet. After data classification, cleaning, fusion, and deduplication, the data is stored and indexed according to the scholar's basic information, scientific research behavior data, and Internet information, and a data exchange interface is established for data exchange. Finally, through statistical analysis, text mining, clustering analysis, and feature extraction, the labels of each scholar are generated in four dimensions, as shown in FIG. 1. Figure 3

[0190] 1.3 Data Collection

[0191] The data generated in the process of scientific research is scattered in the school data center, various application systems, and log systems, and there are also some on the Internet. Except for independent documents and paper files that need to be manually imported, all of them can be integrated through tool software. The data integration tool ODI (Oracle Data Integrator) or DATA-X tool software is used to efficiently realize the extraction, conversion, and loading of batch data, support the integration of mainstream relational databases, and can be refreshed and executed, timed, with high automation, or read by API at regular intervals. The data of the discipline service platform uses the relational database system My-SQL, the network scientific research behavior log uses the firewall access log and the website access log, and the distributed file system HDFS of Hadoop. However, as more and more database vendors use the Https protocol and the database server is not inside the school, it is impossible to obtain complete access logs. At present, the digital resource access control system can be used to obtain complete access logs and operation behaviors such as search terms, so that the data required by the scholar portrait can be directly connected to the system to obtain it, greatly simplifying the workload of analyzing access logs. The data sources and integration methods of the scholar portrait are shown in Table 1.

[0192] ​The keywords are extracted from the research results in the school institutional repository, Baidu Library or other websites as research interests. In order to express the level of research interest, the binary group <keyword, frequency> is used, and the keywords are cross-existed in multiple achievements, and the frequency is the sum of the appearance frequency in all achievements. Hot behavior needs to be obtained from recent scientific research behavior, including digital resource access log, network behavior log, scientific research project declaration form, etc. For universities using digital resource access control system, access behavior data can be directly read, otherwise network exit log and VPN log need to be analyzed, but the log acquisition and analysis are difficult, and currently general not to be used, scientific research project declaration form needs to be obtained through scientific research management platform or offline. Hot behavior also uses binary group <keyword, frequency> to express, and different behavior frequencies need to be normalized.

[0193] 1.4 Scholar portrait label generation

[0194] On the basis of comprehensive data, the portrait label is formed to avoid directly using metadata as a label, and the lack of data mining and clustering calculation will lead to too many labels and inaccurate description. Label design should pay attention to the intuitiveness of the portrait and the convenience of the application. Traditional labels include structured, semi-structured and unstructured three types, basic information and academic achievement dimension are structured data, research interest and hot behavior dimension are also structured data converted to <keyword, frequency>, scholar portrait has multiple labels, and each label has several values. The main label categories and classification values of scholar portrait are shown in Table 2.

[0195] The label of personal information can be directly generated, and each label of academic achievement dimension is first converted to specific score and sorted according to the school academic achievement score. Since the data used to construct the portrait is very different, it is difficult to determine a fixed interval for segmentation like age, so the interval needs to be determined according to the data. The natural breakpoint classification method is used to determine the clustering number N value, and the variance goodness of fit is used for judgment. In order to simplify the calculation, it is uniformly divided into three grades of high, medium and low, that is, N=3, but natural breakpoint method still needs to be used for classification, not one third of high, medium and low, and finally the academic achievement portrait label of the scholar is determined, such as high paper, low book, high longitudinal and medium horizontal.

[0196] Scholar research interest and hot behavior are represented by binary group <research interest, frequency>, research interest can be directly used as label name, and frequency is calculated to generate weight to measure the importance of research interest. Another factor affecting research interest is time decay, which means that the user's behavior gradually weakens with the passage of time and current relevance. The time decay function is established by referring to Newton's cooling law, and the current weight of research interest label is calculated by combining the two.

[0197] 2 Design of subject service platform

[0198] 2.1 Overall Architecture of Discipline Service Platform

[0199] The discipline service platform adopts a three-tier architecture of database layer, intermediate data processing layer, and front-end client layer. The database layer includes a data collection module; the data processing layer includes a scholar portrait module, a discipline data storage and calculation module; the client layer includes a resource recommendation module, a discipline analysis module, which is the entry point for users such as scholars, discipline leaders, functional departments, and subject librarians to use the platform. Scholars can view their research data, customize relevant parameters, and receive academic resource push information through WeChat and email. Discipline leaders can view discipline development data. School functional departments can view the comprehensive development of each discipline. Subject librarians maintain basic discipline data, various platform parameters, scholar lists, system data interfaces, and import a small amount of data, and promptly handle various needs of scholars and functional departments. The overall framework of the discipline service platform is shown in Figure 4 .

[0200] 2.2 Design of Data Tables

[0201] When designing tables in a relational database, attributes and intermediate tables related to the four dimensions of scholar portraits need to be added. The scholar number is the primary key, and the school's faculty and student ID numbers are used. Postdoctoral schools also have numbers, and external scholars can use the ORCID, which has a globally unique 16-digit identification code. The main data tables are described as follows:

[0202] (1) Scholar Basic Information Table: Attributes include ID (VARCHAR2, 16, primary key), name, identity, gender, birth date, title, degree, department and major, discipline, contact information, talent echelon, total academic achievement score, H-index, and number of citations. It is dynamically updated through the school data center interface.

[0203] (2) Scholar Portrait Label Table: Attributes include ID (VARCHAR2, 16, primary key), label, weight, and creation time. A scholar includes multiple records to represent multiple labels and corresponding weights. Recalculation is performed regularly.

[0204] (3) Discipline Basic Information Table: Attributes include discipline code (ID, VARCHAR2, 16, primary key), discipline name, department, discipline leader, discipline nature, establishment time, address, and contact information.

[0205] (4) Discipline Portrait Label Table: Attributes include discipline code (ID, VARCHAR2, 16, primary key), label, weight, and creation time. It is compiled from the scholar portrait labels of all discipline members.

[0206] (5) Academic achievement table: records all the academic achievements of scholars in the discipline, attributes include worker number / student number (ID, VARCHAR2, 16, primary key), name, discipline code, achievement name, nature, ranking, achievement score, and time of achievement.

[0207] (6) Research interest table: attributes include worker number / student number (ID, VARCHAR2, 16, primary key), keyword, and weight. Extracted from the scholar's research achievements in the past ten years, recalculated regularly.

[0208] (7) Hot behavior table: worker number / student number (ID, VARCHAR2, 16, primary key), keyword, and weight. Extracted from the scholar's research behavior in the past year, recalculated regularly.

[0209] (8) Latest academic resource table: resource name, resource nature, author, source, time, and keyword set.

[0210] (9) Recommended results table: worker number / student number (ID, VARCHAR2, 16, primary key), resource name, author, time, keyword, and similarity. It is the result of matching the latest academic resource table with the scholar, recommending academic resources according to similarity, and calculating once before recommending resources each time.

[0211] 2.3 Resource Push

[0212] Academic resource recommendation is one of the core contents of discipline service, which can be divided into three steps: resource collection, matching, and personalized push. Resource collection refers to collecting newly published papers, books, academic conferences, patents, and project application notices, etc. The methods include automatic collection, web crawler, and manual input, such as the latest books in the library are obtained by connecting the library system; academic conference notices and project application notices are crawled from relevant professional websites, and electronic resources are crawled from relevant resource database websites. Specific source journals and websites are set for each discipline, which are read through the institutional repository, obtained by web crawler, and a small amount of resources can also be manually imported. Periodically match the collected academic resources with the scholar's portrait to achieve personalized recommendation, the specific method is as follows:

[0213] For each newly added academic resource and each scholar, execute:

[0214] Extract the keyword set X of the academic resource

[0215] Extract the research interest and hot behavior tag set Y from the scholar's portrait

[0216] Calculate the similarity of the two, and write the resources and scholars with higher similarity into the recommended results table;

[0217] The calculation of similarity selects Jaccard similarity. The greater the value of Jaccard (X, Y), the more similar the academic resources and the scholar's research interests. The formula for calculating Jaccard similarity is:

[0218]

[0219] Where text (X) represents the attribute characteristics of the academic resource keyword set X, text (Y) represents the attribute characteristics of the research interests and hot behavior label set Y in the scholar portrait, text (X) ∩ text (Y) represents the number of the same words in the attribute characteristics of X and Y, and text (X) ∪ text (Y) represents the total number of words in the attribute characteristics of X and Y. Thus, the formula can calculate the text similarity of the two attribute characteristics. The greater the value of Jaccard (X, Y), the closer the resource and the scholar's research interests.

[0220] 2.4 Subject analysis and application

[0221] The subject service platform integrates all the data of scholars, such as basic information, research achievements and research interests, and can realize the functions of subject development data analysis and display, assistance in subject team building and talent introduction, etc.

[0222] (1) Subject data analysis and display

[0223] Using the information of subjects and scholars on the platform, the subject development data can be analyzed and displayed in multiple dimensions, such as correlation analysis of scholars' graduation institutions and research achievements, short board analysis of subject development, subject member echelon analysis, scholar research behavior and research status analysis, research activity analysis, and latest research trend analysis. The results can be displayed on a large screen at a glance. Even a reference list of "fake scholars" (no achievements and no research behavior), "free riders" (with achievements but no research behavior), and "inefficient researchers" (many research behaviors but few achievements) can be provided. However, it is difficult to confirm the research behavior, so it is temporarily not easy to realize.

[0224] Precise division of subject members

[0225] The traditional members of a discipline team mainly come from a certain department, and a few members come from other departments. Due to the existence of school mergers, affiliated hospitals and other relatively independent departments, the members of a discipline are more scattered. Through the scholar portrait, different department personnel can be collected under the name of the discipline, including visiting scholars, post-doctors and graduate students, etc. temporary research forces, so that the members of the discipline are more comprehensive and accurate. Through research achievements, scientific research behavior data, etc. The members of the discipline are marked as “senior scholars”, “core members” and “active researchers”, etc. so that the young teachers, post-doctors and graduate students, etc. who are easily overlooked in the past can get the due attention. Through the scholar portrait, the subject librarian can master the differentiation of the service object, and send the latest information on the forefront of the discipline research to the relevant members of the discipline. It is more targeted than sending information uniformly, and more efficient than individual consultation services. Through research achievements, scientific research behavior data, it is beneficial to form new research teams and carry out scientific research cooperation within the school.

[0226] (3) Assist in the introduction of discipline talents

[0227] Through the analysis of the publication journals, research directions and keywords of the achievements of the members of the discipline, the research theme of the discipline can be obtained. According to the research theme, the list of excellent scholars engaged in research in this direction can be found out, which can help the discipline leaders and personnel departments to lock in the introduction targets, track their scientific research achievements and research interests, understand their potential positions in the discipline, and facilitate the salary and position arrangement when introducing talents, so that the talent introduction work is more proactive, more accurate and data-supported.

[0228] 3 Platform development

[0229] The subject service platform deploys database servers and front-end servers on the school cloud platform, is based on B / S architecture, uses Python, PyCharm development tools and Django Web application development framework for development, finally uses MySQL relational database for data storage, and uses Navicat tool for data management. It can manage and operate various types of databases through intuitive graphical interface, which is convenient for debugging during development. The front end is based on HTML, CSS and JavaScript technologies, and the interface is relatively simple.

[0230] As mentioned earlier, the main functional modules include data collection, scholar portrait, discipline data calculation, resource recommendation, discipline analysis and other five modules. In order to facilitate the use of users such as scholars, discipline leaders, functional departments and subject librarians, user management, data management (including data collection), system parameter configuration and other modules are also designed to facilitate system operation and maintenance.

[0231] 3.1 User management module

[0232] The unified identity authentication system of the docking school is used to send an authentication request to the authentication system after the user inputs the account and password. The system verifies the validity and returns the identity information. The real system should have strict user permission management function, including subject librarian, subject leader, personnel department, scholar, etc. The experimental system only has two types of users: subject librarian (system administrator) and scholar, and simplifies the permission allocation.

[0233] 3.2 Data management module

[0234] It mainly includes four sub-modules: data collection, data preprocessing, data updating, and data export. Data collection and preprocessing are the basis for constructing scholar portraits. Data collection refers to the process of obtaining the required data from various data sources. Data preprocessing is the process of cleaning, deduplicating, and converting raw data to facilitate subsequent analysis and application.

[0235] Data collection. In the absence of a data center, ODI or DATA-X tools are used to achieve this. Otherwise, it is achieved through API interfaces based on the school's data center platform. When calling API interface services, the application identifier and usage key are passed in to sign the parameters for identity authentication, improving security. The data from the data center is more accurate and comprehensive after cleaning, reducing the workload of data preprocessing.

[0236] (2) Data preprocessing. To ensure the integrity, uniformity, and accuracy of the data set, data preprocessing is required before using the data to convert it into data that can be used to construct a university teacher research portrait. The process of data preprocessing varies depending on the attributes and tasks of different data sets. It generally includes three steps: data cleaning, data conversion, and data integration. Data cleaning refers to the process of filtering, filling, and deleting redundant, redundant, missing, and abnormal data to ensure data consistency and integrity. Specific operations include filling in missing values, removing duplicates, and deleting unnecessary fields. Data conversion refers to converting raw data into a uniform format suitable for specific analysis tasks. Data integration integrates data from different data sources into a complete data set for subsequent data analysis and mining.

[0237] (3) Data updating. Python scripts are executed during data collection and data preprocessing to update the data source permissions for administrators, ensuring the accuracy and completeness of the data.

[0238] (4) Data export. The data in each data table can be exported as needed for use by other applications.

[0239] 3.3 Subject data calculation module

[0240] According to the design of the data table in 2.2, the subject service platform has 9 data tables in total, among which the scholar basic information table, subject basic information table, academic achievement table and scientific research behavior log are the original data, which need to be recalculated to generate the research interest table and hot behavior table after regular update. The task of this module is to generate the research interest table and hot behavior table. It is worth noting that the scientific research behavior log can only be obtained through the digital resource access control system at present, and this system is not available in all schools. Even if it is available, not all scholars' resource access goes through it, and scientific research behavior is not only accessing digital resources. Therefore, this part is not implemented at present, although there is no hot behavior table data, the research interest table system can still run normally, but the research interest tags generated are not timely enough, which affects the effect of resource recommendation.

[0241] 3.4 Scholar portrait module

[0242] It includes three sub-modules: scholar portrait generation, portrait combination query and portrait statistical analysis. The scholar portrait generation is to generate the scholar portrait tag table and subject portrait tag table according to the original data and the research interest table and hot behavior table generated by the subject data calculation module according to the method in 1.4. The portrait combination query is to query the scholar portrait tag table, such as querying "non-library" + "subject service" to get the scholars who have related research results of "subject service" and do not belong to the library. The portrait statistical analysis is also similar, which can finally form a statistical report.

[0243] 3.5 Resource recommendation module

[0244] It is divided into four sub-modules: resource acquisition, resource import, resource matching and result sending. ① Resource acquisition. Academic resources are diverse, focusing on new books in the library, project application notices and academic conferences, whose data sources are the library book system, the website of the school science and technology department / sociology department and China Academic Conference Network (conf.cnki.net), and the data integration methods are API docking and web crawler respectively. Compared with the data acquisition of new books and project application, the data obtained by web crawler method needs to be processed before it can be applied, so it is not considered to be implemented at present. ② Resource import. Extract the relevant fields from the resource information and fill them into the latest academic resource table. In addition to automatic import, it also supports manual import, such as academic lecture information. ③ Resource matching. Match according to the method in 1.3, store the results in the recommended result table, and sort them according to the similarity. ④ Result sending. Call the message center to send the corresponding recommended content to the scholar, and save the previous recommended results in the recommended result table for the scholar to check and evaluate the accuracy of the recommendation, which can be used as a reference for improving the recommendation algorithm. In the next step, the subject service platform and subject librarian can communicate with the scholar through WeChat, and the recommended results can be sent to WeChat at regular intervals.

[0245] 3.6 System parameter setting module

[0246] To increase the flexibility of the system, the need to protect privacy, some parameters scholars can set independently: ① scholars can set to open or close resource recommendation, since the opening of the recommended function to collect the individual behavior information need to agree to the default off; ② the number of recommended resources each time, the default 50; ③ recommended information transmission mode, message center support pieces, WeChat and SMS and their combination, the default only send email; ④ the maintenance of the scholar's personal information, such as telephone email, degree and title changes, school personnel can automatically synchronize; ⑤ scholars can define research interests and hot behavior, which recommended academic resources. And the discipline of research interest decay coefficient, journal source directory, conference website, etc. need to be set by subject librarian.

[0247] After the completion of the docking of each data source, the subject service platform generates labels, weights and time decay calculations at regular intervals every day to ensure that the platform maintains the latest discipline data, and sends recommended resources at regular intervals every week. The automatic operation mode reduces the dependence on subject librarians, and the subject leaders and management departments enter the platform to understand the development of the discipline. In addition, the platform records the whole process of subject service, which is conducive to the work handover of subject librarians. The construction of the subject service platform under the support of the scholar portrait can improve the service capacity, reconstruct the service mode, and improve the service level.

[0248] Table 1: Scholar portrait data sources and integration methods

[0249]

[0250] Table 2: Main labels of scholar portrait

[0251]

[0252] Example 1: Personal academic resource recommendation system based on scholar portrait

[0253] 1. System operation process

[0254] This embodiment constructs a personalized academic resource recommendation system based on a scholar portrait, uses data collection, portrait generation, resource matching and recommendation pushing modules, and realizes precise academic resource pushing.

[0255] 2. Specific implementation steps

[0256] (1) User identity authentication and access authorization

[0257] After the user logs in, the user management module sends an authentication request to the school unified identity authentication system, receives an identity verification signal, parses the user identity, and assigns access rights.

[0258] After authentication, the user management module sends user identity data stream to the data management module, including user ID, discipline field, title information, etc.

[0259] (2) Data collection and processing

[0260] The data management module calls the API interface to collect data such as papers published by scholars, research activities, and citation information from data sources such as academic paper libraries, research project databases, and academic conference websites.

[0261] The collected data is pre-processed by the data preprocessing submodule to remove duplicates, standardize formats, and filter outliers, and is finally stored in the database and sent to the subject data calculation module to send data update signals.

[0262] (3) Scholar portrait generation

[0263] The subject data calculation module regularly calculates the research interests of scholars, analyzes paper keywords and research behavior logs, and generates research interest tables and hot behavior tables.

[0264] The scholar portrait module receives research interest table data, clusters the research directions of scholars based on machine learning algorithms, generates scholar portrait labels, and stores them in the portrait database.

[0265] (4) Resource recommendation

[0266] The resource recommendation module regularly obtains the latest academic resources, including new book information, academic conferences, and research projects, and stores them in the academic resource table.

[0267] The resource matching submodule calls the collaborative filtering algorithm, calculates the similarity between resources and the research interests of scholars based on the scholar portrait, and stores the sorted recommendation results table according to the similarity.

[0268] (5) Recommended result pushing

[0269] The result sending submodule calls the message center to push recommended resources to scholars through email, WeChat, or in-site messages.

[0270] Scholars can view the recommended content and rate the accuracy of the recommendations, and the rating data is stored in the recommendation feedback database for subsequent algorithm optimization.

[0271] Example 2: Dynamic scientific research data monitoring and analysis system for university subject service

[0272] 1. System operation process

[0273] This embodiment constructs a university scientific research data monitoring and analysis system, which regularly collects and calculates scientific research behavior data to dynamically monitor subject hotspots and research trends, and provides analysis support for the development of university subjects for subject librarians and managers.

[0274] 2. Specific implementation steps

[0275] (1) Scientific research data collection and preprocessing

[0276] The data management module calls the API interface to obtain the latest scientific research project, paper, patent, and scientific research fund data from the university scientific research management system, national natural science foundation database, and know network platform.

[0277] The collected data is subjected to format standardization, field mapping, and de-duplication operations by the data preprocessing submodule, and the cleaned data is stored in the database and sent to the subject data calculation module.

[0278] (2) Scientific research trend calculation and subject hotspot analysis

[0279] The subject data calculation module receives the data update signal from the data management module, extracts the scientific research behavior log, and analyzes the subject hotspot, including paper publication trend, research direction evolution, and popular cooperation institutions.

[0280] A time series analysis model is used to predict scientific research data and calculate the possible development direction of a subject in the next 3-5 years, and a trend report is generated and stored in the hotspot behavior table.

[0281] (3) Subject portrait generation

[0282] The scholar portrait module clusters the scientific research behavior of scholars at different time periods to generate subject portrait labels.

[0283] The portrait combination query submodule allows subject librarians to query research scholars with specific keywords, such as querying "deep learning + medical imaging" to obtain the main researchers in this field and their cooperation network.

[0284] (4) Data visualization and decision support

[0285] The portrait statistical analysis submodule receives the analysis request of the administrator, performs data calculation, and generates statistical analysis reports.

[0286] The data export submodule supports exporting subject trend analysis data, and administrators can view subject development reports through the visualization platform to support scientific research management and subject planning decisions.

[0287] (5) System parameter setting and automatic update

[0288] The system parameter setting module allows administrators to set the data update period, with the default being to perform data synchronization tasks every morning.

[0289] The research interest customization submodule allows subject librarians to manually adjust subject hotspot keywords to optimize trend analysis results.

[0290] Administrators can view data update records through system monitoring logs to ensure stable operation of the system.

[0291] It should be noted that embodiments of the present application can be realized by hardware, software, or a combination of software and hardware. The hardware portion can be realized by a special logic; the software portion can be stored in a memory and executed by a proper instruction execution system, such as a microprocessor or a specially designed hardware. Those skilled in the art can understand that the above-mentioned apparatus and method can be realized by computer executable instructions and / or included in processor control codes, such as a carrier medium, such as a magnetic disk, CD or DVD-ROM, a programmable memory, such as a read-only memory (firmware), or a data carrier, such as an optical or electronic signal carrier. The apparatus of the present application and its modules can be realized by a hardware circuit, such as a very large scale integrated circuit or a gate array, a semiconductor, such as a logic chip, a transistor, or a programmable hardware device, such as a field programmable gate array, a programmable logic device, or the like, by software executed by various types of processors, or by a combination of the above-mentioned hardware circuit and software, such as firmware.

[0292] The above description is merely a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any modification, equivalent replacement, and improvement within the technical range disclosed by the present application, and within the spirit and principle of the present application, should be covered within the protection scope of the present application.

Claims

1. A high school individual chemistry subject service system, characterized by comprising: The system comprises a user management module, a data management module, a discipline data calculation module, a scholar portrait module, a resource recommendation module, and a system parameter setting module. The user management module is connected with the data management module, receives user input identity authentication information, sends an authentication request to a school unified identity authentication system, receives an identity verification signal returned by the authentication system, analyzes and obtains user identity information, and sends an authentication result signal to the data management module. The data management module is connected with the user management module, the discipline data calculation module, the scholar portrait module, the resource recommendation module, and the system parameter setting module, receives an authentication result signal from the user management module, sends an access authorization signal to the discipline data calculation module, the scholar portrait module, and the resource recommendation module, and performs data collection, data preprocessing, data updating, and data export tasks. The discipline data calculation module is connected with the data management module, receives data signals from the data management module, analyzes the data, performs calculation to generate a research interest table and a hot behavior table, and returns calculation result data streams to the data management module. The scholar portrait module is connected with the data management module, receives user behavior data streams transmitted by the data management module, analyzes and calculates user feature information, generates scholar portrait label data, and transmits portrait data streams to the data management module and the resource recommendation module. The resource recommendation module is connected with the data management module, receives portrait data from the scholar portrait module, combines academic resource library data streams, performs resource matching calculation, generates recommended resource data streams, and sends a recommendation result signal to the system parameter setting module. The system parameter setting module is connected with the data management module, receives parameter data streams set by an administrator, and adjusts data collection, calculation, and recommendation processing rules.

2. The high school individual chemistry subject service system according to claim 1, wherein, The data management module comprises a data collection submodule, a data preprocessing submodule, a data updating submodule, and a data export submodule. The data collection submodule receives raw data streams from multiple data sources, and acquires and stores the raw data streams to a database through an API interface or a data exchange signal. The data preprocessing submodule receives raw data streams from the data collection submodule, performs data cleaning, deduplication, anomaly detection, and format conversion processing, and generates preprocessed data streams and sends the preprocessed data streams to the data management module. The data updating submodule receives task scheduling signals from the data management module, calls a Python interpreter to execute a data updating script, stores updated data in a database, and returns a data updating completion signal to the data management module. The data export submodule receives a data export request signal, extracts data from the database, encapsulates the data into a data stream in a specified format for export, and sends a data transmission completion signal.

3. The high school individual chemistry subject service system according to claim 1, wherein, The discipline data calculation module comprises a scholar basic information table, a discipline basic information table, an academic achievement table, and a scientific research behavior log, and is configured to perform calculation to generate a research interest table and a hot behavior table based on the data tables and update the tables periodically. The research interest computing unit is configured to receive the academic achievement data stream, parse the research direction data, compute the scholar research interest, update the research interest table, and generate an interest update completion signal; The hotspot behavior computing unit is configured to receive the scientific research behavior log data stream, perform statistical analysis, compute the scientific research hotspot trend, update the hotspot behavior table, and generate a hotspot behavior computation completion signal.

4. The high school individual chemistry subject service system according to claim 1, wherein, The scholar portrait module includes a scholar portrait generation sub-module, a portrait combination query sub-module, and a portrait statistical analysis sub-module, wherein: The scholar portrait generation sub-module is configured to receive the research interest table and the hotspot behavior table data stream, perform data clustering and feature extraction, generate scholar portrait label data, and store the data in a portrait database; The portrait combination query sub-module is configured to receive a query request signal, parse the query conditions, perform a query in the scholar portrait database, and return a query result data stream; The portrait statistical analysis sub-module is configured to receive a statistical analysis request signal, perform statistical computation, generate a statistical result data stream, and return the data stream to the data management module or the data visualization display module.

5. The high school individual chemistry subject service system according to claim 1, wherein, The resource recommendation module includes a resource acquisition sub-module, a resource import sub-module, a resource matching sub-module, and a result sending sub-module, wherein: The resource acquisition sub-module is configured to receive a resource update signal from the data management module, call an API interface or a network crawler to collect the latest academic resource data stream, and store the data stream in a resource database; The resource import sub-module is configured to receive the resource data stream, parse the field information, store the standardized information in an academic resource table, and return a resource import completion signal to the data management module; The resource matching sub-module is configured to receive the scholar portrait data stream, perform similarity computation, generate a matching result data stream, store the data stream in a recommendation result table after sorting according to the relevance, and return the data stream to the data management module; The result sending sub-module is configured to call a message center, send the recommendation data stream to a target user, record a sending log, and send a recommendation completion signal to the data management module.

6. The high school individual chemistry subject service system according to claim 1, wherein, The system parameter setting module includes a resource recommendation switch setting sub-module, a recommendation information sending method setting sub-module, a scholar personal information maintenance sub-module, a research interest customization sub-module, and a discipline data update setting sub-module, wherein: The resource recommendation switch setting sub-module is configured to receive a recommendation switch setting signal from a user, modify the user recommendation state, and return a modification success signal; The recommendation information sending method setting sub-module is configured to receive a sending method setting signal from a user, update a message center sending strategy, and return a parameter setting success signal; The scholar personal information maintenance sub-module is configured to receive a user information change signal, update a user database, and return an update success signal to the data management module; The research interest customization sub-module is configured to receive a user-set research interest data stream, update a user portrait data, and return an interest modification success signal to the data management module; The discipline data update setting sub-module is configured to set a data update period, trigger an automatic update process, and return an update strategy setting success signal.

7. A method for providing a high school individual chemistry subject service system according to any one of claims 1 to 6, characterized by, The college individual chemical discipline service method includes: Step 1, through the user management module to the unified identity authentication system of the interface school, the user inputs the account password and sends an authentication request to the authentication system, the system verifies the validity, and after the verification, the system generates a research interest table, a hot behavior table according to the returned identity information; Step 2, through the data management module data acquisition, data preprocessing, data updating and data export; Step 3, through the subject data calculation module, the scholar basic information table, the subject basic information table, the academic achievement table and the scientific research behavior log original data are regularly updated, and the research interest table and the hot behavior table need to be recalculated and generated, which is the task of the module; Step 4, through the scholar portrait module, the scholar portrait generation, portrait combination query and portrait statistical analysis; Step 5, through the resource recommendation module, the resource acquisition, resource import, resource matching and result sending; Step 6, through the system parameter setting module, the system parameters are set.

8. A computer device, comprising: The computer device comprises a memory and a processor, the memory stores a computer program, and the computer program is executed by the processor to make the processor execute the steps of the college personalized subject service method of claim 7.

9. A computer readable storage medium storing a computer program, the computer program being executed by a processor to make the processor execute the steps of the college personalized subject service method of claim 7.

10. An information data processing terminal, characterized by The information data processing terminal is used to realize the college personalized subject service system of any one of claims 1-6.

Images