Student ability analysis and employment recommendation system based on big data

Abstract

The invention discloses a student ability analysis and employment recommendation system based on big data, and relates to the technical field of employment recommendation. Comprising a data acquisition module, a data quality monitoring module, a real-time data updating module, a personalized analysis module, a prospective recommendation module, an interactive platform module, a learning track analysis module and an industry trend monitoring module. The comprehensive ability value of each student is determined according to the course score, the interest label and the community activity of the student, the potential professional interest and the advantage field of the student are mined, then the industry trend is predicted based on the current industry data through the industry trend monitoring module, and the prospective recommendation module combines the learning ability, the interest and the advantages of the student to obtain a recommendation result. Personal growth and occupational development planning of the students are considered, posts capable of improving ability and expanding occupational space are recommended for the students, and the matching degree of employment post recommendation is greatly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of employment recommendation, and in particular to a student ability analysis and employment recommendation system based on big data. Background Art

[0002] In today's society, students' employment issues have attracted much attention, because this is not only related to the personal development of students, but also of great significance to social stability and economic development. With the improvement of education level and increasingly fierce competition in the job market, students need more accurate and personalized employment recommendations to improve their employment success rate and career satisfaction. In order to meet this demand, many educational institutions and enterprises have begun to explore the use of big data technology to analyze students' abilities and potential, so as to provide them with more appropriate employment recommendations. Big data technology has the advantages of large data volume, wide sources, and fast processing speed. It can integrate students' academic performance, extracurricular practice, online learning records, social network behavior and other data, so as to have a more comprehensive understanding of students' abilities and interests;

[0003] The existing student ability analysis and employment recommendation systems have certain defects. First, most recommendation systems will rank students according to their grades, rank positions according to their merits, and select and match them according to students' grades. The degree of personalization and intelligence is insufficient, and industry trends and student development plans are not fully considered. Secondly, data collection channels are limited, data quality monitoring is imperfect, and changes in students' abilities cannot be determined, affecting the accuracy of recommendations. Finally, the communication channels between students and employers in existing technologies are not smooth and interactivity is insufficient. After choosing a position, many students can only understand the job content based on the position name, and are not clear about the company's specific job requirements and work content, resulting in low employment matching and failure to meet the needs of both parties. To this end, we propose a student ability analysis and employment recommendation system based on big data. Summary of the invention

[0004] The purpose of the present invention is to provide a student ability analysis and employment recommendation system based on big data.

[0005] In order to solve the problems raised in the above background technology, the present invention provides the following technical solutions: a student ability analysis and employment recommendation system based on big data, the system includes a data acquisition module, a data quality monitoring module, a real-time data update module, a personalized analysis module, a forward-looking recommendation module, an interactive platform module, a learning trajectory analysis module and an industry trend monitoring module;

[0006] The data acquisition module acquires multi-source data from external data sources, including learning data and internship data. The real-time data update module tracks the students' latest learning and internship data through sensor technology and mobile application interface, and updates the acquired data in time. The data quality monitoring module can acquire data from the data acquisition module and the real-time data update module, and use the autoencoder model and data verification algorithm to detect anomalies in the data, calculate the reconstruction error, modify and check the abnormal data, and transmit the final normal data to the database. The personalized analysis module can extract the learning data in the database, analyze the students' course selection history and grade distribution based on the deep learning algorithm, and calculate the students' comprehensive ability value through the formula. The industry trend monitoring module It can collect industry information, predict industry trends through time series analysis technology, and transmit industry trend data to the forward-looking recommendation module. The learning trajectory analysis module extracts students' learning behavior characteristics through data mining technology, and obtains students' learning trajectory data based on data in the database, and transmits the trajectory data to the personalized analysis module. The forward-looking recommendation module obtains the prediction curve of future talent demand based on the student personalized ability model of the personalized analysis module and the industry trend prediction data of the industry trend monitoring module, and combines the trend fitting function formula to generate a personalized job recommendation list for students, and helps students and enterprises understand each other through the interactive platform module, helps students understand the recommended positions in detail, and helps enterprises screen job personnel.

[0007] As a further solution of the present invention: the data acquisition module is provided with a data acquisition unit and a database unit, and a connection is established between the data acquisition unit and the online learning community and the internship recruitment platform. The data acquisition unit is set with a trigger condition for data acquisition. When new data is generated from the external data source and according to a preset time interval, the acquisition program is started. Different acquisition strategies are adopted for different types of data. For learning progress data, it is directly obtained from the database interface of the online learning platform. For internship project achievement data, it is obtained by interacting with the API of the internship recruitment platform. The collected data is encrypted and transmitted and stored in the database unit. The database unit uses a hash algorithm to deduplicate the data, identifies and removes the same data items, and then, through data format conversion, unifies the data from different sources into a format recognizable by the system. At the same time, the data acquisition module adopts an adaptive data acquisition frequency adjustment mechanism to dynamically adjust the acquisition frequency according to the data update speed and importance of different data sources. For online learning community data that is frequently updated and has a greater impact on employment recommendations, it is collected once every half an hour, while for relatively stable enterprise evaluation data, it is collected once every two days.

[0008] As a further solution of the present invention: the real-time data update module includes a data update unit and a historical data storage unit. The data update unit is provided with sensor technology and a mobile application interface. When students participate in offline activities, the sensor detects an activity signal and triggers a data update process. First, the activity information is converted into a storable data format. The actual activity duration is t1, the planned activity duration is t2, and the activity participation is P. The specific activity participation calculation formula is as follows:

[0009]

[0010] Where: σ is the activity duration fluctuation parameter obtained by statistics based on the historical data stored in the historical data storage unit;

[0011] The data update unit will transmit the calculated data transmission to the historical data storage unit through the message queue technology. When new data is generated, the data will be placed in the message queue and processed asynchronously by the background processing program to avoid the impact of data updates on system performance. At the same time, data caching technology is used to store frequently accessed data in the cache to improve data reading speed and provide users with real-time employment recommendation services. Blockchain technology is introduced to ensure that data updates cannot be tampered with and are traceable. Each data update is recorded on the blockchain to form a data block, which contains detailed information and timestamps of the data. The authenticity and reliability of the data are guaranteed by encryption algorithms. The real-time data update module will transmit the updated data to the data quality monitoring module for detection.

[0012] As a further solution of the present invention: the data quality monitoring module can receive data transmitted by the data acquisition module and the real-time data update module, and the data quality monitoring module uses a data verification algorithm and is provided with an autoencoder model based on deep learning for data reconstruction and anomaly detection. Assume that the input data is X={x1,x2,…,x n}, the hidden layer is represented as H after the encoding function f(X)=W1X+b1 of the autoencoder, and the reconstructed data is obtained through the decoding function g(H)=W2X+b2 Where W1 and W2 represent the weight matrices of input data and hidden layer data respectively, b1 and b2 represent the encoding bias term and encoding bias term respectively, and then calculate the reconstruction error. The specific reconstruction error calculation formula is as follows:

[0013]

[0014] Where: E represents the reconstruction error, x i represents the original input data points, Represents the data points after data reconstruction;

[0015] When the reconstruction error exceeds the threshold preset by the manager, it is judged as abnormal data. At this time, the manual review process is initiated and checked and corrected by professionals. When the reconstruction error does not exceed the threshold preset by the manager, it is judged as normal data. The data quality monitoring module will store the normal data and the abnormal data checked and repaired by the manager in the database.

[0016] As a further solution of the present invention: the personalized analysis module can obtain the course selection history and grade distribution data of students in the database. The personalized analysis module is provided with a deep learning algorithm to analyze the course selection history and grade distribution data of students. For the course selection history data, text analysis technology is used to extract course topics and difficulty characteristics. For the grade distribution data, the statistical analysis method is used to calculate the mean and variance statistics. Then, a personalized ability model is constructed, and the course grade factor is set to F1, the interest tag factor is F2, the community activity factor is F3, and the student comprehensive ability value is C. The student comprehensive ability value C is calculated by a formula. The specific calculation formula of the student comprehensive ability value C is as follows:

[0017]

[0018] Where: d i It represents the adjustment factor calculated based on the different characteristic dimensions of students, and n is the number of characteristic dimensions;

[0019] The model is trained with a large amount of data. During the training process, the stochastic gradient descent algorithm is used to optimize the model parameters and accelerate the model convergence speed. The personalized analysis module transmits the calculation and analysis results of the model to the forward-looking recommendation module.

[0020] As a further solution of the present invention: the forward-looking recommendation module combines big data analysis and industry trend forecasting technology. The forward-looking recommendation module monitors the development trends of different industries in real time, collects data from multiple industry data sources using web crawler technology, including industry news websites and professional forums, and then uses trend fitting functions to predict future changes in talent demand based on historical data and current trends. Talent demand is calculated using trend fitting functions. The specific trend fitting functions are as follows:

[0021] T(x)=A×sin(ωx+φ)+B×cos(ωx+φ)+k

[0022] Where T(x) represents the talent demand at time x, A, B, ω, φ and k represent the fitting parameters;

[0023] By analyzing a large amount of historical data, the values ​​of fitting parameters are determined, thereby obtaining a forecast curve for future talent demand, recommending positions that meet students' development potential, and improving students' employment success rate. The forward-looking recommendation module uses distributed storage technology to store a large amount of industry data and historical recommendation records. Through data mining technology, valuable information is extracted from these data, including popular positions in the industry and skill requirements. A recommendation algorithm is used to generate a personalized job recommendation list for students based on the students' personalized ability model and industry trend prediction results. The recommendation algorithm uses a combination of collaborative filtering and content recommendation. For collaborative filtering, the similarity between students is calculated, and positions are recommended to target students based on the job selections of similar students. For content recommendation, Analyze the characteristics of the positions and the ability characteristics of the students, make matching recommendations, improve the accuracy and diversity of recommendations, introduce reinforcement learning algorithms, let the system learn and optimize recommendation strategies in the continuous recommendation process, adjust the parameters of the recommendation algorithm according to the students' feedback on the recommended positions, if the students are interested in a certain position and apply for it, the system will increase the recommendation weight of this type of position, if the students are not interested in a certain position, the system will reduce the recommendation probability of this type of position, improve the satisfaction of the recommendation, at the same time, use knowledge graph technology to build an industry knowledge graph, model the enterprises, positions and skill entities in the industry and the relationships between them, and use the knowledge graph to deeply understand the industry relationships and job requirements, providing richer information support for forward-looking recommendations.

[0024] As a further solution of the present invention: the interactive platform module is capable of receiving data transmitted by the forward-looking recommendation module. When students accept the recommended corporate positions, they can communicate with the companies through the interactive platform module. The interactive platform module establishes a two-way communication channel between students and employers. Students can submit questions and resumes through the platform, and employers can publish job information and interview invitations. The platform adopts message push technology to push important information to both parties in a timely manner. When new job information is released and students submit resumes, the system screens out relevant information based on the user's settings and interest preferences, and then pushes the information to the corresponding users through the message push algorithm. During the message push process, an intelligent push algorithm is adopted to push personalized information to users based on the user's interests and behavioral characteristics. According to the student's major and internship experience characteristics, job information of related industries is pushed to the student. According to the recruitment needs of the employer, qualified student resumes are pushed to the employer. Virtual reality technology is introduced to provide a virtual interview environment for students and employers. During the virtual interview process, an interview scene is constructed through three-dimensional modeling technology. Students and employers can interact through virtual reality equipment. Students can display their works and projects, and employers can ask questions and make evaluations.

[0025] As a further solution of the present invention: the learning trajectory analysis module tracks and analyzes the students' learning process, extracts the students' learning behavior characteristics, including learning time, learning frequency, and course completion progress, through data mining technology, and divides students into different learning types using clustering algorithms to provide a more accurate basis for personalized recommendations. In the analysis process, the association rule mining algorithm is used to discover the correlation between students' learning behaviors. Based on these correlations, personalized learning suggestions and employment recommendations are provided to students. At the same time, visualization technology is used to display the learning trajectory analysis results in the form of charts, which is convenient for students and teachers to understand the students' learning situation. The learning trajectory analysis module will transmit the analyzed student learning trajectory data to the personalized analysis module.

[0026] As a further solution of the present invention: the industry trend monitoring module regularly collects industry reports and news information, uses natural language processing technology to analyze text data, and extracts key information, including industry hotspots and technology trends. The analysis results are displayed to users through visualization technology to help students and employers understand industry dynamics. The industry trend monitoring module uses web crawler technology to collect industry information from multiple channels, and uses text classification algorithms to classify the collected text data, including industry news, technical reports and market analysis. In the natural language processing process, word vector technology is used to convert text into digital vectors to facilitate subsequent analysis and processing. The topic model algorithm is used to extract the topic information in the text to understand the hotspots and trends of the industry. At the same time, the attention mechanism in deep learning is introduced to give higher weight to important industry information and improve the accuracy of information extraction. The industry trend monitoring module will transmit industry trend prediction information to the forward-looking recommendation module.

[0027] By adopting the above technical solution, compared with the prior art, the beneficial effects of the present invention are:

[0028] 1. The present invention obtains the latest and accurate student learning and enterprise internship data through the setting of the forward-looking recommendation module, the data collection module, the data quality monitoring module and the real-time data update module, and constructs a personalized student ability model through the personalized analysis module. The comprehensive ability value of each student is determined according to each student's own course grades, interest tags and community activities, and their potential professional interests and advantage areas are excavated. Then, the industry trend monitoring module predicts the industry trend based on the current industry data. The forward-looking recommendation module combines the student's learning ability, interest and advantage, and provides students with job recommendations that meet future market demand through real-time monitoring and analysis of the development trends and talent demand changes in different industries. This intelligent job recommendation not only takes into account the individual specificity of students, but also takes into account the industry development trend, and provides students with job recommendations that meet future market demand. At the same time, considering the student's personal growth and career development plan, it recommends positions that can improve their abilities and expand their career space, which greatly improves the matching degree of job recommendation;

[0029] 2. The present invention connects to online learning communities and internship recruitment platforms through a data acquisition module, and can obtain data such as student academic performance, extracurricular practice, and online learning progress. At the same time, a data quality monitoring mechanism is established through a data quality monitoring module to ensure the accuracy of the acquired data, ensure the accuracy and completeness of the data, and provide a reliable basis for employment recommendations. At the same time, the real-time data update module uses sensor technology and mobile applications to track student ability changes in a timely manner, so that recommendations are more in line with students' actual conditions. These measures can ensure the comprehensiveness and accuracy of the data, help to fully understand the ability status of students, provide a more reliable basis for employment recommendations, and help to improve the accuracy of employment recommendations;

[0030] 3. The present invention establishes a two-way communication channel between students and employers through an interactive platform module, which greatly enhances the interactivity of job recommendations. On this platform, students can fully ask questions to employers and gain an in-depth understanding of the details of the position and the corporate culture, so as to better judge whether the position is in line with their career plans. Employers can also conduct interview appointments, online exchanges and other activities with students through the platform to more comprehensively understand students' abilities and potential. Through this interactive mechanism, students can better understand the needs and expectations of employers, and employers can more accurately find talents that meet their requirements, thereby significantly improving the employment matching degree and bringing convenience and benefits to both students and employers. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 Schematic diagram of the system flow in an embodiment of the present invention. DETAILED DESCRIPTION

[0032] The specific embodiments of the present invention will be further described below in conjunction with the accompanying drawings. It should be noted that the description of these embodiments is used to help understand the present invention, but does not constitute a limitation of the present invention.

[0033] In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0034] Embodiment 1

[0035] In today's society, students' employment issues have attracted much attention, because this is not only related to the personal development of students, but also of great significance to social stability and economic development. With the improvement of education level and increasingly fierce competition in the job market, students need more accurate and personalized employment recommendations to improve their employment success rate and career satisfaction. In order to meet this demand, many educational institutions and enterprises have begun to explore the use of big data technology to analyze students' abilities and potential, so as to provide them with more appropriate employment recommendations. Big data technology has the advantages of large data volume, wide sources, and fast processing speed. It can integrate students' academic performance, extracurricular practice, online learning records, social network behavior and other data, so as to have a more comprehensive understanding of students' abilities and interests;

[0036] The existing student ability analysis and employment recommendation system has certain defects. First, most recommendation systems will rank students according to their grades, rank positions according to their merits, and select and match them according to students' grades. The degree of personalization and intelligence is insufficient, and industry trends and student development plans are not fully considered. Secondly, data collection channels are limited, data quality monitoring is imperfect, and changes in students' abilities cannot be determined, affecting the accuracy of recommendations. Finally, the communication channels between students and employers in existing technologies are not smooth and interactive. After choosing a position, many students can only understand the job content based on the job name, and are not clear about the specific job requirements and work content of the company, resulting in low employment matching and failure to meet the needs of both parties. To this end, we propose a student ability analysis and employment recommendation system based on big data;

[0037] Therefore, in order to effectively solve the above problems, this application proposes a student ability analysis and employment recommendation system based on big data, as shown in the accompanying drawings of the specification. Figure 1 , including data collection module, data quality monitoring module, real-time data update module, personalized analysis module, forward-looking recommendation module, interactive platform module, learning trajectory analysis module and industry trend monitoring module;

[0038] The data acquisition module obtains multi-source data from external data sources, including learning data and internship data. The real-time data update module tracks the students' latest learning and internship data through sensor technology and mobile application interface, and updates the acquired data in time. The data quality monitoring module can obtain the data from the data acquisition module and the real-time data update module, and use the autoencoder model and data verification algorithm to detect anomalies in the data, calculate the reconstruction error, modify and check the abnormal data, and transmit the final normal data to the database. The personalized analysis module can extract the learning data in the database, analyze the students' course selection history and grade distribution based on the deep learning algorithm, and use the formula C= Calculate the comprehensive ability value of students. The industry trend monitoring module can collect industry information, predict industry trends through time series analysis technology, and transmit industry trend data to the forward-looking recommendation module. The learning trajectory analysis module extracts students' learning behavior characteristics through data mining technology, obtains students' learning trajectory data based on the data in the database, and transmits the trajectory data to the personalized analysis module. The forward-looking recommendation module uses the student personalized ability model of the personalized analysis module and the industry trend prediction data of the industry trend monitoring module, and cooperates with the trend fitting function formula T(x)=A×sin(ωx+φ)+B×cos(ωx+φ)+k to obtain the prediction curve of future talent demand, generate personalized job recommendation lists for students, and help students and enterprises understand each other through the interactive platform module, help students understand the recommended positions in detail, and help enterprises screen job personnel;

[0039] Specific workflow: The forward-looking recommendation module collects industry data through web crawler technology, uses trend fitting functions to predict future talent needs, combines students' personalized ability models and industry trends to generate personalized job recommendation lists, and uses reinforcement learning algorithms to optimize recommendation strategies. At the same time, it uses knowledge graph technology to provide richer information support. In actual applications, the forward-looking recommendation module collects industry data from multiple data sources through web crawler technology. After cleaning and preprocessing, it uses trend fitting functions to predict changes in talent demand based on historical data. It combines students' personalized ability models with industry trends, uses collaborative filtering and content recommendation to generate personalized job recommendation lists, and uses reinforcement learning algorithms to optimize recommendation strategies based on student feedback. At the same time, it uses knowledge graph technology to build industry knowledge graphs, deeply understand industry relationships and job needs, and provide richer and more accurate information support for recommendations.

[0040] Furthermore, through the setting of the forward-looking recommendation module, the latest and accurate student learning and corporate internship data are obtained through the data collection module, data quality monitoring module and real-time data update module, and a personalized student ability model is constructed through the personalized analysis module. The comprehensive ability value of each student is determined according to each student's own course grades, interest tags and club activities, and their potential professional interests and areas of strength are explored. Then, the industry trend monitoring module predicts industry trends based on current industry data. The forward-looking recommendation module combines students' learning ability, interests and advantages, and provides students with job recommendations that meet future market needs through real-time monitoring and analysis of development trends and changes in talent demand in different industries. This intelligent job recommendation not only takes into account the individual specificity of students, but also takes into account industry development trends, and provides students with job recommendations that meet future market needs. At the same time, it considers students' personal growth and career development plans, and recommends jobs that can enhance their abilities and expand their career space, which greatly improves the matching degree of job recommendations.

[0041] Embodiment 2:

[0042] Based on the first embodiment, as shown in the accompanying drawings of the specification Figure 1 As shown, the data collection module is provided with a data collection unit and a database unit. A connection is established between the data collection unit and the online learning community and the internship recruitment platform. The data collection unit is set with trigger conditions for data collection. When new data is generated from the external data source and according to the preset time interval, the collection program is started. Different collection strategies are adopted for different types of data. For learning progress data, it is directly obtained from the database interface of the online learning platform. For internship project results data, it is obtained by interacting with the API of the internship recruitment platform. The collected data is encrypted and transmitted and stored in the database unit. The database unit uses a hash algorithm to deduplicate the data, identify and remove the same data items, and then, through data format conversion, the data from different sources are unified into a system-recognizable format. At the same time, the data collection module adopts an adaptive data collection frequency adjustment mechanism to dynamically adjust the collection frequency according to the data update speed and importance of different data sources. For online learning community data that are frequently updated and have a greater impact on employment recommendations, the data is collected every half an hour, while for relatively stable enterprise evaluation data, it is collected every two days. The real-time data update module includes a data update unit and a historical data storage unit. The data update unit is equipped with sensor technology and a mobile application interface. When students participate in offline activities, the sensor detects the activity signal and triggers the data update process. First, the activity information is converted into a storable data format. Let the actual activity duration be t1, the planned activity duration be t2, and the activity participation be P. The specific activity participation calculation formula is as follows:

[0043]

[0044] Where: σ is the activity duration fluctuation parameter obtained by statistics based on the historical data stored in the historical data storage unit;

[0045] The data update unit will transmit the calculated data transmission to the historical data storage unit through the message queue technology. When new data is generated, the data will be placed in the message queue and processed asynchronously by the background processing program to avoid the impact of data update on system performance. At the same time, the data caching technology is used to store frequently accessed data in the cache to improve the data reading speed and provide users with real-time employment recommendation services. The blockchain technology is introduced to ensure the non-tamperability and traceability of data updates. Each data update is recorded on the blockchain to form a data block containing detailed information and timestamps of the data. The authenticity and reliability of the data are guaranteed by the encryption algorithm. The real-time data update module will transmit the updated data to the data quality monitoring module for detection. The data quality monitoring module can receive data transmitted by the data acquisition module and the real-time data update module. The data quality monitoring module uses a data verification algorithm and is equipped with a deep learning-based autoencoder model for data reconstruction and anomaly detection. Suppose the input data is X={x1,x2,…,x n}, the hidden layer is represented as H after the encoding function f(X)=W1X+b1 of the autoencoder, and the reconstructed data is obtained through the decoding function g(H)=W2X+b2 Where W1 and W2 represent the weight matrices of input data and hidden layer data respectively, b1 and b2 represent the encoding bias term and encoding bias term respectively, and then calculate the reconstruction error. The specific reconstruction error calculation formula is as follows:

[0046]

[0047] Where: E represents the reconstruction error, x i represents the original input data points, Represents the data points after data reconstruction;

[0048] When the reconstruction error exceeds the threshold preset by the manager, it is judged as abnormal data. At this time, the manual review process is started, and professionals check and correct it. When the reconstruction error does not exceed the threshold preset by the manager, it is judged as normal data. The data quality monitoring module will store the normal data and the abnormal data checked and repaired by the manager in the database. The personalized analysis module can obtain the course selection history and grade distribution data of students in the database. The personalized analysis module is equipped with a deep learning algorithm to analyze the course selection history and grade distribution data of students. For the course selection history data, text analysis technology is used to extract the course theme and difficulty characteristics. For the grade distribution data, statistical analysis methods are used to calculate the mean and variance statistics. Then, a personalized ability model is constructed. The course grade factor is set as F1, the interest tag factor is set as F2, the community activity factor is set as F3, and the student comprehensive ability value is C. The student comprehensive ability value C is calculated by the formula, where d i It represents the adjustment factor calculated based on the different characteristic dimensions of students, and n is the number of characteristic dimensions;

[0049] The model is trained with a large amount of data. During the training process, the stochastic gradient descent algorithm is used to optimize the model parameters and accelerate the model convergence speed. The personalized analysis module transmits the calculation and analysis results of the model to the forward-looking recommendation module.

[0050] Specific workflow: The data collection module is connected to the external data source interface. In practical applications, an intelligent recognition and adaptation strategy can be adopted to automatically adapt the best collection strategy using intelligent recognition technology. For example, machine learning algorithms can be used to automatically identify the importance and timeliness of data, and key data can be collected first. The adaptive data collection frequency adjustment mechanism is based on machine learning algorithms and big data analysis. It can monitor changes in data sources in real time and make dynamic and intelligent adjustments based on the data update frequency, importance, and student behavior patterns. For example, online learning community data that is frequently updated and important to student ability assessment is collected every half an hour to ensure the timeliness and effectiveness of the data, providing a solid data foundation for subsequent analysis and recommendations.

[0051] Furthermore, by connecting the online learning community and internship recruitment platform through the data collection module, it is possible to obtain data such as students' academic performance, extracurricular practice, and online learning progress. At the same time, a data quality monitoring mechanism is established through the data quality monitoring module to ensure the accuracy of the acquired data. The accuracy and completeness of the data can provide a reliable basis for employment recommendations. At the same time, the real-time data update module uses sensor technology and mobile applications to track students' ability changes in a timely manner, making recommendations more in line with their actual conditions. These measures can ensure the comprehensiveness and accuracy of the data, help to fully understand students' ability status, provide a more reliable basis for employment recommendations, and help improve the accuracy of employment recommendations.

[0052] Embodiment three:

[0053] Based on the second embodiment, as shown in the accompanying drawings of the specification Figure 1 As shown in the figure, the forward-looking recommendation module combines big data analysis and industry trend forecasting technology. The forward-looking recommendation module monitors the development trends of different industries in real time and uses web crawler technology to collect data from multiple industry data sources, including industry news websites and professional forums. Then, based on historical data and current trends, the trend fitting function is used to predict future changes in talent demand. The talent demand is calculated through the trend fitting function, where T(x) is the talent demand, and A, B, ω, φ and k represent fitting parameters;

[0054] By analyzing a large amount of historical data, the values ​​of fitting parameters are determined, thereby obtaining a forecast curve for future talent demand, recommending positions that meet students' development potential, and improving students' employment success rate. The forward-looking recommendation module uses distributed storage technology to store a large amount of industry data and historical recommendation records. Through data mining technology, valuable information is extracted from these data, including industry hot positions and skill requirements. The recommendation algorithm is used to generate a personalized job recommendation list for students based on the students' personalized ability model and industry trend prediction results. The recommendation algorithm uses a combination of collaborative filtering and content recommendation. For collaborative filtering, the similarity between students is calculated, and positions are recommended to target students based on the job choices of similar students. For content recommendation, the recommendation algorithm is divided into Analyze the characteristics of the position and the ability characteristics of the students, make matching recommendations, improve the accuracy and diversity of recommendations, introduce reinforcement learning algorithms, let the system learn and optimize recommendation strategies in the continuous recommendation process, and adjust the parameters of the recommendation algorithm according to the students' feedback on the recommended positions. If the student is interested in a certain position and applies for it, the system will increase the recommendation weight of this type of position. If the student is not interested in a certain position, the system will reduce the recommendation probability of this type of position to improve the satisfaction of the recommendation. At the same time, use knowledge graph technology to build an industry knowledge graph, model the enterprises, positions and skill entities in the industry and the relationships between them, and deeply understand the industry relationships and job requirements through the knowledge graph, so as to provide richer information support for forward-looking recommendations. The interactive platform module can receive data transmitted by the forward-looking recommendation module. When students accept the recommended corporate positions, they can communicate with the companies through the interactive platform module. The interactive platform module establishes a two-way communication channel between students and employers. Students can submit questions and resumes through the platform, and employers can publish job information and interview invitations. The platform uses message push technology to push important information to both parties in a timely manner. When new job information is released and students submit resumes, the system filters out relevant information based on the user's settings and interest preferences, and then pushes the information to the corresponding users through the message push algorithm. In the message push process, an intelligent push algorithm is used to push personalized information to users based on their interests and behavioral characteristics. Based on the characteristics of their major and internship experience, we push job information in related industries to them. According to the recruitment needs of employers, we push qualified student resumes to them. We introduce virtual reality technology to provide a virtual interview environment for students and employers. During the virtual interview, we build interview scenes through 3D modeling technology. Students and employers can interact through virtual reality equipment. Students can show their works and projects, and employers can ask questions and make evaluations. The learning trajectory analysis module tracks and analyzes the students' learning process. Through data mining technology, we extract students' learning behavior characteristics, including learning time, learning frequency, and course completion progress. We use clustering algorithms to divide students into different learning types, providing a more accurate basis for personalized recommendations.In the analysis process, the association rule mining algorithm is used to discover the correlation between students' learning behaviors. Based on these correlations, personalized learning suggestions and employment recommendations are provided to students. At the same time, visualization technology is used to display the learning trajectory analysis results in the form of charts, which is convenient for students and teachers to understand the students' learning situation. The learning trajectory analysis module will transmit the analyzed student learning trajectory data to the personalized analysis module. The industry trend monitoring module regularly collects industry reports and news information, uses natural language processing technology to analyze text data, and extracts key information, including industry hot spots and technology trends. The analysis results are displayed to users through visualization technology to help students and employers The unit understands the industry dynamics. The industry trend monitoring module uses web crawler technology to collect industry information from multiple channels and adopts text classification algorithms to classify the collected text data, including industry news, technical reports and market analysis. In the process of natural language processing, the word vector technology is used to convert the text into a digital vector to facilitate subsequent analysis and processing. Through the topic model algorithm, the topic information in the text is extracted to understand the hot spots and trends of the industry. At the same time, the attention mechanism in deep learning is introduced to give higher weight to important industry information and improve the accuracy of information extraction. The industry trend monitoring module will transmit the industry trend forecast information to the forward-looking recommendation module;

[0055] Specific workflow: The learning trajectory analysis module first obtains students’ learning behavior data from the database, including learning time, learning frequency, course completion progress, etc., and uses data mining algorithms to process these data, extract students’ learning behavior characteristics, and classify students into different learning types through clustering algorithms in order to better understand students’ learning habits and characteristics. The association rule mining algorithm is used to discover the correlation between students’ learning behaviors and provide students with personalized learning suggestions. Finally, visualization technology is used to display the analysis results, so that students and teachers can intuitively understand students’ learning situations. The interactive platform module builds a communication bridge between students and employers. Students can enter questions on the platform and ask employers questions. The unit asks for job details and corporate culture, and the employer can publish job information and receive students' questions and applications through the platform. The platform uses message push technology to promptly transmit information from both parties to each other. At the same time, the platform also provides personalized career development advice and training resource recommendations, and accurately pushes them according to students' needs and characteristics. In actual application, the interactive platform module can use advanced real-time communication technology to ensure efficient and undelayed information transmission between students and employers. The platform can establish an evaluation feedback system, where students and employers can evaluate each other and provide references for other users. It uses big data to analyze the behavior patterns of students and employers, predict possible needs, and push relevant information in advance.

[0056] Furthermore, a two-way communication channel between students and employers is established through the interactive platform module, which greatly enhances the interactivity of job recommendations. On this platform, students can fully ask questions to employers and gain an in-depth understanding of the details of the position and the corporate culture, so as to better judge whether the position is in line with their career plans. Employers can also conduct interview appointments, online exchanges and other activities with students through the platform to have a more comprehensive understanding of students' abilities and potential. Through this interactive mechanism, students can better understand the needs and expectations of employers, and employers can more accurately find talents that meet their requirements, thereby significantly improving the employment matching degree and bringing convenience and benefits to both students and employers.

[0057] At the same time, the present application uses specific words to describe the embodiments of the present application. For example, "one embodiment", "an embodiment", and / or "some embodiments" refer to a certain feature, structure or characteristic related to at least one embodiment of the present application. Therefore, it should be emphasized and noted that "one embodiment" or "an embodiment" or "an alternative embodiment" mentioned twice or more in different positions in this specification does not necessarily refer to the same embodiment. In addition, certain features, structures or characteristics in one or more embodiments of the present application can be appropriately combined.

[0058] Some aspects of the present application may be performed entirely by hardware, entirely by software (including firmware, resident software, microcode, etc.), or by a combination of hardware and software. The above hardware or software may be referred to as "data blocks", "modules", "engines", "units", "components" or "systems". The processor may be one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DAPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), processors, controllers, microcontrollers, microprocessors, or combinations thereof. In addition, various aspects of the present application may be expressed as computer products located in one or more computer-readable media, which include computer-readable program codes. For example, computer-readable media may include, but are not limited to, magnetic storage devices (e.g., hard disks, floppy disks, tapes ...), optical disks (e.g., compact disks CDs, digital versatile disks DVDs ...), smart cards, and flash memory devices (e.g., cards, sticks, key drives ...).

[0059] A computer-readable medium may include a propagated data signal containing computer program code, such as in baseband or as part of a carrier wave. The propagated signal may have a variety of manifestations, including electromagnetic, optical, etc., or a suitable combination. A computer-readable medium may be any computer-readable medium other than a computer-readable storage medium, which may be connected to an instruction execution system, device or apparatus to communicate, propagate or transmit a program for use. The program code on the computer-readable medium may be propagated through any suitable medium, including radio, cable, fiber optic cable, radio frequency signal, or similar medium, or a combination of any of the above mediums.

[0060] Similarly, it should be noted that in order to simplify the description disclosed in this application, so as to help understand one or more embodiments of the invention, in the above description of the embodiments of the present application, sometimes multiple features are merged into one embodiment, drawings or descriptions thereof. However, this disclosure method does not mean that the features required by the object of the present application are more than the features mentioned in the claims. In fact, the features of the embodiment are less than all the features of the single embodiment disclosed above. In some embodiments, numbers describing the number of components and attributes are used. It should be understood that such numbers used in the description of the embodiments are modified by the modifiers "approximately", "approximately" or "substantially" in some examples. Unless otherwise specified, "approximately", "approximately" or "substantially" indicate that the number allows a variation of ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may change according to the characteristics required by individual embodiments. In some embodiments, the numerical parameters should take into account the specified significant digits and adopt the general method of retaining the digits. Although the numerical domains and parameters used to confirm the breadth of their range in some embodiments of the present application are approximate values, in specific embodiments, such numerical settings are as accurate as possible within the feasible range.

[0061] Although the present invention is disclosed as above in terms of preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art may make possible changes and modifications without departing from the spirit and scope of the present invention. Therefore, any modification, equivalent change and modification made to the above embodiments according to the technical essence of the present invention without departing from the content of the technical solution of the present invention shall fall within the scope of protection defined by the claims of the present invention.

Claims

1. A student ability analysis and employment recommendation system based on big data, Features: The system includes a data acquisition module, a data quality monitoring module, a real-time data update module, a personalized analysis module, a forward-looking recommendation module, an interactive platform module, a learning trajectory analysis module and an industry trend monitoring module; The data acquisition module acquires multi-source data from external data sources, including learning data and internship data. The real-time data update module tracks the students' latest learning and internship data through sensor technology and mobile application interface, and updates the acquired data in time. The data quality monitoring module can acquire data from the data acquisition module and the real-time data update module, and use the autoencoder model and data verification algorithm to detect anomalies in the data, calculate the reconstruction error, modify and check the abnormal data, and transmit the final normal data to the database. The personalized analysis module can extract the learning data in the database, analyze the students' course selection history and grade distribution based on the deep learning algorithm, and calculate the students' comprehensive ability value through the formula. The industry trend monitoring module It can collect industry information, predict industry trends through time series analysis technology, and transmit industry trend data to the forward-looking recommendation module. The learning trajectory analysis module extracts students' learning behavior characteristics through data mining technology, and obtains students' learning trajectory data based on data in the database, and transmits the trajectory data to the personalized analysis module. The forward-looking recommendation module obtains the prediction curve of future talent demand based on the student personalized ability model of the personalized analysis module and the industry trend prediction data of the industry trend monitoring module, and combines the trend fitting function formula to generate a personalized job recommendation list for students, and helps students and enterprises understand each other through the interactive platform module, helps students understand the recommended positions in detail, and helps enterprises screen job personnel.

2. According to the student ability analysis and employment recommendation system based on big data as described in claim 1, Features: The data acquisition module is provided with a data acquisition unit and a database unit. A connection is established between the data acquisition unit and the online learning community and the internship recruitment platform. The data acquisition unit is set with trigger conditions for data acquisition. When new data is generated from the external data source and at a preset time interval, the acquisition program is started. Different acquisition strategies are adopted for different types of data. The learning progress data is directly obtained from the database interface of the online learning platform. The internship project results data is obtained by interacting with the API of the internship recruitment platform. The collected data is encrypted and transmitted and stored in the database unit. The database unit uses a hash algorithm to deduplicate the data, identifies and removes the same data items, and then, through data format conversion, unifies the data from different sources into a format recognizable by the system. At the same time, the data acquisition module adopts an adaptive data acquisition frequency adjustment mechanism to dynamically adjust the acquisition frequency according to the data update speed and importance of different data sources.

3. According to the student ability analysis and employment recommendation system based on big data as described in claim 1, Features: The real-time data update module includes a data update unit and a historical data storage unit. The data update unit is provided with sensor technology and a mobile application interface. When students participate in offline activities, the sensor detects the activity signal and triggers the data update process. First, the activity information is converted into a storable data format. Assume that the actual activity duration is t 1 , the planned activity duration is t 2 , let activity participation be P, the specific activity participation calculation formula is as follows: Where: σ is the activity duration fluctuation parameter obtained by statistics based on the historical data stored in the historical data storage unit; The data update unit will transmit the calculated data transmission to the historical data storage unit through the message queue technology. When new data is generated, the data will be placed in the message queue and processed asynchronously by the background processing program to avoid the impact of data updates on system performance. At the same time, data caching technology is used to store frequently accessed data in the cache to improve data reading speed and provide users with real-time employment recommendation services. Blockchain technology is introduced to ensure that data updates cannot be tampered with and are traceable. Each data update is recorded on the blockchain to form a data block, which contains detailed information and timestamps of the data. The authenticity and reliability of the data are guaranteed by encryption algorithms. The real-time data update module will transmit the updated data to the data quality monitoring module for detection.

4. According to the student ability analysis and employment recommendation system based on big data as described in claim 1, Features: The data quality monitoring module can receive data transmitted by the data acquisition module and the real-time data update module. The data quality monitoring module uses a data verification algorithm and is provided with an autoencoder model based on deep learning to perform data reconstruction and anomaly detection. Assume that the input data is X={x 1 ,x 2 ,…,x n }, the encoding function f(X) = W after the autoencoder 1 X+b 1 The hidden layer is represented as H, and then through the decoding function g(H) = W 2 X+b 2 Get reconstructed data Where W 1 and W 2 Represents the weight matrix of input data and hidden layer data, b 1 and b 2 They represent the coding bias term and the coding bias term respectively, and then calculate the reconstruction error. The specific reconstruction error calculation formula is as follows: Where: E represents the reconstruction error, x i represents the original input data points, Represents the data points after data reconstruction; When the reconstruction error exceeds the threshold preset by the manager, it is judged as abnormal data. At this time, the manual review process is initiated and checked and corrected by professionals. When the reconstruction error does not exceed the threshold preset by the manager, it is judged as normal data. The data quality monitoring module will store the normal data and the abnormal data checked and repaired by the manager in the database.

5. According to the student ability analysis and employment recommendation system based on big data as described in claim 1, Features: The personalized analysis module can obtain the course selection history and grade distribution data of students in the database. The personalized analysis module is provided with a deep learning algorithm to analyze the course selection history and grade distribution data of students. For the course selection history data, text analysis technology is used to extract course topics and difficulty characteristics. For the grade distribution data, statistical analysis methods are used to calculate the mean and variance statistics. Then, a personalized ability model is constructed. The course grade factor is set as F 1 , the interest tag factor is F 2 , the community activity factor is F 3 , the student's comprehensive ability value is C, and the student's comprehensive ability value C is calculated by the formula. The specific calculation formula of the student's comprehensive ability value C is as follows: Where: d i It represents the adjustment factor calculated based on the different characteristic dimensions of students, and n is the number of characteristic dimensions; The model is trained with a large amount of data. During the training process, the stochastic gradient descent algorithm is used to optimize the model parameters and accelerate the model convergence speed. The personalized analysis module transmits the calculation and analysis results of the model to the forward-looking recommendation module.

6. A student ability analysis and employment recommendation system based on big data according to claim 1, Features: The forward-looking recommendation module combines big data analysis and industry trend forecasting technology. The forward-looking recommendation module monitors the development trends of different industries in real time and uses web crawler technology to collect data from multiple industry data sources, including industry news websites and professional forums. Then, based on historical data and current trends, a trend fitting function is used to predict changes in future talent demand. The talent demand is calculated through the trend fitting function. The specific trend fitting function is as follows: T(x)=A×sin(ωx+φ)+B×cos(ωx+φ)+k Where T(x) represents the talent demand at time x, A, B, ω, φ and k represent the fitting parameters; By analyzing a large amount of historical data, the values ​​of fitting parameters are determined, thereby obtaining a forecast curve for future talent demand, recommending positions that meet students' development potential, and improving students' employment success rate. The forward-looking recommendation module uses distributed storage technology to store a large amount of industry data and historical recommendation records. Through data mining technology, valuable information is extracted from these data, including popular positions in the industry and skill requirements. A recommendation algorithm is used to generate a personalized job recommendation list for students based on the students' personalized ability model and industry trend prediction results. The recommendation algorithm uses a combination of collaborative filtering and content recommendation. For collaborative filtering, the similarity between students is calculated, and positions are recommended to target students based on the job selections of similar students. For content recommendation, Analyze the characteristics of the positions and the ability characteristics of the students, make matching recommendations, improve the accuracy and diversity of recommendations, introduce reinforcement learning algorithms, let the system learn and optimize recommendation strategies in the continuous recommendation process, adjust the parameters of the recommendation algorithm according to the students' feedback on the recommended positions, if the students are interested in a certain position and apply for it, the system will increase the recommendation weight of this type of position, if the students are not interested in a certain position, the system will reduce the recommendation probability of this type of position, improve the satisfaction of the recommendation, at the same time, use knowledge graph technology to build an industry knowledge graph, model the enterprises, positions and skill entities in the industry and the relationships between them, and use the knowledge graph to deeply understand the industry relationships and job requirements, providing richer information support for forward-looking recommendations.

7. According to the student ability analysis and employment recommendation system based on big data as described in claim 1, Features: The interactive platform module can receive data transmitted by the forward-looking recommendation module. When students accept the recommended corporate positions, they can communicate with the companies through the interactive platform module. The interactive platform module establishes a two-way communication channel between students and employers. Students can submit questions and resumes through the platform, and employers can publish job information and interview invitations. The platform uses message push technology to push important information to both parties in a timely manner. When new job information is released and students submit resumes, the system screens out relevant information based on the user's settings and interest preferences, and then pushes the information to the corresponding users through the message push algorithm. During the message push process, an intelligent push algorithm is used to push personalized information to users based on the user's interests and behavioral characteristics. According to the student's major and internship experience characteristics, job information in related industries is pushed to them. According to the recruitment needs of the employer, qualified student resumes are pushed to them. Virtual reality technology is introduced to provide a virtual interview environment for students and employers. During the virtual interview process, an interview scene is constructed through three-dimensional modeling technology. Students and employers can interact through virtual reality equipment. Students can display their works and projects, and employers can ask questions and make evaluations.

8. According to the big data-based student ability analysis and employment recommendation system of claim 1, Features: The learning trajectory analysis module tracks and analyzes the students' learning process, extracts the students' learning behavior characteristics, including learning time, learning frequency, and course completion progress, through data mining technology, and divides students into different learning types using clustering algorithms to provide a more accurate basis for personalized recommendations. In the analysis process, the association rule mining algorithm is used to discover the correlation between students' learning behaviors. Based on these correlations, personalized learning suggestions and employment recommendations are provided to students. At the same time, visualization technology is used to display the learning trajectory analysis results in the form of charts, which is convenient for students and teachers to understand the students' learning situation. The learning trajectory analysis module will transmit the analyzed student learning trajectory data to the personalized analysis module.

9. According to the big data-based student ability analysis and employment recommendation system of claim 1, Features: The industry trend monitoring module regularly collects industry reports and news information, uses natural language processing technology to analyze text data, and extracts key information, including industry hot spots and technology trends. The analysis results are displayed to users through visualization technology to help students and employers understand industry dynamics. The industry trend monitoring module uses web crawler technology to collect industry information from multiple channels, and uses text classification algorithms to classify the collected text data, including industry news, technical reports and market analysis. In the natural language processing process, word vector technology is used to convert text into digital vectors to facilitate subsequent analysis and processing. The topic model algorithm is used to extract the topic information in the text to understand the hot spots and trends of the industry. At the same time, the attention mechanism in deep learning is introduced to give higher weight to important industry information and improve the accuracy of information extraction. The industry trend monitoring module will transmit industry trend forecast information to the forward-looking recommendation module.

Images