Teaching big data visual analysis system and method

Abstract

The invention discloses a teaching big data visualization analysis system and method, and belongs to the technical field of education information, and the system comprises an analysis data collection module which is used for collecting student IDs, student data, tasks and scores, and storing the ID, the data, the tasks and the scores in a database; the preprocessing module is used for carrying out data cleaning and abnormal data processing; the comprehensive comparison module is used for acquiring student IDs needing to be analyzed; the server is also used for obtaining all historical scores according to student IDs and tasks to form a longitudinal comparison report; and the server is also used for obtaining the student ID in a non-database and a data set with the same task or the same student data according to the student data and the task corresponding to the student ID, and comparing the student data, the task and the score corresponding to the student ID needing to be analyzed with the data set to form a transverse comparison report corresponding to the student ID. According to the teaching big data visual analysis system and method, the problem that an existing teaching analysis method cannot comprehensively reflect the real situation of students is solved.

Description

Technical Field

[0001] The present invention relates to the field of educational information technology, and in particular to a teaching big data visualization analysis system and method. Background Art

[0002] With the rapid development of society and industry, the demand for high-quality skilled personnel is growing. As the main places for training such talents, project bases, technical schools and vocational schools bear important educational missions. However, in the traditional teaching model, teachers mainly rely on experience to teach, lack of accurate analysis and scientific feedback of learning data, which limits the quality and efficiency of skilled personnel training.

[0003] Teaching analysis is the interpretation and analysis of the massive data generated by teachers and students during the teaching process, in order to evaluate the students' training or competition situations and discover the regular problems of students. When analyzing students' training or competition behaviors, the existing teaching analysis methods generally only analyze the students' own situations, which leads to insufficient depth in mining students' learning data and cannot fully reflect the students' real situation. Summary of the invention

[0004] In order to overcome the defects of the prior art, an object of the present invention is to provide a teaching big data visualization analysis system to solve the above-mentioned problems.

[0005] In order to overcome the defects of the prior art, another object of the present invention is to provide a method for visual analysis of teaching big data to solve the above-mentioned problems.

[0006] The technical solution adopted by the present invention to solve the technical problem is: a teaching big data visualization analysis system, comprising:

[0007] The analysis data collection module is used to collect student ID, student data, tasks and grades, form a relationship network among the four, and then store it in the database; the student data includes age, learning stage and competition stage, the tasks are divided into training and competition, and the grades are divided into training grades and competition grades;

[0008] The preprocessing module is used to perform data standardization on the data in the database, and then perform data cleaning and abnormal data processing to obtain standardized student data, tasks and grades;

[0009] The comprehensive comparison module is used to obtain the student ID that needs to be analyzed; it is also used to obtain all historical scores based on the student ID and tasks to form a vertical comparison report; it is also used to obtain a student ID that is not in the database and has a data collection with the same task or the same student data based on the student data and tasks corresponding to the student ID, and compare the student data, tasks and scores corresponding to the student ID that needs to be analyzed with the data collection to form a horizontal comparison report corresponding to the student ID.

[0010] Preferably, it also includes: a task performance and ranking statistics module, which is used to compare the performance corresponding to multiple student IDs of each task to form a task performance and ranking report.

[0011] It is worth mentioning that it also includes: a personal ability assessment module, which is used to count the scores of all tasks of each student ID and form a personal ability assessment report.

[0012] Preferably, it also includes: the preprocessing module performs data standardization processing on the data in the database in a manner including unifying data formats, data types and data naming specifications.

[0013] Optionally, when performing abnormal data processing, the preprocessing module is used to obtain the numerical scores corresponding to the student ID to form a score array;

[0014] It is also used to obtain the median of numerical scores and to set the threshold constant a for outliers;

[0015] It is also used to calculate the absolute deviation b of each numerical score from the median, and then obtain the median deviation MAD;

[0016] It is also used to calculate the threshold of outliers c = a*MAD;

[0017] It is also used to extract numerical results with absolute deviation b and abnormal value threshold c that are large into the cause analysis library.

[0018] A method for visual analysis of teaching big data includes the following steps:

[0019] S1: Collect student ID, student data, tasks and scores, form a relationship network among them, and then store it in the database; student data includes age, learning stage and competition stage, tasks are divided into training and competition, and scores are divided into training scores and competition scores;

[0020] S2: After data standardization is performed on the data in the database, data cleaning and abnormal data processing are performed to obtain standardized student data, tasks and grades;

[0021] S3: Obtain the student ID that needs to be analyzed; also used to obtain all historical scores based on the student ID and tasks to form a vertical comparison report; obtain a student ID that is not in the database and a data collection with the same task or the same student data based on the student data and tasks corresponding to the student ID, and compare the student data, tasks and scores corresponding to the student ID that needs to be analyzed with the data collection to form a horizontal comparison report corresponding to the student ID.

[0022] It is worth noting that the step S3 also includes: comparing the scores corresponding to multiple student IDs of each task to form a task score and ranking report.

[0023] Specifically, the step S3 also includes: collecting statistics on the scores of all tasks of each student ID to form a personal ability assessment report.

[0024] Preferably, the data in the database in step S2 is standardized in a manner that includes standardizing data formats, data types, and data naming specifications.

[0025] Optionally, the process of abnormal data processing in step S2 is:

[0026] Get the numerical scores corresponding to the student ID to form a score array;

[0027] Get the median of the numerical scores and set the threshold constant a for abnormal values;

[0028] Calculate the absolute deviation b between each numerical score and the median, and then get the median deviation MAD;

[0029] Calculate the outlier threshold c = a*MAD;

[0030] The numerical results with the absolute deviation b and the outlier threshold c being larger are extracted and stored in the cause analysis database.

[0031] The beneficial effect of the present invention is that: in the teaching big data visualization analysis system, by forming a longitudinal comparison report, the student's performance progress in each module and skill point in different learning stages and the time taken to complete the questions are deeply analyzed. This time series-based analysis method not only helps to judge whether the student studies seriously and effectively improves himself, but also accurately identifies whether there are modules or skill points that are always difficult to master. For the problems found, personalized training reinforcement can be arranged to ensure that each student can steadily improve at a pace that suits him. In addition, the student's usual training results can be compared with the formal assessment test results, and the gap between the two can be analyzed to determine whether there are psychological barriers such as tension and anxiety that affect performance. This analysis not only focuses on the student's knowledge mastery, but also pays more attention to the adjustment of his psychological state, which helps to timely discover and solve potential psychological problems and ensure that students can perform at their best in formal assessments. By forming a horizontal comparison report, the ability level is deeply analyzed from the following three dimensions: the training results at the same age, the training results at the same learning stage, and the competition results at the same competition stage. This comparison not only helps to estimate the student's ability level and training potential, but also provides a scientific basis for the construction of personalized learning paths. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 A system block diagram of a teaching big data visualization analysis system in one embodiment of the present invention;

[0033] Figure 2 It is an intuitive comparison chart of target students' data in various dimensions in one embodiment of the present invention;

[0034] Figure 3 A comparison chart of the ability of a target student and outstanding contestants of previous years in one embodiment of the present invention;

[0035] Figure 4 A skill radar chart in one embodiment of the present invention;

[0036] Figure 5 An analysis chart of a single skill point corresponding to a skill radar chart in one embodiment of the present invention;

[0037] Figure 6 To play a stability curve in one embodiment of the present invention;

[0038] Figure 7 A score detail dimension diagram in one embodiment of the present invention;

[0039] Figure 8 This is a page for adjusting node parameters in a parameterized model in an embodiment of the present invention. DETAILED DESCRIPTION

[0040] The specific embodiments of the present invention are 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. 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.

[0041] like Figure 1-8 As shown, a teaching big data visualization analysis system includes:

[0042] The analysis data collection module is used to collect student IDs, student data, tasks and grades, form a relationship network among the four, and then store it in the database; the student data includes age, learning stage and competition stage, the tasks are divided into training and competition, and the grades are divided into training grades and competition grades; after the relationship network is formed, the student data and tasks can be queried through the student ID, and the corresponding grades can be queried through the student ID and tasks;

[0043] A preprocessing module is used to perform data standardization on the data in the database, and then perform data cleaning and abnormal data processing to obtain standardized student data, tasks and grades; in this embodiment, the preprocessing module is used to delete duplicate records in the student data, tasks and grades to perform data cleaning to avoid data redundancy affecting the analysis results;

[0044] The comprehensive comparison module is used to obtain the student ID that needs to be analyzed; it is also used to obtain all historical scores based on the student ID and tasks to form a vertical comparison report; it is also used to obtain a student ID that is not in the database and has a data collection with the same task or the same student data based on the student data and tasks corresponding to the student ID, and compare the student data, tasks and scores corresponding to the student ID that needs to be analyzed with the data collection to form a horizontal comparison report corresponding to the student ID.

[0045] In the teaching big data visualization analysis system, by forming a longitudinal comparison report, the student's performance progress in each module and skill point in different learning stages and the time taken to complete the questions are deeply analyzed. This time series-based analysis method not only helps to judge whether the student studies seriously and effectively improves himself, but also accurately identifies whether there are modules or skill points that are always difficult to master. For the problems found, personalized training reinforcement can be arranged to ensure that each student can steadily improve at a pace that suits him. In addition, the student's usual training results can be compared with the formal assessment test results to analyze the gap between the two, and it can be determined whether there are psychological barriers such as tension and anxiety that affect performance. This analysis not only focuses on the student's knowledge mastery, but also pays more attention to the adjustment of his mental state, which helps to timely discover and solve potential psychological problems and ensure that students perform at their best in formal assessments. By forming a horizontal comparison report, the ability level is deeply analyzed from the following three dimensions: the training results at the same age, the training results at the same learning stage, and the competition results at the same competition stage. This comparison not only helps to estimate the student's ability level and training potential, but also provides a scientific basis for the construction of personalized learning paths.

[0046] The process of forming a longitudinal comparison report is as follows: by comparing all the results of students from the time they participated in the training to the present, all the training results are divided according to the different learning stages of the students (such as the learning stage from time A to time B is the first round of learning, the learning stage from time C to time D is the second round of learning, and so on. These learning stages will be sorted according to time, and the results of the learning stages can be sorted according to the time order of the learning stages to form historical results), and according to the content of the training results (such as training modules and skill points), analysis is carried out. Through this comprehensive longitudinal comparative analysis, not only can personalized learning paths be provided for students, but also a scientific basis can be provided for educational decision-making, and continuous improvement of educational practice can be promoted.

[0047] The data collection with the same tasks or the same student data and the student ID other than in the database includes the training / competition results of domestic bases, college students and WorldSkills award-winning contestants built into the system or shared by other users. Figure 2 For intuitive comparison of data in each dimension, Figure 3 Comparing the abilities of the current selected students with those of previous outstanding contestants, it can be seen that the gap is not large, and in some aspects they even surpass the previous outstanding contestants, showing a relatively high training value; this shows that through scientific horizontal comparative analysis, it is possible to accurately identify students' potential strengths and weaknesses, thereby tailoring more targeted learning plans for them, helping them to achieve all-round development in their future academic and career paths.

[0048] The scores of multiple students are selected from multiple tasks for comparison. By comparing the scores of students corresponding to the tasks, a comprehensive analysis report is formed in the vertical direction (students' own growth trajectory) and the horizontal direction (comparison with other domestic students and World Skills award-winning contestants). The comparison is used to explore the relationship between the contestant's performance and age, skill mastery, training time, training scenarios, etc.

[0049] It is worth mentioning that it also includes: task score and ranking statistics module, which is used to compare the scores corresponding to multiple student IDs of each task to form a task score and ranking report. Compare the scores of multiple students in a single task, view the ability distribution of all players from a comprehensive perspective, and understand the ability of each player from all aspects of details.

[0050] Optionally, it also includes: a personal ability assessment module, which is used to collect statistics on the results of all tasks of each student ID and form a personal ability assessment report. By measuring the ability indexes such as the student's training or competition results, mastery, and performance, the student's own growth trajectory and personal comprehensive ability index are formed, the player's ability weaknesses are clarified, and a scientific basis is provided for the formulation of an effective training plan, so as to carry out targeted task arrangements and intensive training.

[0051] The process of forming a personal ability assessment report is: by superimposing the scores of multiple trainings collected in chronological order, the scores are converted into score rates, so as to more scientifically and fairly assess the students' learning outcomes, generate skill radar charts and performance stability curves, so as to accurately assess the students' mastery and development trends in various modules and skill points in a data-driven manner. The gradual increase in score rates and the gradual decrease in curve fluctuations intuitively reflect the gradual mastery of students' abilities.

[0052] For example: collect multiple training results, superimpose the score rate of each skill point of the student in chronological order, and obtain the skill radar chart of this player. Figure 4 The skill radar chart in the figure shows that: the student's skill point "3 Package Design" is unstable and not proficient, especially in the formal assessment, and needs targeted training; the student's skill points "17 Fault Evidence" and "12 Peripheral Driver" are stable and proficient; the student's skill point "10 Welding Quality" is steadily improving in repeated training. From the skill radar chart, we know that the student's skill point "3 Package Design" is the worst. Figure 5Analysis of the problem tracking section: phenomenon, this student's skill point 3 score used to be full marks, but it began to decline gradually after the "November end-of-term baseline assessment" task began; analysis of the conclusion, comparison of the skill point 3 scores of other students in the same training batch; if other students' scores are declining, consider whether the difficulty of the test questions associated with the current skill point is increasing, and all students need to strengthen their in-depth understanding of skill point 3; if other students' scores have not declined, and only this student's skill point 3 score is declining, consider whether the current student's in-depth understanding of skill point 3 is not enough, and the grasp of the underlying knowledge base needs to be deepened; if other students' scores have not declined, and this student's scores for multiple skill points are declining, consider whether this student has physical or physiological discomfort, and needs to adjust his or her status to the best in time.

[0053] Figure 6 The performance stability curve in the figure reveals that a student's mastery of skill points 2, work organization and management, and 3, planning and communication is unstable and fluctuates greatly, sometimes full marks and sometimes zero marks, which indicates that these skill points need further training. Skill point 1, partition wall construction accuracy, has a large fluctuation, but its original score is not low, and its score has been steadily rising after the fluctuation, and it is predicted that the score of this skill point will show an upward trend. Combined analysis shows that the student should focus on strengthening the training of skill points with large fluctuations or a downward trend. This analysis method not only provides students with personalized learning paths, but also provides teachers with precise teaching guidance to ensure that students can fully develop in each skill point. Through continuous monitoring and optimization of training plans, we are committed to cultivating students' comprehensive abilities and laying a solid foundation for their future career development and academic achievements.

[0054] Problem Tracking Section: Through the combination of parameterized models and key task indicators (KTM) algorithms, we can accurately screen out the skills that students have not yet mastered, and present them through intuitive color labels to track the training progress of these skills throughout the process until students have mastered all skills. Figure 5 As shown in the figure, the project covers 18 skill points, of which 1 skill point is marked as R to indicate that it has not been mastered, and 5 skill points are marked as Y to indicate that they are not proficient enough. Based on these data, the training plan can be dynamically adjusted to tailor targeted intensive training programs for students to help them make up for their shortcomings and achieve all-round development. This is not only a precise shaping of students' individual abilities, but also a key measure to promote the personalized and intelligent transformation of education, laying a solid foundation for cultivating high-quality talents that can adapt to future development.

[0055] Specifically, it also includes: the preprocessing module performs data standardization processing on the data in the database in a manner including data format unification, data type unification and data naming specification unification. In this embodiment, data standardization processing is performed in the following manners: data format unification (e.g., grade ranking, score, and scoring rate are all displayed in numerical form), data type unification (e.g., grade ranking is a positive integer, score and scoring rate are decimals with decimal points), data naming specification unification (e.g., camel case nomenclature "scoringRate" is used in Java language, and database fields use snake case nomenclature with underscores between words "scoring_rate");

[0056] Preferably, when performing abnormal data processing, the preprocessing module is used to obtain the numerical scores corresponding to the student ID to form a score array;

[0057] It is also used to obtain the median of numerical scores and to set the threshold constant a for outliers;

[0058] It is also used to calculate the absolute deviation b of each numerical score from the median, and then obtain the median deviation MAD;

[0059] It is also used to calculate the threshold of outliers c = a*MAD;

[0060] It is also used to extract numerical scores with absolute deviation b and abnormal value threshold c that are large to the cause analysis library to confirm whether the abnormal data is due to collection errors or abnormal performance of the players, and then correct the erroneous data in time to ensure data accuracy.

[0061] In this embodiment, the numerical grade is a score or a ranking. When performing abnormal data processing, it will first be selected whether to process the score or the ranking. If the score is selected, the system will obtain the historical score corresponding to the student ID. If the ranking is selected, the system will obtain the historical ranking corresponding to the student ID. In this embodiment, for different tasks, there will be different scores and rankings. The scores and rankings of different tasks should be counted independently, so before performing abnormal data processing, it is also necessary to select the corresponding task first. For example, for a certain training, the rankings in the numerical grade of the historical training of a certain student ID are 2, 3, 6, 3, 5, 8 and 4, the median of the ranking is 4, the threshold constant c of the abnormal value is set to 3, the absolute deviations of the individual rankings and the species are 2, 1, 2, 1, 1, 4, 0, respectively, the median deviation MAD is 1, the threshold c of the abnormal value is c = a * MAD = 3 * 1 = 3, where the absolute deviation is 4, which is greater than the threshold c = 3 of the abnormal value, so the data ranked 8 will be extracted to the cause analysis library, and the data in the cause analysis library will be analyzed later.

[0062] In addition, according to the business scenarios and understanding, a complete and highly compatible multi-level data collection and storage structure is designed, such as the evaluation details record level, the summary level by evaluation indicators, the summary level by module, the summary level by sub-module, the summary level by skill points, and the summary level by total score. Therefore, after data processing, according to the multi-level data collection and storage structure designed in the data standardization process, the collected data from different sources and different structures are converted into a unified data structure that can be analyzed in the system and stored in a unified location. Finally, by setting up the API and microservice framework, several major sections such as data collection, data analysis, and resource file classification and precipitation belong to different microservices. Each microservice has its own database to ensure the independence and flexibility of the service, and use distributed transactions to ensure data consistency. A unified API interface standard is defined, including input parameters, output format, and corresponding format. The API gateway is used as a unified entry point to receive requests and forward them to the corresponding microservices.

[0063] In this solution, before executing the comprehensive comparison module, task performance and ranking statistics module, and personal ability assessment module, the data preprocessed by the preprocessing module will be firstly subjected to a variety of statistical methods, including but not limited to the mean, range, standard deviation, and the combination of parameterized models and key task indicators (KTM, for screening and tracking of skills that do not meet the standards). Through multi-dimensional (score details, performance, mastery, etc.) detailed analysis, the overall ability of students can be accurately assessed, providing scientific teaching decision support for teachers:

[0064] 1. Scoring rate, average, and range: Figure 7 Based on the score detail dimension, the highest score, lowest score and average score of each module and skill point in each training are calculated based on the training results of each dimension (module, skill point). The range is calculated by the difference between the highest score and the lowest score, so as to intuitively understand the level differences of this group of students in different skills. At the same time, the average reflects the overall performance level of the students in each module and skill point. In addition, by comparing the score rate of the currently selected student with the set target score rate, the gap between the student and the final training goal can be clearly identified. Through these methods, the weak points of the ability of this group of students can be accurately located so that teachers can adjust the students' training plans in time.

[0065] 2. Standard Deviation: It is used to measure the degree of fluctuation of students' scores in each module and skill point, so as to evaluate the stability of their performance. Through statistical formulas, from the dimension of performance, the standard deviation of each module and skill point is calculated in chronological order using the results of students' multiple trainings. The larger the value, the greater the fluctuation of scores and the more unstable the skill mastery. It is necessary to strengthen the training of related skills to give full play to students' potential.

[0066] 3. Combining parameterized models with key task indicators (KTM): used to screen and track skills that do not meet the standards. By building a parameterized model with parameters, users can adjust the parameters of each node to meet different requirements (such as Figure 8 ),in Figure 1 The "standard value" in the above refers to the target score rate that students need to achieve in various skill training. The model uses different marks to intuitively reflect the students' achievement of key skill indicators under various requirements ( Figure 5 and 8 ): G means met, mastered, Y means nearly met, not proficient, and R means not met, not mastered. This visualization method makes it easy to quickly identify which skills students need further improvement, thereby promoting targeted tracking and improvement.

[0067] A method for visual analysis of teaching big data includes the following steps:

[0068] S1: Collect student ID, student data, tasks and scores, form a relationship network among them, and then store it in the database; student data includes age, learning stage and competition stage, tasks are divided into training and competition, and scores are divided into training scores and competition scores;

[0069] S2: After data standardization is performed on the data in the database, data cleaning and abnormal data processing are performed to obtain standardized student data, tasks and grades;

[0070] S3: Obtain the student ID that needs to be analyzed; also used to obtain all historical scores based on the student ID and tasks to form a vertical comparison report; obtain a student ID that is not in the database and a data collection with the same task or the same student data based on the student data and tasks corresponding to the student ID, and compare the student data, tasks and scores corresponding to the student ID that needs to be analyzed with the data collection to form a horizontal comparison report corresponding to the student ID.

[0071] Preferably, the step S3 further comprises: comparing the scores corresponding to the multiple student IDs of each task to form a task score and ranking report.

[0072] Optionally, step S3 further includes: collecting statistics on the scores of all tasks of each student ID to form a personal ability assessment report.

[0073] Specifically, the data in the database in step S2 is standardized in a manner that includes standardizing data formats, data types, and data naming specifications.

[0074] It is worth noting that the process of abnormal data processing in step S2 is:

[0075] Get the numerical scores corresponding to the student ID to form a score array;

[0076] Get the median of the numerical scores and set the threshold constant a for abnormal values;

[0077] Calculate the absolute deviation b between each numerical score and the median, and then get the median deviation MAD;

[0078] Calculate the outlier threshold c = a*MAD;

[0079] The numerical results with the absolute deviation b and the outlier threshold c being larger are extracted and stored in the cause analysis database.

[0080] In this solution, all analyzable data in the training process are recorded in real time through personnel collection, data collection, result collection, data storage and system architecture. Personnel collection is to group and classify according to different identities, resumes and achievements, build a talent echelon, and assist in talent selection and training. Data collection is to use the system to upload and archive the data of the training task package by type, which is convenient for data review and rapid construction of the training task environment. Result collection is to use the system to quickly and in real time enter the scoring process online on the computer or mobile terminal, and can also directly import the scoring data in multiple dimensions according to evaluation indicators, modules, and skill points to ensure comprehensive data collection. Data storage is to persist the collected data using a relational database to support the storage and efficient retrieval of large-scale data. The system architecture adopts a distributed architecture to improve the scalability of the system.

[0081] Most of the training data in the existing system cannot be effectively recorded, the training materials and task resources are scattered, the practical training experience of previous students is not systematically managed, lacks orderly organization and accumulation, and the efficiency of access is low. This solution comprehensively collects training data, systematically manages training materials and results, dynamically tracks and analyzes data, and forms the growth trajectory of students. This not only provides reference and summary for subsequent training, but also gradually builds a scientific training system that suits itself, improves training efficiency and quality, and realizes the effective use and inheritance of data.

[0082] The existing system lacks effective tools to conduct a comprehensive analysis of students' performance, making it difficult for teachers to accurately grasp students' strengths and weaknesses, limiting the implementation of teaching students in accordance with their aptitude. This solution introduces a variety of statistical methods, including standard deviation, median absolute deviation (MAD, for abnormal data detection) and parameterized models combined with key task indicators (KTM, for screening and tracking of skills that do not meet the standards). Through multi-dimensional detailed analysis, the overall ability of students can be accurately assessed, providing teachers with scientific teaching decision support.

[0083] Existing teaching data is mostly displayed in the form of tables or simple charts, which makes it difficult for teachers to quickly understand the data and students' learning situation. This solution provides an intuitive and highly interactive visualization interface, combining data and graphic display to help teachers quickly gain insight into teaching data, discover its complex relationships and potential problems, and improve teaching efficiency.

[0084] The existing system only processes data on the surface and cannot provide valuable suggestions for improving teaching. This solution uses multi-dimensional in-depth data analysis, both vertically (students’ own growth trajectory) and horizontally (compared with other domestic contestants and World Skills winners), to accurately mine the gaps, potential problems and trends in the data, and provide a scientific basis for decision-making for improving students’ abilities and improving teaching.

[0085] The existing system lacks real-time and interactivity, and cannot dynamically adjust teaching strategies based on teaching data. This solution realizes real-time analysis and dynamic adjustment of teaching data, optimizes the teaching process, specifies personalized teaching plans, and improves teaching effects.

[0086] The embodiments of the present invention are described in detail above with reference to the accompanying drawings, but the present invention is not limited to the described embodiments. For those skilled in the art, various changes, modifications, substitutions and variations of these embodiments are made without departing from the principles and spirit of the present invention, and still fall within the scope of protection of the present invention.

Claims

1. A teaching big data visualization analysis system, characterized in that: include: The analysis data collection module is used to collect student IDs, student data, tasks and grades, form a relationship network among the four, and then store it in the database; The student data includes age, learning stage and competition stage, the tasks are divided into training and competition, and the results are divided into training results and competition results; The preprocessing module is used to perform data standardization on the data in the database, and then perform data cleaning and abnormal data processing to obtain standardized student data, tasks and grades; The comprehensive comparison module is used to obtain the student ID that needs to be analyzed; it is also used to obtain all historical scores based on the student ID and tasks to form a vertical comparison report; it is also used to obtain a student ID that is not in the database and has a data collection with the same task or the same student data based on the student data and tasks corresponding to the student ID, and compare the student data, tasks and scores corresponding to the student ID that needs to be analyzed with the data collection to form a horizontal comparison report corresponding to the student ID.

2. A teaching big data visualization analysis system according to claim 1, characterized in that: Also includes: The task score and ranking statistics module is used to compare the scores corresponding to multiple student IDs of each task to form a task score and ranking report.

3. A teaching big data visualization analysis system according to claim 1, characterized in that: Also includes: The personal ability assessment module is used to compile statistics on the scores of all tasks for each student ID and form a personal ability assessment report.

4. The teaching big data visualization analysis system according to claim 1 is characterized by: Also includes: The preprocessing module performs data standardization processing on the data in the database in a manner including data format unification, data type unification and data naming specification unification.

5. The teaching big data visualization analysis system according to claim 1 is characterized by: When performing abnormal data processing, the preprocessing module is used to obtain the numerical scores corresponding to the student ID to form a score array; It is also used to obtain the median of numerical scores and to set the threshold constant a for outliers; It is also used to calculate the absolute deviation b of each numerical score from the median, and then obtain the median deviation MAD; It is also used to calculate the threshold of outliers c = a*MAD; It is also used to extract numerical results with absolute deviation b and abnormal value threshold c that are large into the cause analysis library.

6. A method for visual analysis of teaching big data, characterized in that: The following steps are involved: S1: Collect student ID, student data, tasks and grades, form a relationship network among them, and then store it in the database; The student data includes age, learning stage and competition stage, tasks are divided into training and competition, and results are divided into training results and competition results; S2: After data standardization is performed on the data in the database, data cleaning and abnormal data processing are performed to obtain standardized student data, tasks and grades; S3: Obtain the student ID that needs to be analyzed; also used to obtain all historical scores based on the student ID and tasks to form a vertical comparison report; obtain a student ID that is not in the database and a data collection with the same task or the same student data based on the student data and tasks corresponding to the student ID, and compare the student data, tasks and scores corresponding to the student ID that needs to be analyzed with the data collection to form a horizontal comparison report corresponding to the student ID.

7. A method for visual analysis of teaching big data according to claim 6, characterized in that: The step S3 also includes: comparing the scores corresponding to multiple student IDs of each task to form a task score and ranking report.

8. A method for visual analysis of teaching big data according to claim 6, characterized in that: The step S3 also includes: collecting statistics on the scores of all tasks of each student ID to form a personal ability assessment report.

9. The method for visual analysis of teaching big data according to claim 6, characterized in that: The data standardization process for the data in the database in step S2 includes standardization of data formats, data types and data naming specifications.

10. The method for visual analysis of teaching big data according to claim 6, characterized in that: The process of abnormal data processing in step S2 is as follows: Get the numerical scores corresponding to the student ID to form a score array; Get the median of the numerical scores and set the threshold constant a for abnormal values; Calculate the absolute deviation b between each numerical score and the median, and then get the median deviation MAD; Calculate the outlier threshold c = a*MAD; The numerical results with the largest absolute deviation b and the largest outlier threshold c are extracted and stored in the cause analysis database.

Images