Training project demand analysis and real-time feedback system

Abstract

The invention discloses a training project demand analysis and real-time feedback system, and relates to the technical field of data processing, and the system comprises the steps: obtaining training demand information inputted by a user, and obtaining user information inputted by the user; according to the training demand information and the user basic information, training depth analysis optimization is carried out on the training items, and the optimal training depth is obtained; training resource demand information of the optimal training depth is obtained, the optimal training depth is adjusted and analyzed in combination with user resource information, and the adjusted training depth is obtained; training scheme analysis and optimization of the training project are carried out according to the user resource information in combination with adjustment of the training depth, and an optimal training scheme is obtained; and generating a training project according to the adjusted training depth and the optimal training scheme, and feeding back to the user. The technical problems that dynamic adjustment cannot be performed based on real-time user information and personalized recommendation ability is lacked in the prior art, so that training items cannot meet actual requirements of users are solved, and the technical effect of improving training efficiency is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a training project demand analysis and real-time feedback system. Background Art

[0002] With the rapid development of information technology, traditional training methods are gradually shifting towards intelligent and personalized directions. Enterprises and educational institutions are paying more and more attention to how to effectively design and implement efficient training programs based on the actual needs, existing capabilities and resource constraints of trainees. Traditional training program design usually relies on human experience and general course planning, and it is difficult to tailor it to the individual needs of each trainee.

[0003] For example, in CN118365485A, the prior art relies on a label model library for course matching, which makes it difficult to accurately respond to changes in students' career goals or needs during the learning process. Summary of the invention

[0004] The present application provides a training program demand analysis and real-time feedback system, which is used to solve the technical problems that the existing technology cannot be dynamically adjusted based on real-time user information and lacks personalized recommendation capabilities, resulting in the training program failing to meet the actual needs of users.

[0005] In view of the above problems, this application provides a training project needs analysis and real-time feedback system.

[0006] In a first aspect of the present application, a training project needs analysis and real-time feedback system is provided, the system comprising: An information acquisition module, wherein the information acquisition module acquires training demand information input by a user, and acquires user information input by a user, wherein the training demand information includes training target information, and the user information includes user basic information and user resource information; a training analysis and optimization module, wherein the training analysis and optimization module performs training depth analysis and optimization of training projects according to the training demand information and user basic information to obtain an optimal training depth; an adjustment analysis module, wherein the adjustment analysis module acquires training resource demand information of the optimal training depth, and performs adjustment analysis on the optimal training depth in combination with the user resource information to obtain an adjusted training depth; a program analysis and optimization module, wherein the program analysis and optimization module performs training program analysis and optimization of training projects according to the user resource information and the adjusted training depth to obtain an optimal training program; a training program generation module, wherein the training program generation module generates a training program according to the adjusted training depth and the optimal training program, and feeds back to the user.

[0007] A second aspect of the present application provides a training project needs analysis and real-time feedback method, the method comprising: Acquire training demand information input by a user, and acquire user information input by a user, wherein the training demand information includes training target information, and the user information includes user basic information and user resource information; perform training depth analysis and optimization on the training project based on the training demand information and user basic information to obtain an optimal training depth; acquire training resource demand information for the optimal training depth, and perform adjustment analysis on the optimal training depth in combination with the user resource information to obtain an adjusted training depth; perform training plan analysis and optimization on the training project based on the user resource information and the adjusted training depth to obtain an optimal training plan; generate a training project based on the adjusted training depth and the optimal training plan, and provide feedback to the user.

[0008] The third aspect of the present application provides an electronic device, comprising: a memory for storing executable instructions; a processor for executing the executable instructions stored in the memory to implement a training project demand analysis and real-time feedback system provided by the present application.

[0009] The technical solution provided in this application has at least the following technical effects or advantages: The present application obtains training demand information input by a user, and obtains user information input by a user, wherein the training demand information includes training target information, and the user information includes user basic information and user resource information; according to the training demand information and user basic information, the training depth analysis and optimization of the training project is performed to obtain the optimal training depth; the training resource demand information of the optimal training depth is obtained, and the optimal training depth is adjusted and analyzed in combination with the user resource information to obtain the adjusted training depth; according to the user resource information, in combination with the adjusted training depth, the training program analysis and optimization of the training project is performed to obtain the optimal training program; according to the adjusted training depth and the optimal training program, a training project is generated and fed back to the user. The present invention solves the technical problem that the prior art cannot be dynamically adjusted based on real-time user information, lacks personalized recommendation capabilities, and causes the training project to fail to meet the actual needs of the user. By obtaining the user's training demand information and resource information in real time, horizontal content optimization and vertical learning progress analysis are performed, and depth and program analysis and prediction are performed. According to the analysis results and the matching of the adaptive optimization module, the training program control parameters are generated, and the training program control parameters are fed back to the training management system, so as to realize the generation and dynamic adjustment of intelligent training projects, optimize the training content and arrangement, and achieve the technical effect of improving training efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0011] Figure 1 A schematic diagram of the structure of a training project demand analysis and real-time feedback system provided in an embodiment of the present application; Figure 2 A flowchart of a training project needs analysis and real-time feedback method provided in an embodiment of the present application; Figure 3 This is a schematic diagram of the structure of an exemplary electronic device of the present application.

[0012] Explanation of the reference numerals: information acquisition module 11 , training analysis and optimization module 12 , adjustment analysis module 13 , program analysis and optimization module 14 , training program generation module 15 , bus 300 , receiver 301 , processor 302 , transmitter 303 , memory 304 , bus interface 305 . DETAILED DESCRIPTION

[0013] The present application provides a training project demand analysis and real-time feedback system to solve the technical problems that the existing technology cannot make dynamic adjustments based on real-time user information and lacks personalized recommendation capabilities, resulting in the inability of training projects to meet the actual needs of users. By obtaining users' training demand information and resource information in real time, horizontal content optimization and vertical learning progress analysis are performed, and in-depth and program analysis and prediction are performed. According to the analysis results and matching with the adaptive optimization module, training program control parameters are generated and fed back to the training management system, so as to realize intelligent training project generation and dynamic adjustment, optimize training content and arrangements, and achieve the technical effect of improving training efficiency.

[0014] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0015] It should be noted that any variations of the terms "include" and "have" are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules that are not explicitly listed or inherent to these processes, methods, products or devices.

[0016] Embodiment 1, as Figure 1 As shown, this application provides a training project needs analysis and real-time feedback system for performing the following Figure 2 A training project demand analysis and real-time feedback method is shown, the system comprising: The information acquisition module 11 acquires the training demand information input by the user, and acquires the user information input by the user, wherein the training demand information includes the training target information, and the user information includes the user basic information and the user resource information.

[0017] In the embodiment of the present application, user input collection is implemented through a form input interface. This is the initial interaction point between the user and the system. The user provides their training demand information and user information by filling out a form. The training demand information includes training target information, that is, the skills and knowledge that the user hopes to master through training. User information includes user basic information and user resource information.

[0018] Next, basic user information is collected, also through a form. In this step, users fill in their current skill and knowledge levels in the form. For example, enter the knowledge and skills they currently have in the form. Then the system requires users to enter user resource information, that is, the time resources and cost resources they can invest. This step is still completed through form input, and users fill in the amount of time and budget they can invest in training. For example, users choose the number of hours of training per week, or directly enter the amount they are willing to pay for training. Through this process, basic user information and user resource information are obtained.

[0019] The training demand information and user information input by the user are obtained through the above process.

[0020] Furthermore, the information acquisition module 11 in the system provided in the application embodiment is also used for: The training skill target information and training knowledge target information input by the user are obtained as training demand information; the basic knowledge information and basic skill information input by the user are obtained as user basic information; the time resource information and cost resource information input by the user are obtained as user resource information, and combined with the user basic information as user information.

[0021] In the embodiment of the present application, the training skill target information and training knowledge target information input by the user are first obtained, which is used as the user's training demand information. This step is implemented through a form input interface, and the user fills in their specific goals in the interface. For example, the user can enter the skills they hope to master through training, such as "improving programming skills" or "learning project management", and the knowledge they expect to acquire, such as "deep understanding of algorithm design" or "familiarity with industry standards."

[0022] Next, users are asked to enter their current basic knowledge information and basic skills information, which constitutes the user's basic information. Still through the form input, users fill in the knowledge and skills they have mastered. For example, if the user is participating in programming training, they can enter their current programming level, such as "understanding basic syntax" and "able to write simple programs." After this, the time resource information and cost resource information entered by the user are obtained, which constitute the user resource information. Users enter the amount of time they can invest and the training cost they can afford through the form.

[0023] Finally, the user basic information and user resource information are combined as complete user information and stored in the system database.

[0024] The training analysis and optimization module 12 performs training depth analysis and optimization of the training project according to the training demand information and user basic information to obtain the optimal training depth.

[0025] In an embodiment of the present application, the training analysis and optimization module first performs a preliminary training content analysis based on the user's training demand information and user basic information. Through keyword matching or extracting training content corresponding to the user's goal from a preset content library, a training list containing core content and extended content is generated. Then, based on the training content, combined with the user's current basic information, multiple training depth options are generated. These depth options divide the training content into different levels, depending on the proportion of core training content and comprehensive training content. Create a depth space containing multiple possible training depth combinations. The core training content is the basic knowledge that must be taught at all training depths, and the comprehensive training content is additional professional content used to expand the user's knowledge breadth. For example, for software development engineers, the core training content is Java basic syntax, object-oriented programming, and exception handling. Comprehensive training content includes Java advanced features (such as generics, annotations, multi-threaded programming, etc.), Java Web development (such as Servlet, JSP, Spring framework) and Java performance optimization (such as JVM tuning, garbage collection mechanism), etc. These contents are comprehensive training content because they provide additional and more professional knowledge.

[0026] The depth fitness function is then used to evaluate the training depth options. On this basis, the training depth is optimized in the depth space through an iterative optimization algorithm, such as gradient descent or random search. Starting from multiple initial depth combinations, the training content is gradually adjusted and the fitness is continuously calculated until the depth combination with the largest fitness is found. This combination is used as the optimal training depth.

[0027] Furthermore, the training analysis and optimization module 12 in the system provided in the application embodiment is also used for: According to the training demand information, the total training content is obtained by indexing, and the total training content includes core training content and comprehensive training content; according to the core training content and the comprehensive training content in different proportions, multiple training depths are obtained, and the training depth space is set; with the goal of maximizing the training effect and minimizing the training difficulty, a training depth function is constructed, as shown in the following formula:

[0028] In the embodiment of the present application, firstly, according to the training demand information input by the user, that is, the skills and knowledge that the user expects to learn, the corresponding total training content is extracted from the pre-built training content database using a keyword matching algorithm or natural language processing technology. The database presets a variety of training content modules according to different training objectives, and the modules include core training content and comprehensive training content.

[0029] After extracting the core and comprehensive training content, a combination algorithm is used to generate multiple different training depth options. The combination algorithm arranges and combines the core training content and the comprehensive training content in different proportions to generate multiple training depths. Each depth option represents the breadth and depth of the training content. The greater the depth, the more comprehensive content is included and the richer the knowledge learned.

[0030] Next, a training depth space is constructed by combining different training depth options generated by the algorithm. The training depth space is a collection of different depth combinations, representing all the paths that users may experience during training. This space is constructed based on a multidimensional matrix representation, in which each dimension corresponds to a different level of training content. The rationality and effectiveness of each training depth are then evaluated using a predefined depth fitness function. The training depth function is calculated based on two core factors: training effect and training difficulty, and ultimately outputs a fitness value that reflects the suitability of the training depth for the user. Among them, the training depth function contains two weights, w1 and w2, which are pre-set by technical experts based on the importance of training depth and training difficulty, such as 0.6 and 0.4.

[0031] Based on the depth fitness function, an optimization algorithm such as gradient descent or genetic algorithm is used in the training depth space to continuously iterate and find the training depth combination with the highest fitness. Starting from the initial training depth, the depth options are continuously adjusted by evaluating the fitness value of each depth combination until the optimal solution is found. Finally, the optimal training depth is output, which not only meets the user's training needs, but also ensures that the training difficulty is moderate and the effect is maximized.

[0032] Furthermore, the training analysis and optimization module 12 in the system provided in the application embodiment is also used for: In the training depth space, a first training depth is randomly selected; a training difficulty is predicted according to the first training depth combined with the user basic information to obtain first training difficulty information; a first depth fitness is calculated based on the training depth function according to the first training difficulty information and the first training depth; the training depth is continuously optimized in the training depth space until convergence, and the training depth with the maximum depth fitness is output to obtain the optimal training depth.

[0033] In the embodiment of the present application, firstly, an initial point is randomly selected in the constructed training depth space, and this point is called the first training depth. Then, combined with the basic information of the user, the difficulty of the randomly selected first training depth is predicted, and the training difficulty prediction is based on the machine learning model to obtain the first training difficulty information of the first training depth through the prediction.

[0034] Then, according to the first training depth and the corresponding first training difficulty information, a depth fitness function is used to calculate the fitness value of the depth, thereby obtaining a first depth fitness.

[0035] Once the first depth fitness is calculated, continue to explore other depth combinations in the training depth space. By using iterative optimization algorithms, such as gradient descent, genetic algorithms, etc., the training depth is gradually adjusted in the depth space to evaluate the fitness of each new depth combination. At each iteration, the depth options are updated and a new fitness value is calculated. This process continues until the fitness value reaches a convergence state, that is, there is no significant improvement. When all iterations are completed in the training depth space and the fitness value reaches the maximum and converges, the optimal training depth is output.

[0036] Furthermore, the training analysis and optimization module 12 in the system provided in the application embodiment is also used for: Based on historical training data of similar total training content, a sample user basic information set and a sample training depth set are collected, and the sample training difficulty information set is annotated according to the learning test results of different sample training depths for different sample user basic information; the sample user basic information set, the sample training depth set and the sample training difficulty information set are used as supervised training data to train and obtain a training difficulty predictor; the training difficulty predictor is used to predict the training difficulty of the user basic information and the first training depth to obtain the first training difficulty information.

[0037] In the embodiment of the present application, a sample user basic information set and a sample training depth set are first collected from the historical training data of the same total training content using SQL query statements. These data include the user's basic knowledge level, skill level, and their learning records at different training depths, such as the complexity of the learning content and the knowledge covered.

[0038] Then, through the rule-based conditional judgment algorithm, the learning performance of different sample users at each training depth is labeled according to the user's learning performance, such as test scores, task completion status, etc. Automatic labeling is performed according to the set labeling rules. For example, if the test score is less than 60% and the task completion rate is less than 50%, it is labeled as "high difficulty", and the difficulty score is considered to be 3; if the test score is between 60%-80% and the task completion rate is between 50%-80%, it is labeled as "medium difficulty", and the difficulty score is considered to be 2; if the test score is higher than 80% and the task completion rate exceeds 80%, it is labeled as "low difficulty", and the difficulty score is considered to be 1. According to these rules, the user data is automatically traversed, and a specific difficulty label is assigned to each user's performance to generate a sample training difficulty information set.

[0039] Then, using data processing libraries and data cleaning techniques, the sample user basic information set, sample training depth set, and sample training difficulty information set are integrated together to form a complete supervised training data set. On this basis, a supervised learning algorithm such as decision tree or random forest is used to train a training difficulty predictor. The input features are user basic information and training depth, and the output label is the labeled difficulty value. By repeatedly adjusting the parameters of the model, the prediction results are as close as possible to the actual historical labeled values. The training process evaluates the generalization ability of the model through cross-validation to ensure that the model can accurately predict difficulty even on unseen data. After the model training is completed, the model is verified using cross-validation technology. By dividing the data set into a training set and a validation set, the performance of the model on different data sets is measured, and the prediction accuracy of the model is evaluated based on evaluation indicators such as mean square error MSE and R² value. Model verification ensures the reliability of the predictor in actual scenarios.

[0040] Finally, in practical applications, the basic information of the current user and the randomly selected first training depth are input into the trained training difficulty predictor. The model generates the first training difficulty information based on the input features.

[0041] The adjustment analysis module 13 obtains the training resource demand information of the optimal training depth, and performs adjustment analysis on the optimal training depth in combination with the user resource information to obtain the adjusted training depth.

[0042] In the embodiment of the present application, firstly, according to the optimal training depth, the training resource demand information related to the depth is extracted from the training content library. This information includes the time, budget, equipment resources, etc. required for training. Then, the user's resource information is obtained, including the time resources and cost resources that the user can invest in the training.

[0043] Next, we analyze whether the user's resources can meet the requirements of the optimal training depth. We compare the user's resource information with the resource requirements of the optimal training depth and calculate the resource deviation between the two. Similarly, we compare the user's budget with the cost requirements of training to determine whether there is a shortage of resources.

[0044] After identifying the resource deviation, the optimal training depth is adjusted and analyzed. This step is usually implemented through a linear programming algorithm or a heuristic adjustment algorithm. The training depth is adjusted according to the user's resource constraints, such as time and budget, to ensure that the user completes the training within the available resources. Finally, the adjusted training depth is generated and output to obtain the adjusted training depth.

[0045] Furthermore, the adjustment analysis module 13 in the system provided in the application embodiment is also used for: According to the optimal training depth, the training resource requirement information is mapped in the training resource requirement table, wherein the training resource requirement table includes a mapping relationship between sample training depths and sample training resource requirement information of the same type of total training content, and the training resource requirement information includes time resource requirement information and cost resource requirement information; the resource deviation between the cost resource information in the user resource information and the cost resource requirement information is calculated; according to the resource deviation, the adjusted training depth is adjusted and calculated to obtain the adjusted training depth.

[0046] In an embodiment of the present application, first, a SQL query statement or an API interface is used to extract training resource demand information from a database. According to the determined optimal training depth, the resource demand information of the relevant comprehensive training content is obtained from the training resource demand table of the database. The training resource demand table stores mapping information of different training depths and training content, covering the time resource demand information and cost resource demand information required for the comprehensive content. Through this process, the time and cost demand information corresponding to the comprehensive training content is extracted from the database, that is, the training resource demand information is obtained.

[0047] Next, the resource deviation is calculated through mathematical operations and logical comparisons. User resource information is obtained from the user's input, including the time resources the user is willing to invest (such as the learning time the user can invest per week, 30 hours) and cost resources (such as the user's training budget, 200 yuan). This user resource information is compared with the training resource demand information obtained in the previous step (such as 50 hours and 300 yuan) to calculate the resource deviation. Specifically, when calculating the time resource deviation, the difference between the time required for comprehensive training content and the time the user can invest is calculated by subtraction. If the comprehensive content requires 50 hours, but the user can only invest 30 hours, the time resource deviation is calculated to be 20 hours. When calculating the cost resource deviation, the system also calculates the difference between the cost requirement of the comprehensive content and the user's budget by subtraction. If the comprehensive content requires 300 yuan, and the user's budget is 200 yuan, the budget deviation is calculated to be 100 yuan.

[0048] Based on these resource deviations, the comprehensive training content is adjusted and calculated, while the core training content remains unchanged. A heuristic adjustment algorithm is used to optimize the comprehensive content. The algorithm will give priority to reducing or simplifying low-priority content modules, such as removing scalability modules with higher time or cost, or simplifying uncritical comprehensive training content. The adjustment rule is to gradually reduce the duration or cost of the comprehensive content according to the resource deviation until these resource requirements match the user's resource information. For example, if the user can only invest 30 hours of time and a budget of 200 yuan, adjust the comprehensive content to reduce its total duration to 30 hours and its total cost to 200 yuan.

[0049] Optionally, the ratio between the user resource information and the training resource demand information is calculated as an adjustment coefficient to adjust the optimal training depth, that is, the adjustment coefficient is multiplied by the optimal training depth to obtain the adjusted training depth. When calculating the adjustment coefficient, the ratio of the time resource information in the user resource information to the time resource demand information is calculated, and the ratio of the cost resource information in the user resource information to the cost resource demand information is calculated, and the average of the two ratios is calculated as the adjustment coefficient.

[0050] Finally, after the adjustment is completed, a new adjusted training depth is generated to obtain the adjusted training depth.

[0051] The program analysis and optimization module 14 performs training program analysis and optimization of the training project according to the user resource information and the adjustment of the training depth to obtain the optimal training program.

[0052] Furthermore, the solution analysis and optimization module 14 in the system provided in the application embodiment is also used for: According to the training content corresponding to the adjustment of the training depth, a training program space is constructed, wherein each training program includes different training times and different numbers of training contents in a single training; with the purpose of maximizing training efficiency and training efficiency, a training program function is constructed as follows:

[0053] Among them, TPF is the fitness of the solution, w 3 and w 4 is the weight, T t T is the training time for training according to the training plan, y is the preset training time, G t G is the training effect information predicted according to the training plan combined with the basic information of the user. y The training effect information is preset; according to the training program function, the training program is optimized in the training program space to obtain the optimal training program.

[0054] In an embodiment of the present application, firstly, a plurality of different training scheme combinations are constructed by data matrix representation or multidimensional array modeling. According to the adjusted training depth, the training content related to the depth is extracted, and a plurality of feasible training schemes are generated according to different combinations. Each scheme defines different training times, the duration of a single training and the specific content covered. The different combinations of these schemes are represented by multidimensional arrays, for example, the dimension of each scheme can represent the frequency of training, the depth of content, the distribution of total time, etc. These data are stored in the form of a matrix, and a complete training scheme space is provided for subsequent evaluation and optimization.

[0055] Then, with the goal of maximizing training efficiency and training efficiency, a training plan function is constructed, in which w 3 and w 4 These weights are customized according to user needs to adapt to different users' preferences for time and effect.

[0056] The scheme fitness function is then used to evaluate the effectiveness of each training scheme. The fitness of each scheme is evaluated in the scheme space and its TPF value is calculated. Prioritize schemes with high fitness, that is, those whose training time is close to the time the user can invest and whose training effect is close to the expected goal. For example, if a scheme can achieve 80% of the expected effect within the time allowed by the user, its fitness value is set to high as a potential candidate for the optimal scheme. According to the calculation results of the fitness function, the scheme is optimized through multiple iterations until the fitness value of the scheme reaches a convergence state. In each iteration, certain parameters in the scheme are adjusted, such as the frequency of training or the depth of a single training content, until an optimal combination is found so that the scheme can maximize the training effect and meet the user's resource constraints at the same time.

[0057] When all iterations are completed, the optimal training plan is finally output.

[0058] Furthermore, the solution analysis and optimization module 14 in the system provided in the application embodiment is also used for: In the training program space, a first training program is randomly generated; a first training time of the first training program is obtained, and a training effect is predicted in combination with the user basic information to obtain first training effect information, wherein a sample training program set, a sample user basic information set and a sample training effect information set are collected in the historical training record data of the same training content, and a training effect predictor is trained to perform prediction; based on the training program function, the fitness of the first program is calculated according to the first training time and the first training effect information; a second training program is continuously randomly generated, and the fitness of the second program is obtained by processing and calculating; whether the fitness of the second program is greater than the fitness of the first program is determined, and if so, the second training program is retained, and if not, the first training times and the second training times of the first training program and the second training program are obtained, and the ratio of the second training times to the first training times is calculated as the retention probability, and the second training program or the first training program is retained according to the retention probability; optimization is continued in the training program space until convergence, and the training program finally retained is output to obtain the optimal training program.

[0059] In an embodiment of the present application, a random sampling algorithm is first used to generate an initial solution from the solution space. Specifically, a first training solution is randomly generated from a pre-constructed multidimensional training solution space by random sampling technology. The solution defines multiple parameters, including the number of trainings, the duration of each training, and the specific training content covered. The generation of the solution is based on the random selection of all possible combinations to ensure that different solutions can be explored. Through this random generation, potential solutions are found in a large-scale solution space.

[0060] Then use machine learning models, such as regression models, random forests, etc. to predict the training effect. The specific steps are to extract the total training time of the generated first training program, which is the time required to complete all the training content calculated based on the program, such as accumulating all training times and the duration of each training. Then use historical training data to predict the training effect. By extracting a sample training program set, a sample user basic information set, and a sample training effect information set from historical data of the same or similar training content, and inputting these data into a machine learning model for training, such as using a random forest or linear regression model. After the training is completed, the model is used to predict the effect of the current user's basic information and the generated first training program, and the predicted first training effect information is output, such as expected test scores, task completion rate, etc.

[0061] After that, the fitness of the first scheme is calculated by the training scheme function to obtain the fitness of the first scheme. Then, the second training scheme is generated by the same random sampling algorithm. This scheme is similar to the first scheme. The training time of the scheme is extracted, and the training effect is predicted by the machine learning model in combination with the basic information of the user. Then, the fitness value of the second scheme is calculated using the same fitness function to ensure that all generated schemes are evaluated using consistent standards. Before retaining the second training scheme, the retention probability calculation and the random number generator are used to decide whether to retain the new scheme. First, the fitness values ​​of the first and second schemes are compared. If the fitness of the second scheme is higher than that of the first scheme, the second scheme is retained directly. Otherwise, the retention probability of the two is calculated, and the calculation formula is the ratio of the training times of the two schemes. Then, a random number generator is used to generate a random number between 0 and 1. If the random number is less than the retention probability, the second scheme is retained; otherwise, the first scheme is retained. In this way, the selection of the scheme is both random and closely related to its fitness. When the retention probability is greater than 1, the second scheme must be retained; when the retention probability is less than 1, whether to retain the second scheme is randomly determined according to the probability.

[0062] Continue to optimize in the solution space until the fitness value of the solution reaches the convergence condition. In this process, a genetic algorithm or a heuristic search algorithm is used. Specifically, multiple new solutions are generated through iteration, and the fitness calculation and retention process of the solution is repeated. In each iteration, a new solution is generated, and the optimal solution is retained based on the retention probability and fitness calculation. When the solution optimization reaches the convergence state, the optimal training solution is output. The optimal solution contains all optimized training parameters, including the best training frequency, duration, content coverage, and expected effect.

[0063] The training project generation module 15 generates a training project according to the adjusted training depth and the optimal training plan, and feeds back to the user.

[0064] In the embodiment of the present application, the adjusted training depth and optimal training plan are first obtained, and the relevant information is extracted from the previous module using data extraction technology to ensure that the core content is retained and the overall content is adjusted to suit the user's time and resource constraints. Then, the training content is divided into multiple tasks through the task generation algorithm, and the time schedule and frequency of each task are matched with the optimization plan. The training project is generated through this process and finally fed back to the user through the visualization tool.

[0065] In the embodiments of the present application, in summary, the embodiments of the present application have at least the following technical effects: The present application obtains training demand information input by a user, and obtains user information input by a user, wherein the training demand information includes training target information, and the user information includes user basic information and user resource information; according to the training demand information and user basic information, the training depth analysis and optimization of the training project is performed to obtain the optimal training depth; the training resource demand information of the optimal training depth is obtained, and the optimal training depth is adjusted and analyzed in combination with the user resource information to obtain the adjusted training depth; according to the user resource information, in combination with the adjusted training depth, the training program analysis and optimization of the training project is performed to obtain the optimal training program; according to the adjusted training depth and the optimal training program, a training project is generated and fed back to the user. The present invention solves the technical problem that the prior art cannot be dynamically adjusted based on real-time user information, lacks personalized recommendation capabilities, and causes the training project to fail to meet the actual needs of the user. By obtaining the user's training demand information and resource information in real time, horizontal content optimization and vertical learning progress analysis are performed, and depth and program analysis and prediction are performed. According to the analysis results and the matching of the adaptive optimization module, the training program control parameters are generated, and the training program control parameters are fed back to the training management system, so as to realize the generation and dynamic adjustment of intelligent training projects, optimize the training content and arrangement, and achieve the technical effect of improving training efficiency.

[0066] Embodiment 2, based on the same inventive concept as a training program demand analysis and real-time feedback system in the above embodiment, Figure 2 As shown, the embodiment of the present application provides a training project demand analysis and real-time feedback method, the method comprising: Acquire training demand information input by a user, and acquire user information input by a user, wherein the training demand information includes training target information, and the user information includes user basic information and user resource information; perform training depth analysis and optimization on the training project based on the training demand information and user basic information to obtain an optimal training depth; acquire training resource demand information for the optimal training depth, and perform adjustment analysis on the optimal training depth in combination with the user resource information to obtain an adjusted training depth; perform training plan analysis and optimization on the training project based on the user resource information and the adjusted training depth to obtain an optimal training plan; generate a training project based on the adjusted training depth and the optimal training plan, and provide feedback to the user.

[0067] Furthermore, obtaining training demand information input by the user and obtaining user information input by the user include: The training skill target information and training knowledge target information input by the user are obtained as training demand information; the basic knowledge information and basic skill information input by the user are obtained as user basic information; the time resource information and cost resource information input by the user are obtained as user resource information, and combined with the user basic information as user information.

[0068] Further, based on the training demand information and user basic information, the training depth analysis and optimization of the training project is performed to obtain the optimal training depth, including: According to the training demand information, the total training content is obtained by indexing, and the total training content includes core training content and comprehensive training content; according to the core training content and the comprehensive training content in different proportions, multiple training depths are obtained, and the training depth space is set; with the goal of maximizing the training effect and minimizing the training difficulty, a training depth function is constructed, as shown in the following formula:

[0069] Among them, TDF is the depth fitness, w 1 and w 2 is the weight, D t is the training depth, K t The training difficulty information is obtained by combining and analyzing the training depth and the user basic information; according to the training depth function, the training depth of the training project is optimized in the training depth space to obtain the optimal training depth.

[0070] Further, according to the training depth function, in the training depth space, the training depth optimization of the training project is performed, including: In the training depth space, a first training depth is randomly selected; a training difficulty is predicted according to the first training depth combined with the user basic information to obtain first training difficulty information; a first depth fitness is calculated based on the training depth function according to the first training difficulty information and the first training depth; the training depth is continuously optimized in the training depth space until convergence, and the training depth with the maximum depth fitness is output to obtain the optimal training depth.

[0071] Further, performing training difficulty prediction according to the first training depth combined with the user basic information includes: Based on historical training data of similar total training content, a sample user basic information set and a sample training depth set are collected, and the sample training difficulty information set is annotated according to the learning test results of different sample training depths for different sample user basic information; the sample user basic information set, the sample training depth set and the sample training difficulty information set are used as supervised training data to train and obtain a training difficulty predictor; the training difficulty predictor is used to predict the training difficulty of the user basic information and the first training depth to obtain the first training difficulty information.

[0072] Further, obtaining the training resource demand information of the optimal training depth, combining the user resource information, adjusting and analyzing the optimal training depth to obtain the adjusted training depth, including: According to the optimal training depth, the training resource requirement information is mapped in the training resource requirement table, wherein the training resource requirement table includes a mapping relationship between sample training depths and sample training resource requirement information of the same type of total training content, and the training resource requirement information includes time resource requirement information and cost resource requirement information; the resource deviation between the cost resource information in the user resource information and the cost resource requirement information is calculated; according to the resource deviation, the adjusted training depth is adjusted and calculated to obtain the adjusted training depth.

[0073] Further, according to the user resource information and in combination with the adjustment of the training depth, the training program analysis and optimization of the training project are performed, including: According to the training content corresponding to the adjustment of the training depth, a training program space is constructed, wherein each training program includes different training times and different numbers of training contents in a single training; with the purpose of maximizing training efficiency and training efficiency, a training program function is constructed as follows:

[0074] Among them, TPF is the fitness of the solution, w 3 and w 4 is the weight, T t T is the training time for training according to the training plan, y is the preset training time, G t G is the training effect information predicted according to the training plan combined with the basic information of the user. y The training effect information is preset; according to the training program function, the training program is optimized in the training program space to obtain the optimal training program.

[0075] Further, according to the training program function, optimizing the training program in the training program space includes: In the training program space, a first training program is randomly generated; a first training time of the first training program is obtained, and a training effect is predicted in combination with the user basic information to obtain first training effect information, wherein a sample training program set, a sample user basic information set and a sample training effect information set are collected in the historical training record data of the same training content, and a training effect predictor is trained to perform prediction; based on the training program function, the fitness of the first program is calculated according to the first training time and the first training effect information; a second training program is continuously randomly generated, and the fitness of the second program is obtained by processing and calculating; whether the fitness of the second program is greater than the fitness of the first program is determined, and if so, the second training program is retained, and if not, the first training times and the second training times of the first training program and the second training program are obtained, and the ratio of the second training times to the first training times is calculated as the retention probability, and the second training program or the first training program is retained according to the retention probability; optimization is continued in the training program space until convergence, and the training program finally retained is output to obtain the optimal training program.

[0076] Embodiment three, based on the inventive concept of a training project demand analysis and real-time feedback system in the aforementioned embodiment, the present application also provides an electronic device, comprising: at least one processor; a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the steps of any one of the methods described in the aforementioned embodiment one.

[0077] Figure 3 This is a schematic diagram of the structure of an exemplary electronic device of this application. Figure 3 In the embodiment, the bus architecture is represented by bus 300, which may include any number of interconnected buses and bridges, and bus 300 connects various circuits including one or more processors represented by processor 302 and memory represented by memory 304. Bus 300 may also connect various other circuits such as peripherals, voltage regulators, and power management circuits, which are well known in the art and are therefore not further described herein. Bus interface 305 provides an interface between bus 300 and receiver 301 and transmitter 303. Receiver 301 and transmitter 303 may be the same element, namely a transceiver, providing a unit for communicating with various other devices over a transmission medium. Processor 302 is responsible for managing bus 300 and general processing, while memory 304 may be used to store data used by processor 302 when performing operations.

[0078] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. The processes depicted in the accompanying drawings do not necessarily require the specific order and continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0079] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0080] This specification and drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A training project demand analysis and real-time feedback system, characterized in that: The system comprises: An information acquisition module, wherein the information acquisition module acquires training demand information input by a user, and acquires user information input by a user, wherein the training demand information includes training target information, and the user information includes user basic information and user resource information; A training analysis and optimization module, wherein the training analysis and optimization module performs training depth analysis and optimization of the training project according to the training demand information and user basic information to obtain an optimal training depth; An adjustment analysis module, wherein the adjustment analysis module obtains the training resource demand information of the optimal training depth, and performs adjustment analysis on the optimal training depth in combination with the user resource information to obtain an adjusted training depth; A program analysis and optimization module, wherein the program analysis and optimization module analyzes and optimizes the training program of the training project according to the user resource information and the adjustment of the training depth to obtain an optimal training program; A training project generation module generates a training project according to the adjusted training depth and the optimal training plan, and feeds back to the user.

2. The training project demand analysis and real-time feedback system according to claim 1 is characterized in that: Obtain training demand information entered by the user, and obtain user information entered by the user, including: Acquire the training skill target information and training knowledge target information input by the user as training demand information; Obtain basic knowledge information and basic skill information input by the user as basic user information; The time resource information and cost resource information input by the user are obtained as user resource information, and combined with the user basic information as user information.

3. The training project demand analysis and real-time feedback system according to claim 1 is characterized in that: According to the training demand information and user basic information, the training depth analysis and optimization of the training project are carried out to obtain the optimal training depth, including: According to the training demand information, the total training content is obtained by indexing, and the total training content includes core training content and comprehensive training content; According to the core training content and the comprehensive training content in different proportions, multiple training depths are obtained, and a training depth space is set; With the goal of maximizing the training effect and minimizing the training difficulty, a training depth function is constructed as follows: ； Among them, TDF is the depth fitness, w1 and w2 are weights, D t is the training depth, K t Training difficulty information obtained by combining training depth and user basic information; According to the training depth function, the training depth of the training project is optimized in the training depth space to obtain the optimal training depth.

4. The training project demand analysis and real-time feedback system according to claim 3 is characterized in that: According to the training depth function, in the training depth space, the training depth optimization of the training project is performed, including: In the training depth space, randomly selecting a first training depth; Predicting the training difficulty according to the first training depth and the user basic information to obtain first training difficulty information; According to the first training difficulty information and the first training depth, based on the training depth function, a first depth fitness is calculated; Continue to optimize the training depth in the training depth space until convergence, output the training depth with the maximum depth fitness, and obtain the optimal training depth.

5. The training project demand analysis and real-time feedback system according to claim 4 is characterized in that: Predicting the training difficulty according to the first training depth and the user basic information includes: Based on the historical training data of the same type of total training content, a sample user basic information set and a sample training depth set are collected, and the learning test results of different sample training depths based on different sample user basic information are annotated to obtain a sample training difficulty information set; Using the sample user basic information set, the sample training depth set and the sample training difficulty information set as supervised training data, training to obtain a training difficulty predictor; The training difficulty predictor is used to predict the training difficulty of the user basic information and the first training depth to obtain first training difficulty information.

6. The training project demand analysis and real-time feedback system according to claim 1 is characterized in that: Acquiring training resource demand information of the optimal training depth, combining the user resource information, adjusting and analyzing the optimal training depth, and obtaining the adjusted training depth, including: According to the optimal training depth, training resource requirement information is mapped in a training resource requirement table, wherein the training resource requirement table includes a mapping relationship between sample training depths and sample training resource requirement information of the same type of total training content, and the training resource requirement information includes time resource requirement information and cost resource requirement information; Calculating a resource deviation between the cost resource information in the user resource information and the cost resource demand information; The adjusted training depth is adjusted and calculated according to the resource deviation to obtain the adjusted training depth.

7. The training project demand analysis and real-time feedback system according to claim 1 is characterized in that: According to the user resource information and in combination with the adjustment of the training depth, the training program analysis and optimization of the training project are performed, including: According to the training content corresponding to the adjusted training depth, a training program space is constructed, wherein each training program includes a different number of training times and a different number of training contents in a single training; With the purpose of maximizing training efficiency and training efficiency, a training plan function is constructed as follows: ； Among them, TPF is the fitness of the solution, w3 and w4 are weights, T t T is the training time for training according to the training plan, y is the preset training time, G t G is the training effect information predicted according to the training plan combined with the basic information of the user. y To preset training effect information; According to the training program function, the training program is optimized in the training program space to obtain the optimal training program.

8. The training project demand analysis and real-time feedback system according to claim 7 is characterized in that: According to the training scheme function, optimizing the training scheme in the training scheme space includes: Randomly generate a first training scheme in the training scheme space; Obtaining a first training time of the first training program, and performing training effect prediction in combination with the user basic information to obtain first training effect information, wherein a sample training program set, a sample user basic information set, and a sample training effect information set are collected in historical training record data of the same training content, and a training effect predictor is trained to perform prediction; Based on the training program function, and according to the first training time and the first training effect information, a first program fitness is calculated; Continue to randomly generate the second training plan, and process and calculate to obtain the fitness of the second plan; Determine whether the fitness of the second program is greater than the fitness of the first program; if so, retain the second training program; if not, obtain the first training times and the second training times of the first training program and the second training program, and calculate the ratio of the second training times to the first training times as the retention probability, and retain the second training program or the first training program according to the retention probability; Continue to optimize in the training scheme space until convergence, output the final retained training scheme, and obtain the optimal training scheme.

9. A training project needs analysis and real-time feedback method, characterized in that: The method is performed by a training project demand analysis and real-time feedback system according to any one of claims 1 to 8, comprising: Acquire training demand information input by a user, and acquire user information input by a user, wherein the training demand information includes training target information, and the user information includes user basic information and user resource information; According to the training demand information and user basic information, perform training depth analysis and optimization of the training project to obtain the optimal training depth; Acquire the training resource demand information of the optimal training depth, and adjust and analyze the optimal training depth in combination with the user resource information to obtain the adjusted training depth; According to the user resource information and in combination with the adjustment of the training depth, the training program of the training project is analyzed and optimized to obtain the optimal training program; The training depth and the optimal training plan are adjusted as described, a training project is generated, and feedback is given to the user.

10. An electronic device, characterized in that: The electronic device comprises: A memory for storing executable instructions; The processor is used to implement a training project demand analysis and real-time feedback system according to any one of claims 1 to 8 when executing the executable instructions stored in the memory.

Images