Intelligent recommendation course selection learning method and system based on multi-dimensional analysis

Abstract

The invention discloses an intelligent recommendation course selection learning method and system based on multi-dimensional analysis, and relates to the technical field of intelligent recommendation, and the method comprises the steps: obtaining the multi-dimensional information of a user; the multi-dimensional information comprises user basic information, historical behavior data, learning related state information and constraint condition configuration information; intelligently identifying a recommendation scene type based on the multi-dimensional information; the recommendation scene type comprises a resource selection recommendation scene and a learning task recommendation scene; according to the recommendation scene type, executing a corresponding recommendation analysis process to obtain a recommendation result; performing constraint condition verification and dynamic optimization adjustment on the recommendation result to obtain an optimized recommendation result; and integrating the optimized recommendation results, and generating and outputting a personalized recommendation scheme. The method and the device can improve the precision and the efficiency of recommended course selection, and improve the learning efficiency and the use experience of the user.

Description

Technical Field

[0001] This application relates to the field of intelligent recommendation technology, and in particular to an intelligent recommendation course selection and learning method and system based on multi-dimensional analysis. Background Technology

[0002] With the rapid development of digital education and the increasing abundance of online learning resources, users' demand for personalized recommendations is becoming more urgent. However, existing recommendation systems suffer from the following shortcomings in practical applications: Traditional recommendation systems often rely on a single recommendation algorithm, lacking comprehensive consideration of users' multi-dimensional characteristics, resulting in low personalization and accuracy of recommendations, making it difficult to match users' specific needs; when dealing with complex constraints, existing recommendation systems often perform poorly, failing to effectively balance multiple constraints such as learning time requirements and resource type limitations, leading to low accuracy of recommendations; simultaneously, traditional recommendation systems often employ a serial processing approach, resulting in low processing efficiency and significant response delays when facing large-scale data, leading to low efficiency in course selection recommendations and impacting user experience; furthermore, traditional recommendation systems lack effective scene recognition capabilities, cannot dynamically adjust recommendation strategies based on the user's stage, and have deficiencies in learning task recommendation and information display methods, lacking effective tracking and personalized guidance of user learning progress, and displaying information that is overly redundant or cluttered, further reducing user learning efficiency and user experience.

[0003] Therefore, there is an urgent need for a method and system that can integrate multi-dimensional information, efficiently process complex constraints, intelligently identify scenarios, and provide personalized recommendations and optimized displays to improve the accuracy and efficiency of course selection recommendations, thereby enhancing users' learning efficiency and user experience, and solving the problems existing in current technologies. Summary of the Invention

[0004] The purpose of this application is to provide an intelligent course selection recommendation method and system based on multi-dimensional analysis, which can improve the accuracy and efficiency of course selection recommendations, and enhance users' learning efficiency and user experience.

[0005] To achieve the above objectives, this application provides the following solution.

[0006] In a first aspect, this application provides an intelligent course selection and learning method based on multi-dimensional analysis, which includes the following steps.

[0007] Obtain multi-dimensional information about the user; the multi-dimensional information includes basic user information, historical behavior data, learning-related status information, and constraint configuration information.

[0008] Based on the multi-dimensional information, the recommendation scenario type is intelligently identified; the recommendation scenario type includes resource selection recommendation scenario and learning task recommendation scenario.

[0009] Based on the recommended scenario type, the corresponding recommendation analysis process is executed to obtain the recommendation results.

[0010] The recommended results are then subjected to constraint verification and dynamic optimization to obtain optimized recommended results.

[0011] The optimized recommendation results are integrated to generate and output a personalized recommendation scheme.

[0012] Optionally, the user basic information includes user basic attribute information, learning background information, and ability level information; the learning-related status information includes learning progress information and task completion status information; and the constraint configuration information includes study hour constraints, resource type constraints, and time constraints.

[0013] Optionally, based on the multi-dimensional information, intelligently identify and recommend scenario types, specifically including the following steps.

[0014] Based on the multi-dimensional information, the user's current learning status and historical behavior patterns are analyzed.

[0015] Based on the current learning status and the historical behavior pattern, combined with the user-triggered event type, the current stage of the user is determined; the stage includes the resource selection stage and the learning stage.

[0016] Based on the user's current stage, determine the corresponding recommendation scenario type and match the appropriate recommendation strategy.

[0017] Optionally, when the recommended scenario type is the resource selection recommended scenario, the recommendation analysis process includes the following steps.

[0018] Based on the constraints corresponding to the constraint configuration information, the resource library is initially screened to obtain the screened resources.

[0019] Develop structured analysis instructions to guide intelligent analysis models in understanding recommendation requirements.

[0020] Based on the structured analysis instructions, a multi-threaded parallel processing method is used to call the intelligent analysis model to perform recommendation analysis on the filtered resources and obtain the recommendation results of each thread.

[0021] The recommendation results from each thread are collected, summarized, and integrated to obtain the final recommendation result.

[0022] Optionally, the multi-threaded parallel processing method includes the following:

[0023] The number of threads is dynamically determined based on resource type and analysis dimensions.

[0024] A synchronization control mechanism is adopted to enable the threads to work together.

[0025] Configure a timeout protection mechanism to avoid system blockage.

[0026] Optionally, when the recommended scenario type is the learning task recommended scenario, the recommendation analysis process includes the following steps.

[0027] Obtain the list of learning tasks associated with the user, classify and sort each task in the list, and obtain the classification and sorting results.

[0028] Based on the classification and sorting results, a batch display algorithm is used to determine the current display batch and corresponding task.

[0029] The task-related information of the currently displayed batch and corresponding task is intelligently completed to obtain the intelligently completed task-related information.

[0030] Based on the task-related information provided by the intelligent auto-completion, recommended operation options matching the task type are generated, and a recommendation result is obtained.

[0031] Optionally, the batch display algorithm includes the following:

[0032] The batch size is dynamically configured based on the data volume and user behavior characteristics.

[0033] Based on a caching mechanism, the user's browsing progress is recorded.

[0034] Based on a cyclic batching mechanism, the system displays content in cyclic batches and supports user-initiated refresh operations.

[0035] Calculate the start and end indices of the current batch being displayed.

[0036] Optionally, intelligent completion is performed on the task-related information of the currently displayed batch and corresponding task to obtain the intelligently completed task-related information, specifically including the following steps.

[0037] Based on the currently displayed batch and corresponding task, determine the task type.

[0038] Based on the task type, generate personalized task prompts.

[0039] Based on the personalized task prompts, the total duration and already studied duration of the course resources are completed.

[0040] Based on the personalized task prompts, the time limit and remaining attempts for the exam task are completed.

[0041] Based on the personalized task prompts, the previous learning progress for the continued learning task is displayed.

[0042] Based on the previous learning progress of the continued learning task, the total duration of the completed course resources and the already learned duration, as well as the time limit and remaining attempts of the completed exam task, the relevant information of the intelligently completed task is obtained.

[0043] Optionally, the recommendation results are subjected to constraint verification and dynamic optimization adjustment to obtain optimized recommendation results, specifically including the following steps.

[0044] The recommendation results are subjected to constraint verification to determine whether they meet the preset constraints, and the judgment result is obtained.

[0045] When the judgment result is negative, the recommendation result is optimized by means of replacement, supplementation and / or adjustment to obtain an optimized recommendation result.

[0046] When the judgment result is yes, the recommendation result is directly used as the optimized recommendation result.

[0047] Secondly, this application provides an intelligent course selection learning system based on multi-dimensional analysis. The intelligent course selection learning system based on multi-dimensional analysis is used to implement the intelligent course selection learning method based on multi-dimensional analysis as described in any of the first aspects. The intelligent course selection learning system based on multi-dimensional analysis includes the following modules.

[0048] The data acquisition module is used to acquire multi-dimensional information about users, including basic user information, historical behavior data, learning-related status information, and constraint configuration information.

[0049] The scene recognition module is used to intelligently identify and recommend scene types based on the multi-dimensional information; the recommended scene types include resource selection recommendation scenes and learning task recommendation scenes.

[0050] The recommendation analysis module is used to execute the corresponding recommendation analysis process according to the recommendation scenario type and obtain the recommendation results.

[0051] The verification and optimization module is used to verify the constraints of the recommendation results and dynamically optimize and adjust them to obtain optimized recommendation results.

[0052] The results output module is used to integrate the optimized recommendation results and generate and output a personalized recommendation scheme.

[0053] According to the specific embodiments provided in this application, this application has the following technical effects.

[0054] This application provides an intelligent course selection recommendation method and system based on multi-dimensional analysis. First, by comprehensively acquiring multi-dimensional information such as user basic information, historical behavior data, learning-related status information, and constraint configuration information, and combining this with multi-dimensional analysis, the recommendation scheme can deeply match the individual characteristics and needs of users, avoiding generalized recommendations and significantly improving the accuracy and suitability of recommended courses, thus enhancing the precision of personalized recommendations. Second, this application intelligently identifies resource selection recommendation scenarios and learning task recommendation scenarios based on multi-dimensional information, executing corresponding recommendation analysis processes for different scenarios to achieve scenario adaptability optimization. This ensures accurate and efficient matching of recommendation strategies with the user's current needs, improving the accuracy and efficiency of recommended course selection and avoiding ineffective recommendations caused by scenario mismatch. Third, this application verifies and dynamically optimizes the recommendation results by applying constraints, ensuring that the final recommendation results fully comply with preset constraint requirements, eliminating unsuitable recommendations, and improving the accuracy, compliance, reliability, and practicality of the recommendation results. In addition, this application generates personalized recommendation schemes by integrating and optimizing the recommendation results, providing users with clear and tailored resource or task guidance, reducing the time cost for users to filter information, improving the convenience and efficiency of users' course selection and learning process, and ultimately improving users' learning efficiency and user experience. Attached Figure Description

[0055] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0056] Figure 1 This is an application environment diagram of an intelligent course selection and learning method based on multi-dimensional analysis, provided as an embodiment of this application.

[0057] Figure 2 This is a flowchart illustrating an intelligent course selection and learning method based on multi-dimensional analysis, provided as an embodiment of this application.

[0058] Figure 3 This is a schematic diagram of a multi-dimensional information collection and scene recognition process provided in an embodiment of this application.

[0059] Figure 4 This is a schematic diagram illustrating the learning task recommendation and batch display process provided in one embodiment of this application.

[0060] Figure 5 This is a schematic diagram of the intelligent completion algorithm provided in one embodiment of this application.

[0061] Figure 6This is a schematic diagram of a recommendation scene recognition decision tree provided in an embodiment of this application.

[0062] Figure 7 This is a schematic diagram of the structure of an intelligent course selection and learning system based on multi-dimensional analysis, provided as an embodiment of this application. Detailed Implementation

[0063] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0064] The purpose of this application is to provide an intelligent course selection and learning method and system based on multi-dimensional analysis. This method can be widely applied in various fields such as educational resource recommendation, learning management systems, and online training platforms, and is particularly suitable for resource recommendation and learning task management in educational learning scenarios. In educational learning scenarios, the system can intelligently recommend suitable learning resources and tasks based on students' learning progress and ability levels. Through a batch display mechanism, it optimizes the user experience, significantly improving learning efficiency and satisfaction. Using scene recognition technology, it can automatically distinguish whether a user is currently in the course selection or learning stage and employs corresponding recommendation algorithms, achieving intelligent and personalized recommendation, providing users with comprehensive learning support and guidance. By comprehensively processing multi-dimensional user feature information, it achieves efficient and accurate personalized recommendations under complex constraints, while providing intelligent learning task recommendations and optimized information display mechanisms. This improves the accuracy and efficiency of course selection recommendations, thereby enhancing user learning efficiency and user experience, and solving the problems of low personalization, weak complex constraint handling capabilities, low processing efficiency, poor scene adaptability, insufficient learning guidance, and unreasonable information display in traditional recommendation systems.

[0065] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0066] The intelligent course selection and learning method based on multi-dimensional analysis provided in this application embodiment can be applied to, for example... Figure 1In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be set up independently, integrated into server 104, or placed in the cloud or on another server. Terminal 102 can send multi-dimensional user information to server 104. After receiving the multi-dimensional user information, server 104 intelligently identifies the recommendation scenario type based on the information; the recommendation scenario types include resource selection recommendation scenarios and learning task recommendation scenarios; according to the recommendation scenario type, it executes the corresponding recommendation analysis process to obtain recommendation results; it verifies the constraint conditions and dynamically optimizes the recommendation results to obtain optimized recommendation results; it integrates the optimized recommendation results to generate and output a personalized recommendation scheme. Server 104 can feed back the obtained personalized recommendation scheme to terminal 102. Furthermore, in some embodiments, the intelligent course selection and learning method based on multi-dimensional analysis can also be implemented by the server 104 or the terminal 102 separately. For example, the terminal 102 can directly perform multi-dimensional analysis and intelligent course selection processing based on the user's multi-dimensional information, or the server 104 can obtain the user's multi-dimensional information from the data storage system and perform multi-dimensional analysis and intelligent course selection processing based on the user's multi-dimensional information.

[0067] The terminal 102 can be, but is not limited to, various desktop computers, laptops, smartphones, tablets, and IoT devices. The server 104 can be implemented using a standalone server or a server cluster consisting of multiple servers, or it can be a cloud server.

[0068] In one exemplary embodiment, such as Figure 2 As shown, a method for intelligent course selection based on multi-dimensional analysis is provided. This method is executed by a computer device, specifically by a terminal or server alone, or by both a terminal and a server. In this embodiment, the method is applied to... Figure 1 Taking server 104 as an example, the explanation includes the following steps S1 to S5.

[0069] S1: Obtain multi-dimensional information about the user. This multi-dimensional information includes basic user information, historical behavioral data, learning-related status information, and constraint configuration information.

[0070] S2: Based on the aforementioned multi-dimensional information, intelligently identify the recommendation scenario type. The recommendation scenario type includes resource selection recommendation scenarios and learning task recommendation scenarios.

[0071] S3: Based on the recommended scenario type, execute the corresponding recommendation analysis process to obtain the recommendation result.

[0072] S4: Perform constraint verification and dynamic optimization on the recommendation results to obtain optimized recommendation results.

[0073] S5: Integrate the optimized recommendation results to generate and output a personalized recommendation scheme.

[0074] By implementing steps S1 to S5, comprehensive multi-dimensional information is acquired, including basic user information, historical behavior data, learning-related status information, and constraint configuration information. Combined with multi-dimensional analysis, the recommendation scheme deeply matches individual user characteristics and needs, avoiding generalized recommendations and significantly improving the accuracy and suitability of course selection, thus enhancing the precision of personalized recommendations. Based on multi-dimensional information, intelligent identification of resource selection and learning task recommendation scenarios is used. Corresponding recommendation analysis processes are executed for different scenarios to optimize scenario adaptability, achieving accurate and efficient matching between the recommendation strategy and the user's current needs. This improves the accuracy and efficiency of course selection and avoids ineffective recommendations due to scenario mismatch. By verifying and dynamically optimizing the recommendation results based on constraints, the final recommendation results are ensured to fully comply with preset constraints, eliminating unsuitable recommendations and improving the accuracy, compliance, reliability, and usability of the results. By integrating and optimizing the recommendation results, personalized recommendation schemes are generated, providing users with clear and tailored resource or task guidance, reducing the time cost for users to filter information, improving the convenience and efficiency of course selection and learning, and ultimately enhancing user learning efficiency and user experience.

[0075] As an optional implementation, in step S1, the user basic information includes user basic attribute information, learning background information, and ability level information; the learning-related status information includes learning progress information and task completion status information; and the constraint configuration information includes learning hour constraints, resource type constraints, and time constraints.

[0076] As an optional implementation method, step S2 intelligently identifies and recommends scenario types based on the multi-dimensional information, specifically including the following steps.

[0077] S21: Based on the multi-dimensional information, analyze the user's current learning-related status and historical behavior patterns.

[0078] S22: Based on the current learning-related status and the historical behavior pattern, combined with the user-triggered event type, determine the current stage of the user; the stage includes the resource selection stage and the learning stage.

[0079] S23: Based on the current stage of the user, determine the corresponding recommendation scenario type and match the corresponding recommendation strategy.

[0080] As an optional implementation, in step S3, when the recommended scenario type is the resource selection recommended scenario, the recommendation analysis process includes the following steps.

[0081] S31: Based on the constraints corresponding to the constraint configuration information, perform preliminary screening of the resource library to obtain the screened resources.

[0082] S32: Construct structured analysis instructions to guide intelligent analysis models in understanding recommendation requirements.

[0083] S33: Based on the structured analysis instructions, a multi-threaded parallel processing method is adopted to call the intelligent analysis model to perform recommendation analysis on the filtered resources and obtain the recommendation results of each thread.

[0084] S34: Collect the recommendation results from each thread, and then summarize and integrate them to obtain the final recommendation result.

[0085] As an optional implementation, in step S33, the multi-threaded parallel processing method includes the following:

[0086] (1) Dynamically determine the number of threads based on resource type and analysis dimension.

[0087] (2) A synchronization control mechanism is adopted to enable the threads to work together.

[0088] (3) Configure a timeout protection mechanism to avoid system blockage.

[0089] As an optional implementation, in step S3, when the recommended scenario type is the learning task recommended scenario, the recommendation analysis process includes the following steps.

[0090] S31: Obtain the learning task list associated with the user, classify and sort each task in the learning task list, and obtain the classification and sorting results.

[0091] S32: Based on the classification and sorting results, a batch display algorithm is used to determine the current display batch and corresponding task.

[0092] S33: Perform intelligent completion on the task-related information of the currently displayed batch and corresponding task to obtain the intelligently completed task-related information.

[0093] S34: Based on the task-related information provided by the intelligent completion, generate recommended operation options that match the task type, and obtain the recommendation result.

[0094] As an optional implementation, in step S32, the batch display algorithm includes the following:

[0095] (1) Dynamically configure the batch size based on the data volume and user behavior characteristics.

[0096] (2) Based on the caching mechanism, record the user's browsing progress.

[0097] (3) Based on the cyclic batching mechanism, the display is realized in cyclic batches and the user can actively refresh the operation.

[0098] (4) Calculate the start and end indices of the current batch being displayed.

[0099] As an optional implementation, step S33 performs intelligent completion on the task-related information of the currently displayed batch and corresponding task to obtain the intelligently completed task-related information, specifically including the following steps.

[0100] S331: Determine the task type based on the currently displayed batch and the corresponding task.

[0101] S332: Generate personalized task prompt information based on the task type.

[0102] S333: Based on the personalized task prompt information, complete the total duration and already studied duration of the course resources.

[0103] S334: Based on the personalized task prompt information, complete the time limit and remaining attempts for the exam task.

[0104] S335: Based on the personalized task prompt information, display the previous learning progress of the continuing learning task.

[0105] S336: Based on the previous learning progress of the continued learning task, the total duration of the completed course resources and the already learned duration, as well as the time limit and remaining attempts of the completed exam task, obtain the relevant information of the intelligently completed task.

[0106] As an optional implementation, step S4 verifies the constraints and dynamically optimizes the recommendation results to obtain optimized recommendation results, specifically including the following steps.

[0107] The recommendation results are subjected to constraint verification to determine whether they meet the preset constraints, and the judgment result is obtained, including the following two cases.

[0108] (1) When the judgment result is negative, the recommendation result is optimized by means of replacement, supplementation and / or adjustment to obtain the optimized recommendation result.

[0109] (2) When the judgment result is yes, the recommendation result is directly used as the optimized recommendation result.

[0110] This application executes the corresponding recommendation analysis process according to the recommendation scenario type to obtain preliminary recommendation results. Based on the preliminary recommendation results, it verifies whether the preliminary recommendation results meet the preset constraints, thereby determining whether the preliminary recommendation results are consistent and reasonable. If they do not meet the preset constraints, the preliminary recommendation results are further optimized and adjusted by means of replacement, supplementation and / or adjustment to obtain more accurate, reliable and optimized recommendation results. By combining constraint verification with recommendation result optimization, the accuracy and reliability of the recommendation results are improved.

[0111] To make the technical solution of this application clearer, the specific implementation process of the technical solution of this application will be explained in detail below with examples. For example... Figure 3 As shown, the implementation steps include the following.

[0112] S1: Multi-dimensional information collection.

[0113] By collecting basic user information, historical behavioral data, learning-related status information, and constraint configuration information, multi-dimensional user information is formed, which can be further used to build multi-dimensional user profiles and resource feature libraries. Integrating this multi-dimensional information creates a complete user profile, providing a data foundation for subsequent intelligent recommendations.

[0114] S2: Intelligent scene recognition.

[0115] Based on the multi-dimensional information of the user collected in step S1, the recommended scene type of the user is currently in is determined by the scene recognition algorithm, including the following steps.

[0116] S21: Obtain the current learning task completion status of the user and count the number of incomplete tasks.

[0117] S22: Analyze users' historical behavior patterns to determine user intent (course selection or learning).

[0118] S23: Detect the user's current operation context and identify the type of event triggered.

[0119] S24: Classify scenarios based on decision rules, including the following two cases.

[0120] (1) If a user has not completed a learning task, it is identified as a “learning task recommendation scenario”.

[0121] (2) If the user has no unfinished tasks or actively triggers the course selection event, it is identified as a "resource selection recommendation scenario".

[0122] S3: Resource selection and recommendation process.

[0123] Based on the "resource selection recommendation scenario" identified in step S2, the following steps are executed using a multi-threaded parallel AI recommendation algorithm.

[0124] S31: Pre-screen the resource library according to the constraints and establish a hierarchical relationship of resource types.

[0125] S32: Construct structured AI analysis prompts, including four levels: role definition, recommendation goal, knowledge base information, and output format. These prompts are used to guide the large AI model to understand the recommendation task and constraints.

[0126] S33: Dynamically create multiple recommendation threads based on resource type. Each thread combines the constructed prompt words with the corresponding resource data and calls the AI ​​large model interface for intelligent analysis. The AI ​​returns a list of recommended resource indexes based on the guidance information in the prompt words.

[0127] S34: Use a synchronization mechanism to wait for all threads to complete and collect the recommended indexes returned by each thread.

[0128] S35: Based on the collected recommendation index, verify the constraints of the recommendation results and make dynamic adjustments based on the hierarchical relationship.

[0129] S36: Generate a personalized resource recommendation list that meets the constraints.

[0130] S4: Learning task recommendation process.

[0131] like Figure 4 As shown, based on the "learning task recommendation scenario" identified in step S2, the following steps are executed through an intelligent task analysis algorithm.

[0132] S41: Get a list of all learning tasks for the user and sort them by task type and priority.

[0133] S42: Employs an intelligent batch display algorithm to calculate the current display batch based on the number of cache refreshes.

[0134] S43: Employs an intelligent completion algorithm to intelligently complete task information, including task prompts, resource duration, and learning progress. The intelligent completion algorithm process is as follows: Figure 5 As shown.

[0135] S44: Recommended action options related to the generated task type.

[0136] S45: Output the personalized learning task recommendation list for the current batch.

[0137] S5: Optimized batch display.

[0138] Based on the learning task recommendation list generated in step S4, the following content is achieved through a batch display optimization algorithm.

[0139] S51: Dynamically calculate the optimal batch size for display based on user behavior patterns and data volume.

[0140] S52: Implements a cyclic batching mechanism to support users' "switch" operation.

[0141] S53: Employs a caching mechanism to record user browsing progress, providing a continuous experience.

[0142] S54: When the user triggers a refresh operation, return to step S4 to recalculate the next batch of tasks.

[0143] S6: Personalized result output.

[0144] Based on the recommended scenario type identified in step S2, the recommendation results from step S3 or S4 are integrated, including the following steps.

[0145] S61: If it is a resource selection and recommendation scenario, output a resource recommendation list and course selection operation guidance.

[0146] S62: If it is a learning task recommendation scenario, output a task recommendation list and learning operation guidance.

[0147] S63: Generate recommended topics and quick operation options related to the recommended scenarios.

[0148] S64: Return the complete personalized recommendation to the user.

[0149] As an optional implementation method, the intelligent course selection learning method based on multi-dimensional analysis proposed in this embodiment mainly includes data collection, scene recognition, resource recommendation, learning task recommendation, batch display, and result output. The above tasks are respectively completed by the data collection module, scene recognition module, resource recommendation module, learning task recommendation module, batch display module, and result output module.

[0150] Step 1: Multi-dimensional information collection.

[0151] The data acquisition module collects multi-dimensional user information using the following algorithms.

[0152] (1) Collection of basic user information: Constructing user feature vector U={u1,u2,...,u n}, where u i This represents the i-th feature dimension of the user, and n is the total number of features. Features include, but are not limited to, learning background, major, and ability level (ability level is an item in the student's basic information, which is filled in by the student according to their own situation).

[0153] (2) Historical behavior data collection: Constructing a behavior sequence H={h1, h2, ..., hm}, where h j =(action, resource, timestamp) represents the j-th historical action, where action represents the action type (course selection, learning, completion, etc.), and completion means that the selected course has passed the assessment and the learning task has ended. resource represents the resource object involved, timestamp represents the time when the action occurred, and m is the total number of historical actions.

[0154] (3) Learning progress information collection: Construct a progress status set P={p1, p2, ..., p k}, where p k =(taskId, status, progress, remainTime) represents the k-th learning task, status ∈ {not started, in progress, completed}, progress ∈ [0, 1] represents the percentage of completion, and remainTime represents the remaining learning time.

[0155] (4) Constraint configuration and collection: Construct a constraint set C = {c1, c2, ..., c...} l}, where c l =(type, value, operator) represents the l-th constraint condition, where type represents the constraint type (study hours, type, time, etc.), value represents the constraint value, and operator∈{≥,≤,=,∈} represents the constraint relationship.

[0156] (5) Construct a comprehensive user profile: User Profile = f(U, H, P, C), where f is a feature fusion function that integrates multi-dimensional information into a unified user profile representation.

[0157] Step 2: Scene recognition.

[0158] The scene recognition module analyzes the user's status to determine the recommended scene type.

[0159] The scene recognition module executes a scene recognition algorithm based on the user information collected in step 1. This algorithm employs a recommended scene recognition decision tree approach, as described below. Figure 6 As shown.

[0160] Scene recognition includes the following operations: obtaining the user's current learning task status, analyzing the user's behavioral intent, obtaining triggering events (triggering events include: tasks that are about to end, continuing to learn the previous course, viewing live streams, downloading certificates, taking exams, making learning plans, and selecting courses), analyzing historical patterns (historical patterns include historical scenarios where students have completed all tasks and where there are incomplete tasks, including tasks such as learning courses, taking exams, writing learning reflections, filling out questionnaires, paying for orders, printing certificates, and live learning), making scene decision judgments, and determining the scene recognition result (resource selection recommended scene or learning task recommended scene).

[0161] Based on the scene recognition results, the system initiates different recommendation processes, including the following situations.

[0162] If the scenario is identified as "resource selection and recommendation scenario", proceed with the resource recommendation process in step 3.

[0163] If the scenario is identified as a "learning task recommendation scenario", proceed with the learning task recommendation process in step 4.

[0164] If the user is in the resource selection phase, initiate the resource recommendation process.

[0165] If the user is in the learning phase, initiate the learning task recommendation process.

[0166] Step 3: Resource Recommendation Process.

[0167] The resource recommendation module executes the following algorithm.

[0168] (1) Obtain the required study hours configuration and type hierarchy relationship.

[0169] (2) Obtain basic student information and historical records. The basic student information and the basic user information are consistent. A student is a user. Historical records refer to historical learning behaviors and outcomes, serving as a basis for analyzing user preferences for different learning types.

[0170] (3) Determine the scope of the query (course / class) and obtain available resources.

[0171] (4) Perform preliminary filtering based on the upper limit of study hours.

[0172] (5) Create an AI client and structured prompts. The architecture of the AI ​​client is as follows.

[0173] Role definition layer: Define AI as a professional learning consultant.

[0174] Target setting layer: Set recommendation strategies.

[0175] Knowledge base layer: Integrates user information and resource data.

[0176] Output format layer: Standardizes the format of returned results.

[0177] (6) Dynamically calculate the number of threads and create a Count Down Latch synchronizer.

[0178] (7) Multi-threaded parallelism is recommended.

[0179] Course recommendation thread (if any courses are available).

[0180] Recommended courses by type (if there is a type hierarchy).

[0181] Recommend a class thread (if any classes are available).

[0182] Recommended thread for class type (if there is a type hierarchy).

[0183] The specific course / class types vary depending on the program. They are generally divided into two main categories: professional courses and general education courses, with a varying number of subcategories within each category.

[0184] (8) Wait for all threads to complete (3-minute timeout).

[0185] (9) Verify and adjust the learning hours (dual mapping + hierarchical optimization).

[0186] (10) Integrate the recommendation results.

[0187] (11) Return to the complete recommended solution.

[0188] Step 4: Learning task recommendation process.

[0189] The learning task recommendation module executes the following learning task recommendation process.

[0190] (1) Get the list of all learning tasks of the user.

[0191] (2) Sort the tasks.

[0192] (3) If there is a learning task, it includes the following.

[0193] a) Set batching parameters: limi=3.

[0194] b) Calculate refresh count: Call the Refresh Times = get Refresh Times(cache key, list size, batch size, refresh flag) program. Perform the following operations.

[0195] (1) Initialize batch number n=1.

[0196] (2) If it is not a refresh operation, clear the cache and return n=1.

[0197] (3) If it is a refresh operation.

[0198] a) Retrieve the current batch number from the cache.

[0199] b) Calculate the total number of batches.

[0200] c) Increment the batch number based on the initial batch number n.

[0201] d) If the total number of batches is exceeded, reset to 1 (for loop display).

[0202] e) Update the cache.

[0203] (4) Return the current batch number.

[0204] Call the Fetch Batch(record list, batch number, batch size) program to perform the following operations.

[0205] a) If the batch size is greater than or equal to the total number of records, return all records.

[0206] b) Calculate the starting index: Start Index = (batch number - 1) × batch size.

[0207] c) Calculate the end index: End Index = min(Start Index + batch size, total number of records).

[0208] d) Returns a sublist: Records.subList(Start Index, End Index).

[0209] e) Get the current batch of tasks.

[0210] Current Batch = Fetch Batch (AllTask, Refresh Times, limit).

[0211] f) Complete the task information.

[0212] Fill Task Head Tips (Current Batch) / / Add task tips.

[0213] Fix Resource Times (Current Batch) / / Fix resource duration Specifically, invoke the Fill Task Head Tips program and perform the following operations.

[0214] For each task in the task list.

[0215] 1) Obtain the task type and group type.

[0216] 2) Generate personalized prompts based on task type.

[0217] Course task: Display learning progress and passing rules.

[0218] Exam task: Displays exam requirements and time limits.

[0219] Continue learning task: Displays the progress of the last learning session.

[0220] 3) Set the task header prompt message.

[0221] Invoke the Fix Resource Times (task list, site, user) program to perform the following operations.

[0222] For each task in the task list.

[0223] 1) Complete the resource information according to the task type.

[0224] Course task: Complete the total course duration and the duration already completed.

[0225] Exam task: Complete the exam duration and remaining attempts.

[0226] 2) Batch update resource duration information in the database.

[0227] The recommended actions generated in this embodiment include the following:

[0228] Call the Fix Study Recommend Operate (Recommendation Results, Task List) program to perform the following operations.

[0229] 1) Initialize the recommended operation list.

[0230] 2) Traverse the task list and extract the unique task grouping types.

[0231] 3) Generate recommended action buttons for each task group type.

[0232] 4) Add general operations.

[0233] "More tasks" button (placed first).

[0234] The "Study Plan" button is located at the end.

[0235] 5) Set recommended actions to the results.

[0236] 6) If there is no learning task, call Completed Results to generate a completion status result.

[0237] 7) Return the recommended results.

[0238] As an optional implementation, the batch display module uses an intelligent batch display algorithm, which includes the following steps.

[0239] Call the Get Refresh Times(cache key, list size, batch size, refresh flag) program to perform the following operations.

[0240] 1) Initialize batch number n=1.

[0241] 2) If it is not a refresh operation.

[0242] Clear the cache and return n=1.

[0243] 3) If it is a refresh operation.

[0244] a) Retrieve the current batch number from the cache.

[0245] b) Calculate the total number of batches: total Batches = ceil (list size / batch size).

[0246] c) Increment the batch number.

[0247] d) If the total number of batches is exceeded, reset to 1 (for loop display).

[0248] e) Update the cache.

[0249] 4) Return the current batch number.

[0250] Call the Fetch Batch(record list, batch number, batch size) program to perform the following operations.

[0251] 1) If the batch size is greater than or equal to the total number of records, return all records.

[0252] 2) Calculate the starting index: start Index = (batch number - 1) × batch size.

[0253] 3) Calculate the end index: end Index = min(start Index + batch size, total number of records).

[0254] 4) Returns a sublist: records.subList(startIndex, endIndex).

[0255] As an optional implementation, the learning task information completion algorithm includes the following steps.

[0256] Invoke the Fill Task Head Tips program and perform the following operations.

[0257] For each task in the task list.

[0258] 1) Obtain the task type and group type.

[0259] 2) Generate personalized prompts based on task type.

[0260] 3) Course tasks: Displays learning progress and passing rules.

[0261] 4) Exam Task: Displays exam requirements and time limits.

[0262] 5) Continue learning task: Displays the progress of the last learning session.

[0263] In this embodiment, the task header prompt information is mainly used to display a summary description of the student's current scenario status. For example: You currently have n learning tasks to complete, and you have completed n learning tasks.

[0264] Invoke the Fix Resource Times (task list, site, user) program to perform the following operations.

[0265] For each task in the task list.

[0266] 1) Complete the resource information according to the task type.

[0267] Course task: Complete the total course duration and the duration already completed.

[0268] Exam task: Complete the exam duration and remaining attempts.

[0269] 2) Batch update resource duration information in the database.

[0270] As an optional implementation, the recommended operation generation algorithm includes the following steps.

[0271] Call the Fix Study Recommend Operate (Recommendation Results, Task List) program to perform the following operations.

[0272] 1) Initialize the recommended operation list.

[0273] 2) Traverse the task list and extract the unique task grouping types.

[0274] 3) Generate recommended action buttons for each task group type.

[0275] 4) Add general operations.

[0276] "More tasks" button (placed first).

[0277] The "Study Plan" button is located at the end.

[0278] 5) Set recommended actions to the results.

[0279] Based on the same inventive concept, this application also provides a multi-dimensional analysis-based intelligent course selection learning system for implementing the aforementioned multi-dimensional analysis-based intelligent course selection learning method. The solution provided by this multi-dimensional analysis-based intelligent course selection learning system is similar to the solution described in the above method. Therefore, the specific limitations in the embodiments of the multi-dimensional analysis-based intelligent course selection learning system provided below can be found in the limitations of the multi-dimensional analysis-based intelligent course selection learning method described above, and will not be repeated here.

[0280] In one exemplary embodiment, such as Figure 7 As shown, an intelligent course selection and learning system based on multi-dimensional analysis is provided, which includes the following modules.

[0281] The data acquisition module is used to acquire multi-dimensional information about users, including basic user information, historical behavior data, learning-related status information, and constraint configuration information.

[0282] The scene recognition module is used to intelligently identify and recommend scene types based on the multi-dimensional information; the recommended scene types include resource selection recommendation scenes and learning task recommendation scenes.

[0283] The recommendation analysis module is used to execute the corresponding recommendation analysis process according to the recommendation scenario type and obtain the recommendation results.

[0284] The verification and optimization module is used to verify the constraints of the recommendation results and dynamically optimize and adjust them to obtain optimized recommendation results.

[0285] The results output module is used to integrate the optimized recommendation results and generate and output a personalized recommendation scheme.

[0286] This application proposes an intelligent course selection and learning method and system based on multi-dimensional analysis. This method acquires multi-dimensional information and resource configuration requirements from users, uses an artificial intelligence large language model for multi-threaded parallel analysis, and optimizes and adjusts the recommendation results by combining a hierarchical verification mechanism. At the same time, it provides intelligent learning task recommendation and batch display mechanism to achieve accurate personalized resource recommendation and learning path planning. It solves the technical problems of low personalization, poor processing efficiency, insufficient matching accuracy, and lack of learning progress tracking in existing recommendation systems. It has broad application prospects in the fields of educational resource recommendation, learning management, and e-commerce recommendation.

[0287] The intelligent course selection and learning method and system proposed in this application based on multi-dimensional analysis has the following technical advantages.

[0288] (1) Significantly improved recommendation accuracy: Through multi-dimensional information comprehensive analysis and artificial intelligence deep learning, the recommendation accuracy is improved by more than 30% compared with traditional methods.

[0289] (2) Significantly optimized processing efficiency: The multi-threaded parallel processing architecture is adopted, and the response time is recommended to be reduced by more than 50% compared with serial processing.

[0290] (3) Constraint matching is accurate and reliable: Through hierarchical verification and dynamic adjustment mechanism, the recommendation results are ensured to meet the constraint requirements 100%.

[0291] (4) Significantly improved learning experience: Through intelligent learning task recommendation and batch display, users’ learning efficiency is improved by more than 40%.

[0292] (5) Improved information display efficiency: The intelligent batch display mechanism improves the efficiency of user information acquisition by more than 60%.

[0293] (6) Strong system stability: The system adopts anomaly isolation and timeout protection mechanism, and the system availability reaches more than 99.9%.

[0294] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0295] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for intelligent course recommendation and learning based on multi-dimensional analysis, characterized in that, The intelligent course selection and learning method based on multi-dimensional analysis includes: Acquire multi-dimensional information about the user; the multi-dimensional information includes basic user information, historical behavior data, learning-related status information, and constraint configuration information. Based on the multi-dimensional information, the recommendation scenario type is intelligently identified; the recommendation scenario type includes resource selection recommendation scenario and learning task recommendation scenario. Based on the recommended scenario type, execute the corresponding recommendation analysis process to obtain the recommendation result; The recommended results are subjected to constraint verification and dynamic optimization adjustment to obtain optimized recommended results; The optimized recommendation results are integrated to generate and output a personalized recommendation scheme.

2. The intelligent course selection and learning method based on multi-dimensional analysis according to claim 1, characterized in that, The user basic information includes user basic attribute information, learning background information, and ability level information; the learning-related status information includes learning progress information and task completion status information; the constraint configuration information includes learning hour constraints, resource type constraints, and time constraints.

3. The intelligent course selection and learning method based on multi-dimensional analysis according to claim 1, characterized in that, Based on the aforementioned multi-dimensional information, the system intelligently identifies and recommends scenario types, specifically including: Based on the aforementioned multi-dimensional information, analyze the user's current learning status and historical behavior patterns; Based on the current learning status and the historical behavior pattern, combined with the user-triggered event type, the current stage of the user is determined; the stage includes the resource selection stage and the learning stage. Based on the user's current stage, determine the corresponding recommendation scenario type and match the appropriate recommendation strategy.

4. The intelligent course selection and learning method based on multi-dimensional analysis according to claim 1, characterized in that, When the recommended scenario type is the resource selection recommended scenario, the recommendation analysis process includes: Based on the constraints corresponding to the constraint configuration information, the resource library is initially screened to obtain the screened resources. Develop structured analysis instructions to guide intelligent analysis models in understanding recommendation needs; Based on the structured analysis instructions, a multi-threaded parallel processing method is used to call the intelligent analysis model to perform recommendation analysis on the filtered resources and obtain the recommendation results of each thread. The recommendation results from each thread are collected, summarized, and integrated to obtain the final recommendation result.

5. The intelligent course selection and learning method based on multi-dimensional analysis according to claim 4, characterized in that, The multi-threaded parallel processing method includes: The number of threads is dynamically determined based on resource type and analysis dimensions; A synchronization control mechanism is adopted to enable the threads to work together. Configure a timeout protection mechanism to avoid system blockage.

6. The intelligent course selection and learning method based on multi-dimensional analysis according to claim 1, characterized in that, When the recommended scenario type is the learning task recommendation scenario, the recommendation analysis process includes: Obtain the learning task list associated with the user, classify and sort each task in the learning task list, and obtain the classification and sorting results; Based on the classification and sorting results, a batch display algorithm is used to determine the current display batch and corresponding task; The task-related information of the currently displayed batch and corresponding task is intelligently completed to obtain the intelligently completed task-related information. Based on the task-related information provided by the intelligent auto-completion, recommended operation options matching the task type are generated, and a recommendation result is obtained.

7. The intelligent course selection and learning method based on multi-dimensional analysis according to claim 6, characterized in that, The batch display algorithm includes: Dynamically configure batch size based on data volume and user behavior characteristics; Based on a caching mechanism, the user's browsing progress is recorded; Based on a cyclic batching mechanism, the system displays content in cyclic batches and supports user-initiated refresh operations. Calculate the start and end indices of the current batch being displayed.

8. The intelligent course selection and learning method based on multi-dimensional analysis according to claim 6, characterized in that, The task-related information of the currently displayed batch and corresponding task is intelligently completed to obtain the intelligently completed task-related information, specifically including: Based on the currently displayed batch and corresponding task, determine the task type; Based on the task type, generate personalized task prompts; Based on the personalized task prompts, the total duration and already learned duration of the course resources are completed. Based on the personalized task prompts, complete the time limit and remaining attempts for the exam task; Based on the personalized task prompt information, the previous learning progress of the continuing learning task is displayed; Based on the previous learning progress of the continued learning task, the total duration of the completed course resources and the already learned duration, as well as the time limit and remaining attempts of the completed exam task, the relevant information of the intelligently completed task is obtained.

9. The intelligent course selection and learning method based on multi-dimensional analysis according to claim 1, characterized in that, The recommended results are subjected to constraint verification and dynamic optimization to obtain optimized recommended results, specifically including: The recommendation results are subjected to constraint verification to determine whether they meet the preset constraints, and the judgment result is obtained. When the judgment result is negative, the recommendation result is optimized by means of replacement, supplementation and / or adjustment to obtain an optimized recommendation result; When the judgment result is yes, the recommendation result is directly used as the optimized recommendation result.

10. An intelligent course selection and learning system based on multi-dimensional analysis, characterized in that, The intelligent course selection and learning system based on multi-dimensional analysis is used to implement the intelligent course selection and learning method based on multi-dimensional analysis as described in any one of claims 1-9, wherein the intelligent course selection and learning system based on multi-dimensional analysis includes: The data acquisition module is used to acquire multi-dimensional information about the user, including basic user information, historical behavior data, learning-related status information, and constraint configuration information. The scene recognition module is used to intelligently identify and recommend scene types based on the multi-dimensional information; the recommended scene types include resource selection recommendation scenes and learning task recommendation scenes. The recommendation analysis module is used to execute the corresponding recommendation analysis process according to the recommendation scenario type and obtain the recommendation result; The verification and optimization module is used to verify the constraints and dynamically optimize the recommendation results to obtain the optimized recommendation results. The results output module is used to integrate the optimized recommendation results and generate and output a personalized recommendation scheme.

Images