Methods, devices, electronic equipment and storage media for analyzing the evolution of learning outcomes

Abstract

This disclosure provides a method, apparatus, electronic device, and storage medium for analyzing learning progress. The method includes: acquiring a user dataset from a platform; determining an instant learning performance score for each learner based on their learning outcomes and learning objectives within the user dataset; acquiring valid course selection information for all learners; determining a dynamic relationship network based on the valid course selection information and the instant learning performance score; acquiring the clustering characteristics and community structure within the dynamic relationship network; and determining the learning progress based on the clustering characteristics and community structure. Determining the learning progress through a dynamic network based on the instant learning performance score effectively grasps the learning progress patterns of open online courses, thereby enabling real-time monitoring of the learning progress of all users on the platform. This plays a crucial role for educators in real-time evaluation of teaching quality, control of teaching direction, course recommendations, and learning progress early warnings.

Description

Technical Field

[0001] This disclosure relates to the field of educational informatization technology, and in particular to a method, device, electronic device, and storage medium for analyzing student learning evolution. Background Technology

[0002] MOOCs (Massive Open Online Courses) offer a new model of online education. MOOC platforms are developing rapidly both domestically and internationally, and their influence continues to expand. Unlike traditional offline education, MOOCs attract a large number of learners due to their openness and autonomy in participation; however, the lack of effective teaching management has also resulted in a high dropout rate. Increasingly, scholars both at home and abroad are focusing their research on the analysis and mining of massive amounts of data from MOOC platforms.

[0003] Learners' learning and interaction behaviors on MOOC platforms generate a large amount of data, forming a complex relationship network. Analyzing this MOOC relationship network helps to reveal the inherent learning patterns of MOOC learners and the evolution of MOOC learning outcomes.

[0004] Currently, researchers both domestically and internationally tend to study the impact of certain learning behavior data on learners' learning outcomes. However, the main evaluation method for existing MOOC learning outcomes is to calculate the weighted composite score of learners' video viewing time, participation in discussions, daily assignments, and final exams. But this method can only score the courses that learners complete and cannot provide timely and dynamic feedback on learners' learning progress. Summary of the Invention

[0005] In view of this, the purpose of this disclosure is to propose a method, device, electronic device and storage medium for analyzing the evolution of learning status, so as to grasp the learning status of all users on the platform in real time.

[0006] Based on the above objectives, this disclosure provides a method for analyzing the evolution of student learning, including:

[0007] Obtain the platform's user dataset, and determine an instant score for the learner's learning effect based on the learning outcomes and learning objectives in the user dataset;

[0008] Obtain valid course selection information for all learners, score learning outcomes in real time based on the valid course selection information, and determine a dynamic relationship network;

[0009] Obtain the clustering characteristic community structure in the dynamic relationship network, and determine the evolution of learning situation based on the clustering characteristic community structure.

[0010] Optionally, obtaining all learners' valid course selection information and the real-time learning performance scores, and determining the dynamic relationship network, includes:

[0011] Obtain course offering information, course selection information, and course cancellation information, and determine the valid course selection information;

[0012] Determine the similarity in learning among the target learner and all other learners at a specific time.

[0013] Based on the similarity of the learning situations, the dynamic relationship network is determined.

[0014] Optionally, obtaining all learners' valid course selection information and the real-time learning performance scores, and determining the dynamic relationship network, includes:

[0015] Obtain course offering information, course selection information, and course cancellation information, and determine the valid course selection information;

[0016] Determine the similarity in learning among the target learner and all other learners at a specific time.

[0017] Based on the similarity of the learning situations, the dynamic relationship network is determined.

[0018] Optionally, determining the dynamic relationship network based on the similarity of learning situations includes:

[0019] The adjacency matrix is ​​determined based on the aforementioned similarity relationships among students;

[0020] The target learner is used as a network node, and the dynamic relationship network is constructed based on the adjacency matrix.

[0021] Optionally, obtaining the clustering characteristics and community structure in the dynamic relationship network to determine the evolution of learning progress includes:

[0022] The nodes in the dynamic relationship network are adjusted according to the clustering characteristics and community structure.

[0023] The rationality of the dynamic relationship network is tested to determine the evolution of the learning situation.

[0024] Optionally, the step of detecting the rationality of the dynamic relationship network and determining the evolution of learning progress includes:

[0025] The community structure of the dynamic relationship network is adjusted using a faction filtering algorithm;

[0026] For each community in the adjusted community structure, construct the corresponding evolutionary sequence for that community;

[0027] The target community is determined from the community structure, and the communities adjacent to the target community are determined as adjacent communities;

[0028] The evolutionary sequence corresponding to the target community is compared with the evolutionary sequence corresponding to the neighboring communities to determine the community evolutionary morphology of the target community;

[0029] Based on the community's evolutionary pattern, the evolution of students' learning situation is determined.

[0030] Optionally, adjusting the community structure of the dynamic relationship network using a faction filtering algorithm includes:

[0031] The community structure of the dynamic relationship network is detected to obtain the community distribution of the subgraphs in the dynamic relationship network;

[0032] The order of the subgraphs is adjusted according to their size;

[0033] The adjusted subgraphs are merged, and the merged subgraphs are assigned to communities to obtain the adjusted community structure.

[0034] Based on the same inventive concept, corresponding to any of the above embodiments, one or more embodiments of this specification also provide a learning evolution analysis device, including:

[0035] The evaluation module is configured to acquire the platform's user dataset and determine an instant score for the learner's learning performance based on the learning outcomes and learning objectives in the user dataset.

[0036] The module is configured to acquire all learners’ valid course selection information, and determine a dynamic relationship network based on the valid course selection information and the learning effect in real time.

[0037] The analysis module is configured to obtain the clustering characteristics and community structure in the dynamic relationship network, and determine the evolution of learning status based on the clustering characteristics and community structure.

[0038] Based on the same inventive concept, corresponding to any of the above embodiments, one or more embodiments of this specification also provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the program to implement the method described in any of the above embodiments.

[0039] Based on the same inventive concept, corresponding to any of the above embodiments, one or more embodiments of this specification also provide a non-transitory computer-readable storage medium, characterized in that the non-transitory computer-readable storage medium stores computer instructions for causing a computer to perform the method as described in any of the above embodiments.

[0040] As described above, the learning progress analysis method, apparatus, electronic device, and storage medium disclosed herein firstly assess learners' learning outcomes in real-time based on course distribution information and viewing activity; then, it obtains valid course selection information for all learners on the platform and determines a dynamic relationship network based on this information and the real-time learning outcome assessment; finally, it obtains the clustering characteristics and community structure within the dynamic relationship network and determines the learning progress based on this structure. This solution, by determining the learning progress based on the dynamic network established through real-time learning outcome assessment, effectively grasps the learning progress patterns of open online courses, thereby enabling real-time monitoring of the learning status of all users on the platform. This plays a crucial role for educators in real-time evaluation of teaching quality, control of teaching direction, course recommendation, and learning progress early warning applications. Attached Figure Description

[0041] To more clearly illustrate the technical solutions in this disclosure or related technologies, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only embodiments of this disclosure. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0042] Figure 1 This is a flowchart of the learning situation evolution analysis method according to an embodiment of the present disclosure;

[0043] Figure 2 This is a schematic diagram illustrating the learning progress analysis of an embodiment of this disclosure;

[0044] Figure 3 This is a schematic diagram of the evolutionary sequence of embodiments of this disclosure;

[0045] Figure 4 This is a schematic diagram of the electronic device structure according to an embodiment of the present disclosure. Detailed Implementation

[0046] To make the objectives, technical solutions, and advantages of this disclosure clearer, the following detailed description is provided in conjunction with specific embodiments and the accompanying drawings.

[0047] It should be noted that, unless otherwise defined, the technical or scientific terms used in the embodiments of this disclosure should have the ordinary meaning understood by one of ordinary skill in the art to which this disclosure pertains. The terms "first," "second," and similar words used in the embodiments of this disclosure do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Words such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, but do not exclude other elements or objects.

[0048] As mentioned in the background section, MOOCs (Massive Open Online Courses) differ from traditional offline education. MOOCs attract a large number of learners due to their openness and autonomy of participation. However, due to the lack of effective teaching management, they also have a high dropout rate. Therefore, designing an effective method for analyzing the evolution of MOOC learning is an urgent problem to be solved.

[0049] The main reason for the above problems is that learners' learning and interaction behaviors on the platform generate a large amount of data, thus forming a learner relationship network. This network evolves continuously as learners' learning progress changes. Without being able to describe this dynamically evolving MOOC learner relationship network, it is impossible to grade the courses completed by learners, provide timely and dynamic feedback on their learning progress, and therefore, it is impossible to analyze learners' learning status in a timely and effective manner.

[0050] In developing this disclosure, the applicant discovered that learners' learning outcomes are not only influenced by their own learning behaviors but also by other learners closely connected to them within their social networks. Furthermore, the learning situations of MOOC learners evolve over time, and these changes inevitably impact the overall structure of the MOOC social network. As a novel educational model distinct from traditional education, MOOCs currently lack effective methods for analyzing the evolution of MOOC learner situations; therefore, designing an effective method for analyzing the evolution of MOOC learner situations is a pressing issue.

[0051] In view of the problems existing in the prior art and based on the applicant's findings, one or more embodiments of this specification provide a learning situation evolution analysis method. This method can analyze the learning situation evolution pattern of MOOC learner relationship network and can reflect the learner's current learning situation and learning effect in real time, providing decision support and intervention means for MOOC learning situation analysis, personalized recommendation and learning situation early warning and other applications.

[0052] The technical solutions of one or more embodiments of this specification will be further described in detail below through specific examples.

[0053] First, this specification provides an embodiment of a learning progress analysis method that can reflect learners' current learning status and learning outcomes in real time, providing decision support and intervention methods for applications such as MOOC learning progress analysis, personalized recommendations, and learning progress early warning.

[0054] Furthermore, the application scenario of the learning evolution analysis method in this disclosure is a crowdsourcing-based learning platform. Crowdsourcing refers to the practice of a company or organization outsourcing work tasks that were previously performed by employees to a large number of non-specific (and often large) volunteers on a voluntary basis. For example, a company seeking a marketing idea does not outsource it to a specific or designated unit, but instead publishes it to the public via the Internet.

[0055] First, refer to Figure 1 This specification provides an embodiment of a method for analyzing the evolution of learning outcomes, including:

[0056] Step 101: Obtain the platform's user dataset, and determine the learner's immediate learning performance score based on the learning outcomes and learning objectives in the user dataset.

[0057] In some implementations, learners' learning outcomes on MOOC learning platforms are significantly correlated with video viewing time. Therefore, learners' instant ratings should reflect their learning outcomes. According to mastery learning theory, feedback formed by comparing learning outcomes with learning objectives constitutes the learner's goal feedback. This disclosure uses the distribution information of course videos within a specific time period as the learning objective, specifically watching complete videos within that time period; it uses the learner's video completion status within that time period as the learning outcome, specifically a comprehensive rating based on the learner's video viewing time and number of viewing sessions; and it performs instant evaluation of the learner's learning outcomes based on the distribution information and the video completion status to determine the learner's instant learning outcome rating.

[0058] Furthermore, the platform's user dataset is obtained, and the learners' learning outcomes are evaluated in real time based on the learning outcomes and learning objectives in the user dataset. The calculation of the learner's real-time learning outcome score is performed through the following steps:

[0059] Step (1): First, let a specific time period be T, and T be divided into s different time slices T = (t1, t2, ..., t3). s Let the time slice interval be σ; let course c occur in a certain time slice t. b (t b ∈(t1, t sThe time period includes n videos, and the total duration of all videos is represented as... Let the time for a learner to register for a course be t. a (t a ∈[t1,t b )), it is in [t a , t b The duration of video viewing within a given time period is expressed as follows: Where (o≤n); Let the learner watch the i-th video v i The number of views is m (i∈[1,0]), and the viewing time for each video is m / 2. Then the average viewing time of the learner for the i-th video is

[0060] Step (2), calculate the number of times the learner has watched the video, [t a , t b During the time period, there were a total of If there are several time slices, then the average number of study sessions per time slice is:

[0061] Step (3) calculate the learner's response to the i-th video v i The completion rate is The average completion rate of learners within time T is: Course completion rate:

[0062] Step (4): Calculate the percentage of each completed video segment out of all completed videos. Its vector representation is in The closer the distribution is to a uniform distribution, the higher the knowledge conversion rate.

[0063] Step (5), record the uniform distribution. For vectors Conversion rate k is defined as and Cosine similarity:

[0064] Step (6), the learner in time slice t b The instant rating is Where μ is the average number of learning sessions per time slice, θ is the course completion rate (proportion), and k is the conversion rate. Let D be the average completion rate of videos within a specific time period; then the learner's instant rating at different time slices within time period T can be expressed as D = (d1, d2, ..., d...). s ).

[0065] Step 102: Obtain valid course selection information for all learners, and determine the dynamic relationship network based on the valid course selection information and the real-time learning performance score.

[0066] In some implementations of MOOC education, learners' course selection information and learning outcomes can reveal their learning preferences. Other learners with similar course selections represent those with similar learning preferences (e.g., students in related majors often choose similar courses). The higher the similarity between two learners, the more similar their learning situations are, and the more likely they are to be influenced by each other.

[0067] In some implementations, obtaining all learners' valid course selection information and real-time learning performance scores to determine a dynamic relationship network includes: obtaining course offering information, course selection information, and course dropout information; determining the valid course selection information; determining the learning similarity relationship between the target learner and all other learners at a specific time; determining an adjacency matrix based on the learning similarity relationship; and constructing the dynamic relationship network based on the adjacency matrix, with the target learner as a network node.

[0068] Specifically, the calculation of the dynamic relationship network involves obtaining valid course selection information for all learners, and then using this information and real-time learning performance assessments to determine the dynamic relationship network. This is done through the following steps:

[0069] Step (1), firstly, in a certain time slice t b The courses already offered are represented as (c1, c2, ..., c n ); whether any learner selects a course is represented by (0 / 1), if their selected course c i (i∈[1,n]), then c i =1; otherwise, c i =0; the learner's course selection list is represented as a vector using one-hot encoding. For example, learner U selects courses C2 and C n Then learner u's course selection list is

[0070] Step (2): Calculate t using the methods from steps (1) to (6) in step 101. b Instant ratings for all courses selected by Time Learner u

[0071] Step (3): Since learners may drop out of courses after selection, it is necessary to determine whether their selected courses have analytical value: set a threshold M1 and calculate the list of valid course selections. like If the element is greater than the threshold M1, then... The element is set to 1 otherwise; the element represents the ratio of similarity in learning situations.

[0072] Step (4), calculate t b Always up to the valid course selection list for all learners Calculate the cosine similarity between the target learner and the effective course selection lists of all other learners, resulting in a two-dimensional matrix. The formula for calculating cosine similarity is:

[0073]

[0074] Where A is the target learner and B is all other learners;

[0075] Table 1. Two-dimensional matrix of learner similarity

[0076]

[0077] Setting all diagonal elements of the similarity matrix to 0 results in a symmetric matrix with all diagonal elements set to 0.

[0078] Step (5): Set a threshold M2. If the value of a matrix element is greater than M, set the value of that element to 1; otherwise, set it to 0. Use the resulting new matrix as the adjacency matrix. b Learners at any time act as network nodes to construct an unweighted and undirected graph G. b The dynamic relationship network at different time slices within time period T can then be represented as G = (G1, G2, ..., G...). s ).

[0079] Step 103: Obtain the clustering characteristic community structure in the dynamic relationship network, and determine the evolution of learning situation based on the clustering characteristic community structure.

[0080] In some implementations, researchers have proposed numerous community detection algorithms on graph G to discover community structures on the network. However, real-world networks often have overlapping structures, meaning a node may belong to multiple communities simultaneously. Traditional faction filtering algorithms find all subgraphs of size k on graph G and then merge connected subgraphs (two subgraphs share k-1 nodes). Each set of k subgraphs after merging represents a community. However, traditional faction filtering algorithms cannot assign all nodes in the network, making it difficult to compare community structures across adjacent time slices. To address the overlap issues in real-world networks, this disclosure employs an improved faction filtering algorithm to detect community structures in MOOC relationship networks. Specifically, it uses a local merging approach, continuously merging subgraphs from large to small based on the k value until all nodes are assigned to communities, resulting in a community structure more suitable for subsequent dynamic network analysis.

[0081] Furthermore, this invention employs the overlapping community module degree Q. e To evaluate the quality of community testing, Q e A higher value indicates a more rational community structure. (Q represents the degree of overlap in community modules.) e for:

[0082] Where m is the number of edges in the network, C is the community, and A is the number of communities. ij Let A be an element in the adjacency matrix A (if nodes i and j have edges, then A...). ij =1, otherwise A ij =0), C i C represents the community to which node i belongs. j This indicates the community to which node j belongs. If nodes i and j belong to the same community, then δ(C) i C j ) = 1, otherwise δ(C) i C j ) = 0, O i O j d represents the number of communities including nodes i and j, and d represents the instant rating.

[0083] In some implementations, researchers have found that relational networks consist of isolated clusters of nodes, and that nodes exhibit a critical probability of forming clusters. Therefore, relational networks also possess a critical probability, where the critical probability threshold is denoted as p. c When the probability does not exceed the critical probability threshold, it means that the network consists of isolated clusters of nodes. However, when the probability exceeds the critical probability threshold, it means that the clusters of nodes will expand and connect to the entire network.

[0084] In MOOC networks, critical probability thresholds include thresholds M1 and M2. M1 represents the threshold for immediate ratings of effective courses. When M1 reaches a certain threshold, the community structure on the network changes significantly, while other values ​​of M1 show relatively stable changes. Analysis of the M1 threshold provides a reference value for immediate ratings; most learners' immediate ratings are greater than the threshold M1, serving as a timely warning for learners with ratings below M1. M2 also represents the threshold for learner course selection similarity. Analyzing the M1 threshold reveals the impact of learner course selection similarity on the network structure. When M2 reaches a critical value, the community structure on the network becomes clearly apparent, indicating a high degree of similarity in course selections among learners, thus providing suggestions for course recommendations within the community.

[0085] In some implementations, nodes in the dynamic relationship network are adjusted based on clustering characteristics and community structure; the community structure of the dynamic relationship network is adjusted using a faction filtering algorithm; and for each community in the adjusted community structure, an evolutionary sequence corresponding to that community is constructed (evolutionary sequence reference). Figure 3 The process involves: identifying target communities from the community structure; defining adjacent communities as neighboring communities; comparing the evolutionary sequence of the target community with the evolutionary sequences of the neighboring communities to determine the community evolutionary morphology of the target community; and determining the evolution of learning conditions based on the community evolutionary morphology. Figure 3 The diagram below is a reference diagram of the evolution sequence of embodiments of this disclosure, which includes n communities, (E1, E2, ..., E...). n () represents the community number, E represents the community, and the subscripts 1, 2, n are the serial numbers, (t1, t2, ..., t s The equation (t) represents multiple time points, each corresponding to an evolutionary sequence as shown in the figure. Except for the first evolutionary sequence, all other evolutionary sequences are derived from the previous one. For example, in community E1, the evolutionary sequence at time t1 includes three nodes: 1, 2, and 3. At time t2, the evolutionary sequence corresponding to community E1 adds nodes 4 and 5, including five nodes: 1, 2, 3, 4, and 5. Depending on the changes in community structure, two evolutionary sequences may combine into one, and one evolutionary sequence may split into two. For example, the evolutionary sequences of communities E1 and E2 at time t... s The moments merge into an evolutionary sequence, while community E n The evolutionary sequence in t s It splits into two evolutionary sequences at a certain point.

[0086] Furthermore, adjusting the community structure of the dynamic relationship network using the faction filtering algorithm includes: detecting the community structure of the dynamic relationship network to obtain the community distribution of subgraphs in the dynamic relationship network; adjusting the order of the subgraphs according to their size; merging the adjusted subgraphs and assigning the merged subgraphs to communities to obtain the adjusted community structure.

[0087] In some implementations, the dynamic network structure is analyzed based on the clustering characteristics and community structure to determine the evolution of learning outcomes, specifically including the following steps:

[0088] First, construct the evolutionary sequence based on the adjusted community structure: establish the evolutionary sequence E for each community in C1. x =(C 1x C 2x ,...,C sx),(x∈[1,n]). Set a threshold M3, and set C i Each community structure is compared for similarity with the community at the end of each evolutionary cohort. The formula for calculating community similarity is:

[0089]

[0090] If the similarity is greater than the threshold M3, the community is added to the evolutionary sequence; if there is no matching evolutionary sequence, a new evolutionary sequence is created.

[0091] Then, community evolution pattern analysis: for evolutionary queue E i By comparing the community evolution patterns at adjacent time points, the experiment revealed that the dynamic evolution of communities can be categorized into six types: community emergence, community growth, community shrinkage, community merger, community splitting, and community disappearance. By observing the evolution of the learner's community within time period T, the evolution of their learning can be determined.

[0092] As described above, the learning progress analysis method, apparatus, electronic device, and storage medium disclosed herein firstly assess learners' learning outcomes in real-time based on course distribution information and viewing activity; then, it obtains valid course selection information for all learners on the platform and determines a dynamic relationship network based on this information and the real-time learning outcome assessment; finally, it obtains the clustering characteristics and community structure within the dynamic relationship network and determines the learning progress based on this structure. This solution, by determining the learning progress based on the dynamic network established through real-time learning outcome assessment, effectively grasps the learning progress patterns of open online courses, thereby enabling real-time monitoring of the learning status of all users on the platform. This plays a crucial role for educators in real-time evaluation of teaching quality, control of teaching direction, course recommendation, and learning progress early warning applications.

[0093] It should be noted that the method of this disclosure embodiment can be executed by a single device, such as a computer or server. The method of this embodiment can also be applied to a distributed scenario, where multiple devices cooperate to complete the task. In such a distributed scenario, one of these devices may execute only one or more steps of the method of this disclosure embodiment, and the multiple devices will interact with each other to complete the method described.

[0094] It should be noted that the above description describes some embodiments of this disclosure. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in a different order than that shown in the above embodiments and still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0095] Based on the same inventive concept, corresponding to any of the above embodiments, this disclosure also provides a learning evolution analysis device.

[0096] refer to Figure 2 The learning progress analysis device includes:

[0097] The evaluation module is configured to acquire the platform's user dataset and determine an instant score for the learner's learning performance based on the learning outcomes and learning objectives in the user dataset.

[0098] The module is configured to acquire all learners’ valid course selection information, and determine a dynamic relationship network based on the valid course selection information and the learning effect in real time.

[0099] The analysis module is configured to obtain the clustering characteristics and community structure in the dynamic relationship network, and determine the evolution of learning status based on the clustering characteristics and community structure.

[0100] For ease of description, the above apparatus is described in terms of its functions, divided into various modules. Of course, in implementing this disclosure, the functions of each module can be implemented in one or more software and / or hardware.

[0101] The apparatus described above is used to implement the corresponding learning situation evolution analysis method in any of the foregoing embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0102] Based on the same inventive concept, corresponding to the methods of any of the above embodiments, this disclosure also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the learning situation evolution analysis method described in any of the above embodiments.

[0103] Figure 4This embodiment illustrates a more specific hardware structure of an electronic device. The device may include a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040, and a bus 1050. The processor 1010, memory 1020, input / output interface 1030, and communication interface 1040 are interconnected internally via the bus 1050.

[0104] The processor 1010 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this specification.

[0105] The memory 1020 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 1020 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented by software or firmware, the relevant program code is stored in the memory 1020 and is called and executed by the processor 1010.

[0106] The input / output interface 1030 is used to connect input / output modules to realize information input and output. Input / output modules can be configured as components within the device (not shown in the figure) or externally connected to the device to provide corresponding functions. Input devices may include keyboards, mice, touchscreens, microphones, various sensors, etc., while output devices may include displays, speakers, vibrators, indicator lights, etc.

[0107] The communication interface 1040 is used to connect a communication module (not shown in the figure) to enable communication between this device and other devices. The communication module can communicate via wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0108] Bus 1050 includes a pathway for transmitting information between various components of the device (e.g., processor 1010, memory 1020, input / output interface 1030, and communication interface 1040).

[0109] It should be noted that although the above-described device only shows the processor 1010, memory 1020, input / output interface 1030, communication interface 1040, and bus 1050, in specific implementations, the device may also include other components necessary for normal operation. Furthermore, those skilled in the art will understand that the above-described device may only include the components necessary for implementing the embodiments of this specification, and not necessarily all the components shown in the figures.

[0110] The electronic devices described above are used to implement the corresponding learning situation evolution analysis methods in any of the foregoing embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0111] Based on the same inventive concept, corresponding to the methods of any of the above embodiments, this disclosure also provides a non-transitory computer-readable storage medium storing computer instructions for causing the computer to execute the learning situation evolution analysis method as described in any of the above embodiments.

[0112] The computer-readable medium of this embodiment includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transfer medium that can be used to store information accessible by a computing device.

[0113] The computer instructions stored in the storage medium of the above embodiments are used to cause the computer to execute the learning situation evolution analysis method as described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0114] Those skilled in the art should understand that the discussion of any of the above embodiments is merely exemplary and is not intended to imply that the scope of this disclosure (including the claims) is limited to these examples; within the framework of this disclosure, the technical features of the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations of different aspects of the embodiments of this disclosure as described above, which are not provided in detail for the sake of brevity.

[0115] Additionally, to simplify the description and discussion, and to avoid obscuring the embodiments of this disclosure, the provided drawings may or may not show well-known power / ground connections to integrated circuit (IC) chips and other components. Furthermore, the apparatus may be shown in block diagram form to avoid obscuring the embodiments of this disclosure, and this also takes into account the fact that the details of implementation of these block diagram apparatuses are highly dependent on the platform on which the embodiments of this disclosure will be implemented (i.e., these details should be fully understood by those skilled in the art). While specific details (e.g., circuitry) have been set forth to describe exemplary embodiments of this disclosure, it will be apparent to those skilled in the art that the embodiments of this disclosure may be implemented without these specific details or with variations thereof. Therefore, these descriptions should be considered illustrative rather than restrictive.

[0116] Although this disclosure has been described in conjunction with specific embodiments thereof, many substitutions, modifications, and variations of these embodiments will be apparent to those skilled in the art from the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) may be used with the embodiments discussed.

[0117] This disclosure is intended to cover all such substitutions, modifications, and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A method for analyzing evolution of learning situation, comprising: obtaining a user data set of a platform, determining an instant score of learning effect of a learner according to learning results and learning goals in the user data set; wherein the learning goals comprise watching a complete video in a target time period, and the learning results comprise a watching time length in a specific time period and a learning frequency score; determining the instant score of learning effect of the learner according to the learning results and the learning goals in the user data set comprises: Step (1), first set a certain period of time for T, T is divided into s different time slices T= (t1, t2, …, t s ), time slice interval period is σ; set course c at a certain time slice moment includes n video, all video total duration is expressed as ; set the learner registration course time , its watching video duration in [t a， t b ] time period is expressed as , wherein ; set the learner watching the i-th video watching frequency is m, each video watching time is , then the learner watching the i-th video average duration is ; Step (2), calculate the number of times of video learning of the learner, [t a， t b The total number of time slices in the time period is The average number of times of learning of the learner in each time slice is ; Step (3) Calculate the completion rate of the learner for the i-th video is Then the average completion rate of the learner for videos in T is ; the course completion rate is ; Step (4), calculate the ratio of the completed proportion of each video to the total completed videos whose vector representation is where The closer to the uniform distribution, the higher the knowledge conversion rate; Step (5), note that the uniform distribution is satisfied For vector , the conversion rate k is defined as The cosine similarity of : ; Step (6), the instant score of the learner at time slice t b is where, is the average number of learning times on each time slice, is the course completion rate, is the conversion rate, is the average completion rate of the video in a specific time period; then the instant score of the learner at different time slices in the time period T can be expressed as ; obtaining effective course selection information of all learners, determining a dynamic relationship network according to the effective course selection information and the instant score of learning effect; obtaining a community structure with clustering characteristics in the dynamic relationship network, determining an evolution of learning situation according to the community structure with clustering characteristics; wherein the determination comprises detecting and adjusting nodes in the dynamic relationship network according to the community structure with clustering characteristics, and evaluating quality of community detection by using an overlapping community modularity Qe as shown below: where m is the number of edges in the network, C is a community, A ij is an element in the adjacency matrix A, A ij = 1 if nodes i, j have an edge, otherwise A ij = 0, C i represents the community to which node i belongs, C j represents the community to which node j belongs, δ(C i , C j ) = 1 if nodes i, j belong to the same community, otherwise δ(C i , C j ) = 0, O i , O j is the number of communities containing nodes i, j, and d is the immediate score. detecting rationality of the dynamic relationship network to determine the evolution of learning situation; wherein the determination comprises: detecting and adjusting a community structure of the dynamic relationship network by using a faction filtering algorithm, comprising: detecting the community structure of the dynamic relationship network to obtain a community distribution of subgraphs in the dynamic relationship network; adjusting an order of the subgraphs according to sizes of the subgraphs; merging the adjusted subgraphs and distributing the merged subgraphs to communities to obtain the adjusted community structure; constructing an evolution sequence corresponding to each of the adjusted community structures; determining a target community from the community structure, and determining a neighboring community adjacent to the target community; comparing the evolution sequence corresponding to the target community with an evolution sequence corresponding to the neighboring community to determine a community evolution pattern of the target community; determining the evolution of learning situation according to the community evolution pattern; wherein the community evolution pattern comprises community appearance, community growth, community atrophy, community merger, community split and community disappearance.

2. The analytical method of claim 1, wherein, the learning goals comprise distribution information of course videos in a specific time period, the learning results are video completion conditions of the learners in the specific time period; the obtaining of the user data set of the platform and the determining of the instant score of learning effect of the learner according to the learning results and the learning goals in the user data set comprise: instantly evaluating learning effect of the learner according to the distribution information and the video completion conditions to determine the instant score of learning effect of the learner.

3. The analytical method of claim 1, wherein, the obtaining of the effective course selection information of all learners and the instant score of learning effect and the determining of the dynamic relationship network comprise: obtaining course offering information, course selection information and course dropping information to determine the effective course selection information; determining a learning situation similar relationship between a target learner and all other learners at a specific time; determining the dynamic relationship network according to the learning situation similar relationship.

4. The analytical method of claim 3, wherein, the determining of the dynamic relationship network according to the learning situation similar relationship comprises: determining an adjacency matrix according to the learning situation similar relationship; The target learner is taken as a network node, and the dynamic relationship network is constructed according to the adjacency matrix.

5. A learning situation evolution analysis device, comprising: An evaluation module configured to obtain a user data set of a platform, determine an instant score of learning effect of a learner according to learning results and learning goals in the user data set; Wherein the learning goals include watching a complete video within a target time period, and the learning results include watching time length and learning frequency score within a specific time period; determining the instant score of learning effect of the learner according to the learning results and the learning goals in the user data set includes: Step (1): First, let a specific time period be T, and T be divided into s different time slices T = (t1, t2, ..., t3). s The time slice interval is σ; let course c be in a certain time slice. The time period includes n videos, and the total duration of all videos is represented as . Let the learner's course registration time be... , it is in [t a， t b The duration of video viewing within a given time period is expressed as follows: ,in Let the learner watch the i-th video. The number of views is m, and the viewing time of each video is m. Then the average viewing time of the learner for the i-th video is ; Step (2), calculate the number of times of video learning of the learner, [t a， t b ] total number of time slices in the time period, then the average number of times of learning of the learner in each time slice is ;​ Step (3) Calculate the completion rate of the learner for the i-th video is Then the average completion rate of the learner for videos in T is The course completion rate is ; Step (4), calculate the ratio of the completed proportion of each video to the total completed videos whose vector representation is where The closer to the uniform distribution, the higher the knowledge conversion rate; Step (5), note that the uniform distribution is satisfied For the vector The conversion rate k is defined as The cosine similarity of : ; Step (6), the instant score of the learner at time slice t b is where, is the average number of learning times on each time slice, is the course completion rate, is the conversion rate, is the average completion rate of the video in a specific time period; then the instant score of the learner at different time slices in the time period T can be expressed as ; A construction module configured to obtain effective course selection information of all learners, determine a dynamic relationship network according to the effective course selection information and the instant score of learning effect; An analysis module configured to obtain a community structure with clustering characteristics in the dynamic relationship network, determine a learning situation evolution condition according to the community structure with clustering characteristics; wherein the analysis module includes detecting and adjusting nodes in the dynamic relationship network according to the community structure with clustering characteristics, and using an overlapping community module degree Qe to evaluate the quality of community detection as shown below: where m is the number of edges in the network, C is a community, A ij is an element in the adjacency matrix A, A ij = 1 if nodes i, j have an edge, otherwise A ij = 0, C i represents the community to which node i belongs, C j represents the community to which node j belongs, δ(C i , C j ) = 1 if nodes i, j belong to the same community, otherwise δ(C i , C j ) = 0, O i , O j is the number of communities containing nodes i, j, and d is the immediate score. Detecting the rationality of the dynamic relationship network to determine the learning situation evolution condition; wherein the analysis module includes: Detecting and adjusting the community structure of the dynamic relationship network through a faction filtering algorithm, including detecting the community structure of the dynamic relationship network to obtain community distribution of subgraphs in the dynamic relationship network, adjusting the order of the subgraphs according to the size of the subgraphs, merging the adjusted subgraphs, and distributing the merged subgraphs to communities to obtain the adjusted community structure; For each community in the adjusted community structure, constructing an evolution sequence corresponding to the community; Determining a target community from the community structure, and determining a neighboring community adjacent to the target community; Comparing the evolution sequence corresponding to the target community with the evolution sequence corresponding to the neighboring community to determine a community evolution pattern of the target community; Determining the learning situation evolution condition according to the community evolution pattern; wherein the community evolution pattern includes community appearance, community growth, community atrophy, community merger, community split, and community disappearance.

6. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor implements the method of any one of claims 1 to 4 when executing the program.

7. A non-transitory computer-readable storage medium, comprising: The non-transitory computer readable storage medium stores computer instructions for causing a computer to execute the method of any one of claims 1 to 4.

Images