Online education method and system based on block chain

Abstract

The invention belongs to the technical field of online education, and particularly relates to an online education method and system based on a block chain. According to the method, the identity information, the learning behavior data and the course resource data of the learner are stored by constructing the alliance chain network, multi-dimensional verification is performed by using the smart contract, the authenticity and the integrity of the data are ensured, and the learning report containing the timestamp is generated by performing quantitative processing on the interactive behavior data, so that the learning efficiency is improved. According to the method, comprehensive and accurate learning feedback is provided for the learner, in addition, personalized learning path recommendation is carried out by combining historical learning performance of the learner and utilizing a course knowledge graph, the learning effect and satisfaction are effectively improved, and finally, consensus verification is carried out on learning behavior data and course resource data through a consensus mechanism of a block chain, so that the learning efficiency is improved. And public and transparent authentication is carried out on the learning report of the learner, so that the credibility and the security of the data are further enhanced.

Description

Technical Field

[0001] The present invention belongs to the technical field of online education, and particularly relates to an online education method and system based on blockchain. Background Art

[0002] With the continuous development of information technology, in the field of education, it has gradually changed from the traditional education mode to a mode combined with online education. Online education is favored by a large number of learners for its flexible and convenient characteristics. However, there are still some problems in aspects such as the sharing of educational resources, the storage of learning data, and the certification of learning achievements. Traditional online education platforms often adopt a centralized data storage method, which leads to doubts about the security and credibility of data. At the same time, the learning behavior data and course resource data of learners are easily tampered with or deleted. The decentralized, tamper-proof, and transparent and public characteristics of blockchain technology provide new ideas for solving these problems. By applying blockchain technology to the field of online education, reliable sharing of educational resources, secure storage of learning data, and fair certification of learning achievements can be realized, thereby further improving the quality and efficiency of online education.

[0003] In the prior art, although there are some online education platforms applying blockchain technology, these platforms still have deficiencies in aspects such as data processing, learning path recommendation, and learning achievement certification. For example, when processing the learning behavior data of learners, the dynamics and real-time nature of the data are often ignored, and only static processing rules preset are followed, without fully considering the changes in the behavior characteristics of learners at different learning stages. This will undoubtedly lead to non-personalized learning path recommendations and unable to meet the actual needs of learners, thereby affecting the learning experience and effect of learners. Based on this, the present invention proposes an online education method based on blockchain to solve the above problems. Summary of the Invention

[0004] The purpose of the present invention is to provide an online education method and system based on blockchain, which can dynamically process the learning behavior data and course resource data of learners, and at the same time, combined with blockchain technology, ensure the security and credibility of data and the fair certification of learning achievements.

[0005] The technical solutions adopted by the present invention are specifically as follows:

[0006] An online education method based on blockchain, comprising:

[0007] Constructing a consortium blockchain network, and uploading the identity information, learning behavior data, and course resource data of learners to the consortium blockchain network for storage;

[0008] Multi-dimensionally verify the interaction events of learners through smart contracts, and generate a learning report including timestamps based on the verification results;

[0009] Based on the learning report and combined with the learner's historical learning performance, recommend a personalized learning path for the learner;

[0010] Introduce the consensus mechanism of the blockchain to conduct consensus verification on learning behavior data and course resource data, and conduct publicly transparent authentication on the learning reports of learners.

[0011] In a preferred solution, the consortium blockchain network includes three types of participating entities: educational institution nodes, regulatory nodes, and learner nodes. Among them, educational institution nodes are responsible for providing course resource data, regulatory nodes are responsible for supervising the behaviors of educational institutions and learners, and learner nodes are used to upload and query personal learning behavior data.

[0012] In a preferred solution, the smart contract is automatically triggered after the learner logs in, monitors the learner's interaction behaviors on the learning platform in real time, and outputs them as a multi-dimensional interaction behavior dataset;

[0013] Add timestamps to the interaction behavior data in the multi-dimensional interaction behavior dataset to generate an immutable learning record.

[0014] In a preferred solution, the steps of multi-dimensionally verifying the interaction events of learners through smart contracts and generating a learning report including timestamps based on the verification results include:

[0015] Obtain all the interaction behavior data of the learner on the consortium blockchain network, and the interaction behavior data includes learning time periods, course chapter jump paths, and exercise interaction response times;

[0016] Verify the interaction behavior data according to preset multi-dimensional verification rules, and quantify the verification results of each interaction behavior data as learning performance scores;

[0017] Perform fusion processing on the learning performance scores corresponding to each interaction behavior data to obtain the learner's comprehensive learning performance score;

[0018] Compare the comprehensive learning performance score with a preset evaluation threshold. When the comprehensive learning performance score is lower than the evaluation threshold, trigger an early warning mechanism and send a review request to the regulatory node. Otherwise, synchronize the interaction behavior data with the blockchain timestamp service and synchronously generate a learning report;

[0019] Encrypt and shard the storage of the learning report through the distributed storage nodes of the consortium blockchain network, and at the same time return verifiable digital vouchers to the educational institution nodes and learner nodes.

[0020] In a preferred embodiment, the step of validating the interaction behavior data according to a preset multi-dimensional validation rule and quantifying the validation results of each interaction behavior data into a learning performance score includes:

[0021] Collect the learning time period of the learner and the opening time of the learning course, calculate the matching degree between the learning time period and the possible opening time of learning, and record it as the first characteristic parameter;

[0022] Obtain the chapter learning order of the learner and the preset curriculum syllabus, calculate the degree of fit between the chapter learning order and the curriculum syllabus, and record it as the second characteristic parameter;

[0023] Calculate the deviation amount between the answering correct rate of the learner during the exercise interaction process and the historical correct rate during the interaction process of the same type of exercises, and record it as the third characteristic parameter;

[0024] Perform normalization processing on the first characteristic parameter, the second characteristic parameter, and the third characteristic parameter to obtain the learning performance scores of each interaction behavior data under the same dimension.

[0025] In a preferred embodiment, the step of performing fusion processing on the learning performance scores corresponding to each interaction behavior data to obtain the comprehensive learning performance score of the learner includes:

[0026] Collect the learning performance scores of each interaction behavior data and perform initial weight assignment;

[0027] According to the curriculum progress stage of the learner, extract the recent interaction behavior data fluctuation characteristics through a sliding time window to generate a dynamic weight adjustment factor;

[0028] Combined with the real-time difficulty coefficient of the curriculum resources, differentially correct the weights of the first characteristic parameter, the second characteristic parameter, and the third characteristic parameter;

[0029] Based on the historical weight assignment records stored in the blockchain, predict the optimal weight combination for the current curriculum stage through linear regression;

[0030] According to the optimal weight combination, perform weighted summation on the learning performance scores of each interaction behavior data, and output the comprehensive learning performance score of the learner.

[0031] In a preferred embodiment, the step of recommending a personalized learning path for the learner based on the learning report and combining the learner's historical learning performance includes:

[0032] Obtain the historical learning performance of the learner stored in the blockchain, and the historical learning performance includes knowledge mastery, learning efficiency characteristics, and curriculum progress deviation;

[0033] Vectorize the knowledge mastery level, learning efficiency characteristics, and curriculum progress deviation to generate a three-dimensional feature vector reflecting historical learning performance;

[0034] Call the preset curriculum knowledge graph, match the three-dimensional feature vector with the curriculum nodes in the curriculum knowledge graph, and generate an initial recommended path;

[0035] Obtain a historical successful learning path dataset from the regulatory nodes in the consortium chain network. Among them, the historical successful learning path dataset contains the learner identity identification, path trajectory hash value, and key learning behavior characteristic parameters for completing the curriculum objectives;

[0036] Obtain the interaction behavior data of the current learner, generate a standardized feature set, and calculate the matching degree index between the initial recommended path and the historical successful path based on the standardized feature set and the historical successful learning path dataset;

[0037] When the matching degree index is lower than the preset reconstruction threshold, trigger path reconstruction. The specific process of path reconstruction includes:

[0038] Identify redundant paths in the historical successful path through the curriculum knowledge graph, and eliminate the redundant paths to obtain multiple candidate paths;

[0039] Calculate the knowledge correlation degree score between each candidate path and the initial recommended path, and use the candidate path with the highest knowledge correlation degree score as the personalized recommended path.

[0040] In a preferred solution, the step of introducing the consensus mechanism of the blockchain to perform consensus verification on learning behavior data and curriculum resource data, and perform public and transparent authentication on the learning reports of learners includes:

[0041] Perform a hash operation on the learning behavior data and curriculum resource data to generate a unique feature identifier, and broadcast the feature identifier to the educational institution nodes, regulatory nodes, and learner nodes in the consortium chain network;

[0042] The educational institution nodes, regulatory nodes, and learner nodes perform consensus verification on the received feature identifier. After the consensus verification passes, pack the learning behavior data and curriculum resource data into a block and add it to the blockchain;

[0043] The educational institution and the regulatory party add digital signatures to the learning performance scores and learning achievements in the learning report, generate an anti-counterfeiting authentication label, and store it in the blockchain.

[0044] The present invention also provides an online education system based on the blockchain, using the above-mentioned online education method based on the blockchain, including:

[0045] A resource on-chain storage module, which is used to build a consortium chain network and upload the identity information, learning behavior data, and course resource data of learners to the consortium chain network for storage;

[0046] A verification module, which is used to perform multi-dimensional verification on the interaction events of learners through a smart contract and generate a learning report containing a timestamp based on the verification results;

[0047] A personalized recommendation module, which is used to recommend a personalized learning path for learners based on the learning report and in combination with the historical learning performance of the learners;

[0048] A consensus verification module, which is used to introduce the consensus mechanism of the blockchain to perform consensus verification on the learning behavior data and course resource data, and to perform open and transparent authentication on the learning reports of learners.

[0049] And, an electronic device, which includes:

[0050] At least one processor;

[0051] And a memory communicatively connected to the at least one processor;

[0052] Wherein, the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the above-mentioned blockchain-based online education method.

[0053] The technical effects achieved by the present invention are as follows:

[0054] By introducing blockchain technology, the present invention realizes the transparent, secure, and trustworthy management of online education data. Specifically, by building a consortium chain network, storing the identity information, learning behavior data, and course resource data of learners, and using a smart contract for multi-dimensional verification, the authenticity and integrity of the data are ensured. At the same time, by quantifying the interaction behavior data, a learning report containing a timestamp is generated, providing comprehensive and accurate learning feedback for learners. In addition, the present invention also combines the historical learning performance of learners and uses a course knowledge graph to recommend personalized learning paths, effectively improving the learning effect and satisfaction. Finally, through the consensus mechanism of the blockchain, consensus verification is performed on the learning behavior data and course resource data, and open and transparent authentication is performed on the learning reports of learners, further enhancing the credibility and security of the data. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 It is a schematic flowchart of the invention process of the present invention;

[0056] Figure 2 is a schematic diagram of the system module of the present invention;

[0057] Figure 3 is a schematic diagram of the structure of the electronic device of the present invention. Specific Embodiments

[0058] To make the above objects, features, and advantages of the present invention more obvious and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings of the specification.

[0059] In the following description, many specific details are set forth to facilitate a thorough understanding of the present invention. However, the present invention may be implemented in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0060] Secondly, the so-called "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The appearances of "in a preferred embodiment" in different places in this specification do not all refer to the same embodiment, nor are they separate or alternative embodiments that exclude each other with other embodiments.

[0061] Please refer to Figure 1 As shown, the present invention provides an online education method based on blockchain, including:

[0062] S1. Construct a consortium blockchain network, and upload the identity information, learning behavior data, and course resource data of learners to the consortium blockchain network for storage;

[0063] In the step S1, online education refers to educational activities carried out through a network platform, which breaks the geographical and time limitations of traditional education, enabling learners to study at any time and place. Due to its convenience and flexibility, it has gradually received extensive attention. However, online education platforms need to process a large amount of learner data, including identity information, learning behavior data, and course resource data, etc. This solution constructs a consortium blockchain network jointly maintained by multiple participants, and securely uploads the learner's identity information, detailed learning behavior data, and rich course resource data to this consortium blockchain network for distributed storage to ensure the immutability and high security of the data. Among them, the consortium blockchain network includes three types of participating entities: educational institution nodes, regulatory nodes, and learner nodes. Among them, educational institution nodes are responsible for providing course resource data, specifically publishing and updating course resources such as teaching courseware and video lectures, and uploading various teaching materials and data to the network platform to ensure the richness and timeliness of resources. Regulatory nodes are responsible for supervising the behaviors of educational institutions and learners, specifically monitoring the interaction process between educational institutions and learners, and auditing the compliance of educational quality and teaching activities to ensure the accuracy of information and the legality of operations. Learner nodes are used to upload and query personal learning behavior data, such as learning progress, grade records, etc., for personal learning management and data analysis.

[0064] S2. Perform multi-dimensional verification on the interaction events of learners through smart contracts, and generate a learning report containing a timestamp based on the verification results;

[0065] In the step S2, after a learner logs in to the online education platform, various interaction events will occur between the learner and the platform, such as logging in, watching learning videos, submitting homework, taking exams, etc. Use smart contracts to conduct comprehensive and multi-dimensional verification on various interaction events of the learner on the platform, and automatically generate a learning report containing an accurate timestamp based on the verification results to reflect the learner's learning process and achievements. Among them, the smart contract is automatically triggered after the learner logs in, monitors the interaction behaviors of the learner on the learning platform in real time, and outputs them as a multi-dimensional interaction behavior data set;

[0066] In order to ensure the authenticity and immutability of learning records, accurate timestamp information will be added to each piece of interaction behavior data in the multi-dimensional interaction behavior data set one by one. Add timestamps to the interaction behavior data in the multi-dimensional interaction behavior data set to generate immutable learning records.

[0067] Secondly, the steps of performing multi-dimensional verification on the interaction events of learners through smart contracts and generating a learning report containing a timestamp include:

[0068] Obtain all the interaction behavior data of the learner on the consortium blockchain network. The interaction behavior data includes learning time periods, course chapter jump paths, and exercise interaction response times;

[0069] Verify the interaction behavior data according to the preset multi-dimensional verification rules, and quantify the verification results of each interaction behavior data into learning performance scores;

[0070] Perform a fusion process on the learning performance scores corresponding to each interaction behavior data to obtain the learner's comprehensive learning performance score;

[0071] Compare the comprehensive learning performance score with the preset evaluation threshold. When the comprehensive learning performance score is lower than the evaluation threshold, trigger the warning mechanism and send a review request to the regulatory node. Otherwise, synchronize the interaction behavior data with the blockchain timestamp service and generate a learning report synchronously;

[0072] Encrypt and shard the storage of the learning report through the distributed storage nodes of the consortium blockchain network, and at the same time return verifiable digital vouchers to the educational institution node and the learner node;

[0073] Specifically, when comprehensively and multi-dimensionally verifying various interaction events of the learner during the learning process through a smart contract, it is first necessary to obtain all the interaction behavior data of the learner on the consortium blockchain network. The interaction behavior data covers multiple aspects, including but not limited to the learning activity records of the learner at different time periods, the jump paths between course chapters, and the interaction response times to exercises during the learning process, etc. Then, according to the preset multi-dimensional verification rules, the interaction behavior data will be verified one by one. During this verification process, the verification results of each interaction behavior data will be quantified and converted into specific learning performance scores for subsequent comprehensive evaluation. Then, a fusion process will be performed on the learning performance scores corresponding to each interaction behavior data, and the learner's comprehensive learning performance score will be obtained through comprehensive analysis. Among them, the expression of the comprehensive learning performance score is:

[0074]

[0075] In the formula, S represents the learner's comprehensive learning performance score, ω i represents the weight of the i-th verification dimension, v iDenote the verification score of the \(i\)-th verification dimension, and \(n\) represents the total number of verification dimensions. Based on the comprehensive learning performance score, it can comprehensively reflect the learning status and effect of the learner. Subsequently, the comprehensive learning performance score will be compared with a preset evaluation threshold. If the comprehensive learning performance score is lower than the evaluation threshold, the early warning mechanism will be immediately triggered, and a review request will be sent to the relevant regulatory nodes to promptly discover problems and intervene. On the contrary, if the comprehensive learning performance score reaches or exceeds the evaluation threshold, the interaction behavior data will be synchronized with the blockchain timestamp service. The timestamp synchronization function is:

[0076] H(t) = Hash(D event ||T nonce ||T UTC ) ;

[0077] In the formula, H(t) represents the event hash value with timestamp, D event represents the original data of the interaction event, T nonce represents the random number generated by the blockchain, T UTC represents the time encoding. Through the synchronization of the timestamp, it can ensure the authenticity and immutability of the data, and a detailed learning report will be generated synchronously. Finally, the generated learning report will be encrypted and sharded for storage through the distributed storage nodes of the consortium chain network to ensure the security and privacy of the data. At the same time, a verifiable digital certificate will be returned to the educational institution node and the learner node so that all parties can conveniently verify the authenticity and effectiveness of the learning report.

[0078] Next, the step of verifying the interaction behavior data according to the preset multi-dimensional verification rules and quantifying the verification results of each interaction behavior data into learning performance scores includes:

[0079] Collect the learning period of the learner and the opening time of the learning course, calculate the matching degree between the learning period and the possible opening time of learning, and record it as the first characteristic parameter;

[0080] Obtain the chapter learning order of the learner and the preset curriculum syllabus, calculate the degree of fit between the chapter learning order and the curriculum syllabus, and record it as the second characteristic parameter;

[0081] Calculate the deviation amount between the correct answer rate of the learner during the exercise interaction process and the historical correct answer rate during the interaction process of the same type of exercises, and record it as the third characteristic parameter;

[0082] Normalize the first characteristic parameter, the second characteristic parameter, and the third characteristic parameter to obtain the learning performance scores of each interaction behavior data under the same dimension;

[0083] Specifically, when validating the interaction behavior data of learners, the actual learning period of the learners and the opening time information of the corresponding learning courses are first collected. By calculating the matching degree between the learning period and the course opening time, and recording this matching degree as the first characteristic parameter. The calculation expression of the first characteristic parameter is as follows:

[0084]

[0085] In the formula, P1 represents the first characteristic parameter, N represents the number of learning days within the statistical period, represents the overlapping duration between the learning period and the opening time information of the learning course, T open represents the total daily opening duration of the course. For example, if a certain course is open for 8 hours every day, and the total overlapping duration of the learner's learning period and the opening period within a week is 20 hours for 5 days, then P1 = 20 / (5 × 8) = 0.5;

[0086] Then, the chapter learning order followed by the learner during the learning process and the preset course syllabus content are also obtained. The actual chapter jump path of the learner is recorded as [c1, c1,..., c m , and the preset path in the course syllabus is recorded as [C1, C1,..., C m . By comparing and analyzing, the matching degree between the chapter learning order and the course syllabus is calculated, and this matching degree is recorded as the second characteristic parameter to provide a basis for evaluating the learner's learning path. The calculation expression of the second characteristic parameter is as follows:

[0087]

[0088] In the formula, P2 represents the second characteristic parameter, Next(C i ) represents the set of legal subsequent chapters of C i in the course syllabus chapters, takes 1 when the condition is true, and 0 otherwise. For example, if there are 3 deviations from the syllabus order in the learner's jump path and the total number of jumps is 10 times, then P2 is approximately equal to 0.67,

[0089] After that, the answering accuracy rate of the learner during the exercise interaction process is statistically analyzed, and the historical accuracy rate during the interaction process of the same type of exercises is also referred to. By calculating the deviation amount between the current answering accuracy rate and the historical accuracy rate, and recording this deviation amount in detail as the third characteristic parameter to reflect the change situation of the learner's knowledge mastery. The calculation expression of the third characteristic parameter is as follows:

[0090]

[0091] In the formula, P3 represents the third characteristic parameter, R currentIt represents the correct rate of answering questions during the exercise interaction of the learner, and σ represents the standard deviation of the historical correct rate. It represents the average correct rate of historical exercises of the same type. For example, if the correct rate of answering questions is 80%, the average value of the historical average correct rate is 70%, and σ is 0.2, then the calculation result of P3 is 0.5.

[0092] Finally, the first feature parameter, the second feature parameter, and the third feature parameter collected will be normalized. Specifically, the minimum-maximum normalization method or the Z-score standardization method can be used to convert the values of each feature parameter into the range of 0 to 1, or into a distribution with a mean of 0 and a standard deviation of 1, so as to obtain the learning performance scores of each interaction behavior data under the same dimension, ensure that the first feature parameter, the second feature parameter, and the third feature parameter are compared and analyzed under the same dimension, so as to obtain the learning performance scores of each interaction behavior data under the unified standard, and provide data support for comprehensively evaluating the learning effect of the learner.

[0093] And the steps of fusing the learning performance scores corresponding to each interaction behavior data to obtain the comprehensive learning performance score of the learner include:

[0094] Collect the learning performance scores of each interaction behavior data and allocate initial weights.

[0095] According to the course progress stage of the learner, extract the fluctuation characteristics of recent interaction behavior data through a sliding time window to generate a dynamic weight adjustment factor.

[0096] Combined with the real-time difficulty coefficient of the course resources, differentially correct the weights of the learning performance scores of each interaction behavior data.

[0097] Based on the historical weight allocation records stored in the blockchain, predict the optimal weight combination for the current course stage through linear regression.

[0098] According to the optimal weight combination, perform weighted summation on the learning performance scores of each interaction behavior data, and the output is the comprehensive learning performance score of the learner.

[0099] In the above, after the learning performance scores corresponding to the interaction behavior data are determined, preliminary initial weight allocation will be performed on each learning performance score, and then according to the specific progress stage of the learner in the course, the sliding time window technology is used to extract the fluctuation characteristics of the interaction behavior data in the recent period, and a dynamic weight adjustment factor is generated based on the data fluctuation characteristics to adapt to the behavior changes of the learner in different learning stages. First, the length of the time window needs to be defined, and the time series data of the first feature parameter, the second feature parameter, and the third feature parameter within the window are extracted. Then calculate the mean value of the window data. Sum of variances Then, based on the mean and variance of the window data, the dynamic weight adjustment factor can be calculated. The calculation formula is as follows:

[0100]

[0101] In the formula, α represents the dynamic weight adjustment factor. After the output of the dynamic weight adjustment factor, combined with the real-time difficulty coefficient of the course resources, the weights of each learning performance score will be differentially corrected accordingly. Specifically, a course difficulty correction coefficient is introduced for correction. This correction process can be achieved by calculating the difficulty coefficient of each course resource and comparing it with the preset difficulty standard to determine the size of the course difficulty correction coefficient. If the difficulty coefficient of a certain course resource is higher than the preset standard, the weight corresponding to its learning performance score will be increased accordingly; otherwise, the weight will be decreased. For example, if the difficulty coefficient of a certain mathematics course chapter is higher than the average difficulty standard, it indicates that the content of this chapter is relatively profound, and more attention should be paid to the learning performance score of learners in this chapter. Therefore, its weight can be appropriately increased to 1.2 times. On the contrary, if the difficulty coefficient of a certain English course chapter is lower than the average difficulty standard, it indicates that the content of this chapter is relatively simple. Although the learning performance score of learners in this chapter is high, it may not be sufficient to represent their true learning level. Therefore, its weight can be appropriately reduced to 0.8 times. Through this kind of differential correction, the actual performance of learners on course resources with different difficulties can be more accurately reflected. Then, based on the historical weight allocation records stored by blockchain technology, using the method of linear regression analysis, the optimal weight combination for the current course stage is predicted to ensure the rationality of weight allocation. The prediction formula for the optimal weight is as follows:

[0102]

[0103] In the formula, W t represents the optimal weight of the interaction behavior data, r represents the total number of time steps, γ τ represents the time decay coefficient of the historical weight, W hist (τ) represents the weight allocation value of the τ-th historical stage stored by blockchain, W0 represents the initial weight, β represents the difficulty correction coefficient. Finally, based on the determined optimal weight combination, the learning performance scores of each interaction behavior data are weighted and summed, and the final result is output as the comprehensive learning performance score of the learner, so as to comprehensively reflect the learning status and effectiveness of the learner.

[0104] S3. Based on the learning report, combined with the historical learning performance of the learner, recommend a personalized learning path for the learner;

[0105] In step S3, after the learner's learning report is output, based on the generated learning report and combined with the learner's past historical learning performance, a personalized learning path will be recommended for the learner, providing more targeted learning suggestions and resource recommendations, enabling the learner to supplement knowledge and improve skills according to their own situation.

[0106] Among them, the steps of recommending a personalized learning path for the learner based on the learning report and combined with the learner's historical learning performance include:

[0107] Obtain the learner's historical learning performance stored in the blockchain, where the historical learning performance includes knowledge mastery, learning efficiency characteristics, and course progress deviation;

[0108] Perform vectorization processing on knowledge mastery, learning efficiency characteristics, and course progress deviation to generate a three-dimensional feature vector reflecting the historical learning performance;

[0109] Call the preset course knowledge graph, perform similarity matching between the three-dimensional feature vector and the course nodes in the course knowledge graph to generate an initial recommended path;

[0110] Obtain the historical successful learning path dataset from the regulatory nodes in the consortium chain network, where the historical successful learning path dataset contains the learner identity identification, path trajectory hash value, and key learning behavior feature parameters for completing the course objectives;

[0111] Obtain the interaction behavior data of the current learner, generate a standardized feature set, and calculate the matching degree index between the initial recommended path and the historical successful path based on the standardized feature set and the historical successful learning path dataset;

[0112] When the matching degree index is lower than the preset reconstruction threshold, trigger path reconstruction, where the specific process of path reconstruction includes:

[0113] Identify redundant paths in the historical successful path through the course knowledge graph and eliminate the redundant paths to obtain multiple candidate paths;

[0114] Calculate the knowledge correlation degree score between each candidate path and the initial recommended path, and use the candidate path with the highest knowledge correlation degree score as the personalized recommended path;

[0115] Specifically, after the learner's learning report is output, the historical learning performance records of the learner stored in the blockchain will be comprehensively obtained first. The historical learning performance records cover multi-dimensional information such as the learner's mastery degree in various knowledge fields, the efficiency characteristics during the learning process, and the deviation degree between the actual course progress and the expected progress. Then, corresponding vectorization processing will be performed on the knowledge mastery degree, learning efficiency characteristics, and course progress deviation degree, so as to output a three-dimensional feature vector that comprehensively reflects the learner's historical learning performance, providing a prerequisite data basis for subsequent path recommendation. Then, the pre-constructed and improved course knowledge graph will be called, and the generated three-dimensional feature vector will be subjected to corresponding similarity matching analysis with each course node in the course knowledge graph, and an initial recommended path that conforms to the learner's characteristics will be generated based on the matching results. Among them, the calculation formula for the corresponding similarity matching analysis between the three-dimensional feature vector and each course node in the course knowledge graph is:

[0116]

[0117] In the formula, S(V,U i ) represents the similarity score between the three-dimensional feature vector and the course node, cos(V,U i ) represents the cosine similarity between the three-dimensional feature vector and the course node, and λ1, λ2 represent the weight factors of different course types (for example, theoretical courses and practical courses, etc.);

[0118] Then, the historical successful learning path dataset will be obtained from the regulatory nodes in the consortium chain network. The historical successful learning path dataset records the learner identity identification, the hash value of the path trajectory, and the key behavior feature parameters during the learning process of the learners who have successfully completed the course objectives in the past. At the same time, the interactive behavior data of the current learner will be obtained in real time, and after being standardized, a feature set will be formed. Based on this, combined with the historical successful learning path dataset, the matching degree index between the initial recommended path and the historical successful path will be calculated. Among them, the calculation formula for the matching degree index is:

[0119]

[0120] In the formula, M represents the matching degree index between the initial recommended path and the historical successful path, F 初始 represents the feature set of the initial recommended path, F 历史uDenote the feature set of the \(u\)-th historical path. When the matching degree index is lower than the preset reconstruction threshold, the path reconstruction mechanism is immediately triggered. Specifically, redundant paths are identified and removed based on the curriculum knowledge graph, so as to filter out candidate paths that meet the requirements. Then, the knowledge association degree score between each candidate path and the initial recommended path is further calculated. The calculation formula is: Knowledge association degree score = \(\kappa\) coverage (the proportion of core knowledge points included in the candidate path) + \((1 - \kappa)\) complexity (the fitness of the candidate path difficulty to the learner's current ability), where \(\kappa\) represents the balance factor (set by the monitoring node according to the curriculum objective). After the knowledge association degree score is output, the candidate path with the highest knowledge association degree score is selected as the personalized recommended learning path that best meets the learner's needs.

[0121] S4. Introduce the consensus mechanism of the blockchain to conduct consensus verification on the learning behavior data and curriculum resource data, and conduct public and transparent authentication on the learner's learning report;

[0122] In step S4, after processing and analyzing the learner's learning behavior data and curriculum resource data, in order to ensure the accuracy and credibility of the data, introduce the consensus mechanism of the blockchain to conduct multi-party consensus verification on the learning behavior data and curriculum resource data to ensure the authenticity and reliability of the data. At the same time, conduct public and transparent authentication on the learner's learning report to enhance the credibility and public trust of the learning results. Among them, the steps of introducing the consensus mechanism of the blockchain to conduct consensus verification on the learning behavior data and curriculum resource data, and conducting public and transparent authentication on the learner's learning report include:

[0123] Perform a hash operation on the learning behavior data and curriculum resource data to generate a unique feature identifier, and broadcast the feature identifier to the educational institution nodes, regulatory nodes, and learner nodes in the consortium chain network;

[0124] The educational institution nodes, regulatory nodes, and learner nodes conduct consensus verification on the received feature identifier. After the consensus verification passes, the learning behavior data and curriculum resource data are packaged into a block and added to the blockchain;

[0125] The educational institution and the regulatory party add digital signatures to the learning performance scores and learning results in the learning report to generate an anti-counterfeiting authentication label and store it in the blockchain.

[0126] Specifically, to ensure the authenticity and credibility of learning behavior data and course resource data, the consensus mechanism of the blockchain is introduced. First, the learning behavior data and course resource data are subjected to a hashing operation to generate a unique feature identifier through this process. The feature identifier can ensure the uniqueness and immutability of the data. Subsequently, the generated feature identifier is broadcast to each node in the consortium blockchain network, including educational institution nodes, regulatory nodes, and learner nodes. Through broadcasting, it is ensured that all relevant parties can receive the feature identifier. Then, the educational institution nodes, regulatory nodes, and learner nodes perform a consensus verification on the received feature identifier. The consensus verification process ensures the authenticity and consistency of the data through the joint confirmation of multiple parties. Only after the consensus verification is passed can the learning behavior data and course resource data be packaged into a block and added to the blockchain according to the rules of the blockchain. This not only ensures the integrity of the data but also ensures the traceability of the data. Then, the educational institution and the regulatory party further authenticate the learning performance scores and learning outcomes in the learning report. The specific approach is to add a digital signature to these key information to generate an anti-counterfeiting authentication label. The introduction of the digital signature improves the authenticity and credibility of the learning report accordingly. The generated anti-counterfeiting authentication label is then stored in the blockchain, ensuring the openness, transparency, and immutability of the learning report, and providing a trustworthy authentication system for learners.

[0127] Please refer to Figure 2 , an online education system based on the blockchain, using the above-mentioned online education method based on the blockchain, includes:

[0128] Resource on-chain storage module, which is used to build a consortium blockchain network and upload the learner's identity information, learning behavior data, and course resource data to the consortium blockchain network for storage;

[0129] Verification module, which is used to perform multi-dimensional verification on the learner's interaction events through a smart contract and generate a learning report containing a timestamp based on the verification results;

[0130] Personalized recommendation module, which is used to recommend a personalized learning path for the learner based on the learning report and combined with the learner's historical learning performance;

[0131] Consensus verification module, which is used to introduce the consensus mechanism of the blockchain to perform consensus verification on the learning behavior data and course resource data, and to perform an open and transparent authentication on the learner's learning report.

[0132] Among the above, the main responsibility of the resource uploading and storing module is to build a stable and secure consortium blockchain network. Through the consortium blockchain network, the identity information of learners, detailed learning behavior data, and rich course resource data are efficiently uploaded and stored in the consortium blockchain to ensure data transparency and traceability. The verification module uses a preset smart contract to conduct comprehensive and multi-dimensional verification on various interaction events of learners on the platform. The verification process covers multiple aspects such as learning time, learning content, and interaction status to ensure data authenticity and accuracy. Based on the verification results, the system will automatically generate a learning report, which contains timestamps for subsequent query and analysis. The personalized recommendation module, based on the previously generated learning report, combines the learner's past learning performance and historical data to customize a personalized learning path for each learner, thereby improving the learner's learning efficiency and enhancing the learner's learning experience and satisfaction. The consensus verification module introduces the unique consensus mechanism of blockchain to conduct comprehensive and fair consensus verification on learning behavior data and course resource data, and at the same time conducts open and transparent authentication on the learner's learning report to ensure the authenticity and credibility of each report, further enhancing the credibility of the system.

[0133] Please refer to Figure 3 , an electronic device, which includes:

[0134] At least one processor;

[0135] And a memory communicatively connected to the at least one processor;

[0136] Wherein, the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the above-mentioned blockchain-based online education method.

[0137] The processor of the above-mentioned electronic device can be a central processing unit (CPU), a graphics processing unit (GPU), a digital signal processor (DSP), etc. These processors have powerful computing capabilities and data processing capabilities, and can meet the complex data processing and blockchain operation requirements in the online education method. The memory can be a random access memory (RAM), a read-only memory (ROM), a flash memory, or a hard disk drive (HDD), etc., and is used to store computer programs, data, and blockchain information, etc., to ensure the normal operation of the online education method and the security of data. The electronic device can also include an arithmetic unit, an input device, and an output device. The arithmetic unit can provide support for various arithmetic and logical operations to ensure that complex calculation tasks in the online education method can be efficiently completed. The input device, such as a keyboard, a mouse, or a touch screen, enables learners and administrators to conveniently input information and operation instructions. The output device, such as a display, a printer, etc., is used to display information such as learning reports and course resources, and to print necessary documents.

[0138] It should be noted that in this article, the terms "include", "comprise", or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, device, article, or method including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or further includes elements inherent to such a process, device, article, or method. Without more limitations, an element defined by the statement "including one..." does not exclude the existence of another identical element in the process, device, article, or method including that element.

[0139] The above are only the preferred embodiments of the present invention. It should be pointed out that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present invention. The structures, devices, and operation methods not specifically described and explained in the present invention, unless otherwise specified and limited, are implemented according to the conventional means in the art.

Claims

1. An online education method based on blockchain, characterized in that: include: Build a consortium chain network and upload learners’ identity information, learning behavior data, and course resource data to the consortium chain network for storage; Through smart contracts, learners’ interaction events are verified in multiple dimensions, and learning reports with timestamps are generated based on the verification results; Based on the learning report and the learner's historical learning performance, we recommend personalized learning paths for learners. Introduce the consensus mechanism of blockchain to conduct consensus verification of learning behavior data and course resource data, as well as openly and transparently certify learners' learning reports.

2. The online education method based on blockchain according to claim 1, characterized in that: The alliance chain network includes three types of participants: educational institution nodes, regulatory nodes, and learner nodes. Among them, educational institution nodes are responsible for providing course resource data, regulatory nodes are responsible for supervising the behavior of educational institutions and learners, and learner nodes are used to upload and query personal learning behavior data.

3. The online education method based on blockchain according to claim 1, characterized in that: The smart contract is automatically triggered after the learner logs in, monitors the learner's interactive behavior on the learning platform in real time, and outputs it as a multi-dimensional interactive behavior data set; Add timestamps to the interactive behavior data in the multidimensional interactive behavior dataset to generate tamper-proof learning records.

4. An online education method based on blockchain according to claim 1, characterized in that: The step of performing multi-dimensional verification of the learner's interaction events through smart contracts and generating a learning report containing a timestamp based on the verification results includes: Obtain all interactive behavior data of learners on the alliance chain network, including learning period, course chapter jump path and exercise interaction response time; Verify the interactive behavior data according to the preset multi-dimensional verification rules, and quantify the verification results of each interactive behavior data into learning performance scores; The learning performance scores corresponding to each interactive behavior data are integrated to obtain the learner's comprehensive learning performance score; The comprehensive learning performance score is compared with the preset evaluation threshold. When the comprehensive learning performance score is lower than the evaluation threshold, the early warning mechanism is triggered and a review request is sent to the supervision node. Otherwise, the interaction behavior data is synchronized with the blockchain timestamp service and a learning report is generated synchronously. The learning reports are encrypted and sharded and stored through the distributed storage nodes of the alliance chain network, and verifiable digital credentials are returned to the educational institution nodes and learner nodes.

5. The online education method based on blockchain according to claim 4, wherein: The step of verifying the interactive behavior data according to the preset multi-dimensional verification rules and quantifying the verification results of each interactive behavior data into a learning performance score includes: Collect the learner's study period and the opening time of the learning course, calculate the matching degree between the study period and the possible opening time of the learning, and record it as the first characteristic parameter; Obtain the learner's chapter learning sequence and the preset course outline, calculate the degree of fit between the chapter learning sequence and the course outline, and record it as the second feature parameter; For the learner's correct answer rate during the exercise interaction process and the historical correct rate during the same type of exercise interaction process, calculate the deviation between the correct answer rate and the historical correct rate, and record it as the third characteristic parameter; The first characteristic parameter, the second characteristic parameter, and the third characteristic parameter are normalized to obtain the learning performance score of each interactive behavior data under the same dimension.

6. The online education method based on blockchain according to claim 5, characterized in that: The step of fusing the learning performance scores corresponding to the various interactive behavior data to obtain the learner's comprehensive learning performance score includes: Collect the learning performance scores of each interactive behavior data and assign initial weights; According to the learner's course progress stage, the recent interactive behavior data fluctuation characteristics are extracted through a sliding time window to generate a dynamic weight adjustment factor; Combined with the real-time difficulty coefficient of the course resources, the weights of the first characteristic parameter, the second characteristic parameter and the third characteristic parameter are modified differentially; Based on the historical weight distribution records stored in the blockchain, the optimal weight combination for the current course stage is predicted through linear regression; According to the optimal weight combination, the learning performance scores of each interactive behavior data are weighted and summed, and the output is the learner's comprehensive learning performance score.

7. An online education method based on blockchain according to claim 1, characterized in that: The step of recommending a personalized learning path for the learner based on the learning report and the learner's historical learning performance includes: Obtain the learner's historical learning performance stored in the blockchain, wherein the historical learning performance includes knowledge mastery, learning efficiency characteristics, and course progress deviation; Vectorize the knowledge mastery, learning efficiency characteristics, and course progress deviation to generate a three-dimensional feature vector that reflects historical learning performance; Call the preset course knowledge graph, match the three-dimensional feature vector with the course nodes in the course knowledge graph for similarity, and generate the initial recommended path; Obtain a historical successful learning path dataset from the supervision node in the alliance chain network, where the historical successful learning path dataset contains the identity of the learner who has completed the course objectives, the path trajectory hash value, and key learning behavior feature parameters; Obtain the interactive behavior data of the current learner and generate a standardized feature set, and calculate the matching index between the initial recommended path and the historical successful path based on the standardized feature set and the historical successful learning path data set; When the matching index is lower than the preset reconstruction threshold, path reconstruction is triggered. The specific process of path reconstruction includes: Identify redundant paths in historical successful paths through the course knowledge graph, and eliminate them to obtain multiple candidate paths; The knowledge relevance score between each candidate path and the initial recommended path is calculated, and the candidate path with the highest knowledge relevance score is used as the personalized recommended path.

8. The online education method based on blockchain according to claim 1, characterized in that: The steps of introducing the consensus mechanism of blockchain, conducting consensus verification on learning behavior data and course resource data, and conducting open and transparent certification of learners' learning reports include: Perform hash operations on learning behavior data and course resource data to generate unique feature identifiers, and broadcast the feature identifiers to educational institution nodes, regulatory nodes, and learner nodes in the alliance chain network; The educational institution nodes, regulatory nodes, and learner nodes conduct consensus verification on the received feature identifiers. After the consensus verification is passed, the learning behavior data and course resource data are packaged into blocks and added to the blockchain; The educational institution and the regulatory party add digital signatures to the learning performance scores and learning outcomes in the learning report, generate anti-counterfeiting authentication labels, and store them in the blockchain.

9. An online education system based on blockchain, characterized in that: Using the blockchain-based online education method according to any one of claims 1 to 8, comprising: A resource on-chain storage module, which is used to construct a consortium chain network and upload the learner's identity information, learning behavior data, and course resource data to the consortium chain network for storage; A verification module, which is used to perform multi-dimensional verification on the learner's interaction events through a smart contract and generate a learning report containing a timestamp based on the verification results; A personalized recommendation module, which is used to recommend a personalized learning path for the learner based on the learning report and in combination with the learner's historical learning performance; A consensus verification module, which is used to introduce the consensus mechanism of the blockchain to perform consensus verification on the learning behavior data and course resource data, and to perform publicly transparent authentication on the learner's learning report.

10. An electronic device, characterized in that: The electronic device includes: At least one processor; And a memory communicatively connected to the at least one processor; Wherein, the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the blockchain-based online education method according to any one of claims 1 to 8.

Images