Attribute encryption and smart contract-based collaborative human-rearing platform construction method

Abstract

The invention discloses a collaborative human-rearing platform construction method based on attribute encryption and a smart contract. The method comprises the following steps: completing the construction and initialization process of a collaborative human-rearing platform; the enterprise generates and issues a collaborative child-rearing request ciphertext; the student uploads a personal access strategy file; the students decide whether to participate in the collaborative human breeding plan of the enterprise or not according to own demands; the enterprise screens out selectable excellent students willing to participate; entering a personalized culture process, realizing dynamic adjustment, evaluating the culture process, and the like. According to the method, fine-grained authority management specific to the user can be realized, privacy leakage and excessive data sharing in the information exchange process are avoided, and the requirement for compliance of user data is met; the cooperation efficiency of enterprises, schools and students is improved in an online cooperation mode, and the fairness and transparency of the process are ensured through verification and screening of an intelligent contract; an intelligent contract technology is adopted, a data collaboration mode based on a block chain is adopted, a non-tampering trusted collaboration environment is provided, and the trust basis of all parties is enhanced.

Description

Technical Field

[0001] The present invention belongs to the field of information technology, and in particular relates to a method for constructing a collaborative education platform based on attribute encryption and smart contracts. Background Art

[0002] With the renewal of educational concepts and the needs of social development, collaborative education platforms have emerged and have gradually become an integral part of the modern education system. These platforms aim to break the one-way output model of traditional education and establish a close partnership among schools, businesses, and families to jointly promote students' all-round development. However, with the increasing amount of data and the increasing number of processing requirements within collaborative education platforms, these platforms are facing severe security challenges and privacy protection problems. Sensitive information such as students' personal data, learning behaviors, and grades is widely collected, stored, and analyzed. This information falls within the scope of students' privacy and must be properly protected. Furthermore, collaborative education platforms may be subject to security threats such as cyberattacks and data leaks. Therefore, designing a method for constructing a collaborative education platform based on attribute encryption and smart contracts is particularly urgent and important. Summary of the Invention

[0003] In order to solve the above problems, the purpose of the present invention is to provide a method for constructing a collaborative education platform based on attribute encryption and smart contracts.

[0004] To achieve the above objectives, the method for constructing a collaborative education platform based on attribute encryption and smart contracts provided by the present invention includes the following steps performed in sequence:

[0005] 1) Determine the participating roles of the collaborative education platform composed of enterprises and students. Each participating role serves as a participating node, generates a key and access policy file, and has each participating role build a localized collaborative education platform. Then, a master node is generated from all participating nodes of the collaborative education platform. The master node generates a genesis block, thus completing the construction and initialization process of the collaborative education platform.

[0006] 2) The enterprise uses the access policy file and key to generate and publish the collaborative education request ciphertext, determine the number of students selected, and upload it to the smart contract;

[0007] 3) The student obtains the ciphertext of the collaborative education request and uploads the personal access policy file. Then, the student attempts to decrypt the ciphertext of the collaborative education request using their personal key. If the decryption succeeds, the student proceeds to step 4). If the decryption fails, the student proceeds to step 5.

[0008] 4) Students decide whether to participate in the company's collaborative education and training program based on their own needs. If they do, they use the generated personal access policy file and key to encrypt and confirm their participation information and submit it to the smart contract, then jump to step 6). If they do not participate, jump to step 5.

[0009] 5) Repeat steps 3) and 4) until no new collaborative education request ciphertext is released;

[0010] 6) After receiving the students’ confirmation of participation, the company will use a screening algorithm to select outstanding students who are willing to participate and are guaranteed to be selected by the company;

[0011] 7) After selecting outstanding students, they will enter the personalized training process, implement dynamic adjustments, and evaluate the training process.

[0012] In step 1), the participating roles of the collaborative education platform composed of enterprises and students are determined, each participating role serves as a participating node, and a key and access policy file are generated. Each participating role is then asked to build a localized collaborative education platform, and then a master node is generated from all participating nodes of the collaborative education platform. The master node generates a genesis block, thereby completing the construction and initialization process of the collaborative education platform as follows:

[0013] 101) Determine the x and student Stu x The collaborative education platform CE Participating roles, each participating role is a participating node CE_i , participating nodes CE_i Including enterprise node Parties E_i and student nodes Parties S_i ;

[0014] 102) Collaborative Education Platform CE All participating nodes on Node CE_i Generate a private key CE_S and public key CE_P Asymmetric key Key CE_A And access strategy files, and let each participating role build a localized collaborative education platform Platform N ;

[0015] 103) The public key Key CE_P Distribute to all participating nodes CE_i ;

[0016] 104) All participating nodes CE_i Joint consultation and collaborative education platformCE Consensus Node CON_i And generate the master node Node1, the master node Node1 generates the genesis block Block CE_1 , and the public key Key CE_P Write to the genesis block CE_1 Centralize and broadcast, participating nodes CE_i Update the localized collaborative education platform after receiving the broadcast information N , thus completing the construction and initialization process of the collaborative education platform.

[0017] In step 102), the asymmetric key Key CE_A The generation method is as follows:

[0018] 10201) Select a large prime number P, which is usually 1024 bits or longer;

[0019] 10202) Randomly select an integer X as the private key Key CE_S , where 1<X<P-1,X∈R{2,3,…,P-2};

[0020] 10203) Using the private key CE_S , large prime number P and the primitive root G of large prime number P to calculate the public key Key CE_P , the formula is as follows:

[0021] Key CE_P =G X mod P

[0022] In step 104), the method for generating a master node is as follows:

[0023] 10401) Determine the collaborative education platform CE System parameters: time period T1, difficulty value D;

[0024] 10402) Calculate participating nodes CE_i Equity ratio P CE_i , the formula is as follows:

[0025]

[0026] Among them, R CE_i A is a quantitative score for participating node resource contribution. CE_i is the activity score, C CE_i For credibility score, are the weights of the above scores respectively, and z is the total number of participating nodes;

[0027] 10403) According to the above difficulty value D and equity ratio P CE_i , calculate the participating nodes Node CE_i The probability of becoming the master node Node1 is P i , the formula is as follows:

[0028]

[0029] 10404) In each time period T1, the system will CE_i The probability of becoming the master node Node1 is P i Make a random selection;

[0030] 10405) The selected master node Node1 is responsible for collecting the public key Key CE_P , package it into the genesis block BLock CE_1 , and sign and broadcast; Participating Node CE_i Verify the genesis block CE_1 If the verification is successful, the genesis block CE_1 Synchronize to the localized collaborative education platform Platform N middle;

[0031] In step 10404), the method of randomly selecting according to the probability of each participating node becoming a master node is as follows:

[0032] 1040401) Generate a random number Random in the range [0,1] x ;

[0033] 1040402) According to each participating node CE_i The probability of becoming the master node Node1 is P i Calculate the cumulative sum of probabilities where k = 1, 2, ..., n;

[0034] 1040403) Find the one that satisfies Y k <Random x <Y k+1 Participating Node CE_i , then the participating node Node CE_i Become the master node Node1 in the current time period T1.

[0035] In step 3), the student Stu x Obtain the ciphertext of the collaborative education request, upload the personal access policy file, and then try to decrypt the ciphertext of the collaborative education request using your personal key. If the decryption is successful, jump to step 4); if the decryption fails, jump to step 5) as follows:

[0036] 301)Student x Get the collaborative education request ciphertext from the smart contract;

[0037] 302)Student x Upload personal access policy file and query student Stu x The encrypted attribute information corresponding to the individual's private key;

[0038] 303) Try to use your personal private key to decrypt the encrypted attribute information you have queried. If the decryption is successful, the decrypted enterprise Co x The collaborative education request information is returned to the student for review, and the process goes to step 4).

[0039] 304) If decryption fails, jump to step 5);

[0040] In step 302), the student Stu x Upload personal access policy file and query student Stu x The method of encrypting attribute information corresponding to an individual's private key is as follows:

[0041] 30201)Student x Use your private key to digitally sign your personal access policy file and upload it, then use the collaborative education platform Platform CE Perform node verification signature validation to confirm student Stu x The legitimacy of the identity and the authenticity of the operation;

[0042] 30202) Using TLS protocol, x Encrypt the personal access policy file to ensure the security of data during transmission and storage;

[0043] 30203) Broadcast the encrypted personal access policy file to the collaborative education platform Platform in the form of a transaction CE After verification, it will be permanently stored in the collaborative education platform Platform CE and collaborate with the education platform Platform CE The address is associated;

[0044] 30204) on the collaborative education platform Platform CE Establish an index relationship between private keys and encrypted attribute information, and realize structured storage of encrypted attribute information through smart contracts to facilitate subsequent rapid retrieval;

[0045] 30205)Student xInitiate a request to view the encrypted attribute information corresponding to an individual's private key. The private key is also used for digital signature. The encrypted attribute information request contains the private key identification information to be queried.

[0046] 30206) Collaborative Education Platform CE After receiving the above-mentioned encrypted attribute information request, the smart contract in the encrypted attribute information verification process verifies the validity of the signature and, based on the private key identification information in the encrypted attribute information request, queries the corresponding encrypted attribute information including the permission level, validity period and applicable scope from the constructed index relationship.

[0047] In step 6), the enterprise Co x Received student Stu x After confirming the participation information, the guarantee enterprise Co x Excellent students who are willing to participate can be selected b The method is as follows:

[0048] 601) Enterprise Co x Received student Stu x After confirming the participation information, collect the student x The relevant information of evaluation indicators including academic performance, competition awards and skill certificates is then organized into a feature vector F i , i=1,2,…,n, where n is the total number of students, each eigenvector F i ∈R d , d is the number of evaluation indicators;

[0049] 602) Determine each student Stu x Whether the student meets the requirements of the excellent student and is labeled B i Indicates that B i ∈{+1, -1}, +1 means an excellent student, -1 means not;

[0050] 603)Student x Use your personal private key to find the feature vector F i and label B i Perform digital signature and then convert the signed feature vector F i and label B i Submit to the collaborative education platform Platform CE After the smart contract verifies the validity of the signature, it will i and label B i Storage forms an unalterable record;

[0051] 604) Training prediction student Stu xBehavioral model Model1;

[0052] 605) For new student data stu_new , use the above trained model Model1 to perform student Stu x Category prediction, and student category screening is performed through the function shown in the following formula:

[0053] g(F)=sign(δ *T F i +ε * )

[0054] If δ *T F i +ε * >0, then g(F)=+1, indicating that the student Stu x Stu is an excellent student b ;like

[0055] In step 604), the method for training the model for predicting student behavior is as follows:

[0056] 60401) Construct the objective function and find a hyperplane δ for the linearly separable case. T x+ε=0, so that the interval is maximized; solve the following optimization problem:

[0057]

[0058] sB i (δ T F i +ε)≥1, i=1, 2,…, n

[0059] Where δ is the weight vector, ε is the bias, ||δ|| 2 =δ T δ represents the squared norm of the weight vector δ;

[0060] 60402)Introducing the multiplier σ i≥0, i=1, 2, …, g, constructor:

[0061]

[0062] 60403) Take partial derivatives with respect to the weight vector δ, the bias ε, and the multiplier σ, and set them to 0, and we get the dual problem:

[0063]

[0064] 60404) Solve the dual problem and obtain the optimal multiplier

[0065] 60405) According to the optimal multiplier σ * Calculate the weight vector δ * and the bias ε * :

[0066]

[0067] Choose a satisfying Sample (F i ,B i ), calculate the bias ε * :

[0068]

[0069] In step 7), the excellent students Stu b After that, the personalized training process is entered and dynamic adjustment is achieved. The method for evaluating the training process is as follows:

[0070] 701) From the collaborative education platform Platform CE Select outstanding students b The feature vector F in the confirmation participation information i and outstanding students Stu b Divide into x clusters, so that the outstanding students Stu in the cluster b The similarity is the highest, and the inter-cluster similarity is the lowest;

[0071] 702) Define the cluster center vector θ j (j=1,2,…,s), by minimizing the objective function Continuously iterate and update the cluster center until convergence, so as to achieve the goal of b Clustering;

[0072] 703) Enterprise Co x According to the outstanding student Stu b Based on the clustering results, basic training directions are assigned to different categories, and the weights of each training element are determined;

[0073] 704) From the collaborative education platform Platform CE Admiral Student Stu x Dynamically adjust the cultivation process;

[0074] 705) Enterprise Co x For students x Evaluate and summarize the training effect;

[0075] 706) Enterprise Co x Stu x The culture effect score Si Signed with the enterprise private key and uploaded to the collaborative education platform Platform CE , preserved permanently;

[0076] 707) Score S based on the above culture effect i Summary Enterprise Co x Collaborative teaching experience provides reference for subsequent training work;

[0077] In step 703), the enterprise Co x According to the outstanding student Stu b Based on the clustering results, the method for specifying basic training directions for different categories and determining the weights of each training factor is as follows:

[0078] 70301) Constructing a judgment matrix Where m is the number of training elements including theoretical courses, practical projects and academic seminars, a ij is the importance of the i-th training factor relative to the j-th training factor;

[0079] 70302) Calculate the maximum eigenvalue α of the judgment matrix J max and the corresponding eigenvector F;

[0080] 70303) The feature vector F is normalized using L2 to obtain the weight vector ω = (ω1, ω2, ..., ω m ) T ;

[0081] 70304) For each student x , set the expected improvement target value of the training factors to g ij (j=1, 2, ..., m)), and calculate the personalized training priority score E according to the following formula based on the weights: i , and score E based on personalized training priority i Develop student x Detailed personalized training plan;

[0082]

[0083] In step 704), the collaborative education platform Platform CE Admiral Student Stu x The method for dynamically adjusting the cultivation process is as follows:

[0084] 70401) During the training process, regularly obtain CE Collect students Stu x Data including learning progress, assessment scores and project performance form a feedback data set;

[0085] 70402) will be students x The training state is taken as the environment state C, the adjustment measures of the training program are taken as actions A, and the student Stu x The improvement of learning results is used as reward R; the function P(C,A) is defined to represent the expected cumulative reward after performing action A in the environment state C;

[0086] 70403) Using the formula P(C,A)←P(C,A)+T s [R+F d max(P(C′, A′)-P(C, A))] updates the value of the function P, finds the optimal action strategy, and dynamically adjusts the training plan according to the optimal action strategy, where T s is the learning rate, F d is the discount factor;

[0087] In step 705), the enterprise Co x For students x The methods for evaluating the culture effect are as follows:

[0088] 70501) Establish an evaluation system that includes evaluation indicators such as knowledge mastery, skill improvement rate, and practical results, and set corresponding scoring standards for each evaluation indicator;

[0089] 70502) Let the evaluation index be A j (j=1, 2, ..., h), where h is the number of evaluation indicators and the weight is w j , use the weighted average method to calculate the student Stu x Comprehensive score of the training effect Among them A ij is the score of student i on evaluation indicator j.

[0090] The collaborative education platform construction method based on attribute encryption and smart contracts provided by the present invention has the following features:

[0091] Beneficial effects:

[0092] 1. It can realize fine-grained permission management specific to users, avoid privacy leakage and excessive data sharing during information exchange, and meet the requirements for user data compliance; 2. It improves the collaboration efficiency of enterprises, schools and students through online collaboration, and ensures the fairness and transparency of the process through smart contract verification and screening; 3. The use of smart contract technology and blockchain-based data collaboration methods provides a tamper-proof and trusted collaboration environment, enhances the trust basis of all parties, and solves problems such as forged evaluations and false education. BRIEF DESCRIPTION OF THE DRAWINGS

[0093] Figure 1 Flowchart of the method for constructing a collaborative education platform based on attribute encryption and smart contracts provided by the present invention. DETAILED DESCRIPTION

[0094] The present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0095] like Figure 1 As shown, the method for constructing a collaborative education platform based on attribute encryption and smart contracts provided by the present invention includes the following steps performed in sequence:

[0096] 1) Determine the participating roles of the collaborative education platform composed of enterprises and students. Each participating role serves as a participating node, generates a key and access policy file, and has each participating role build a localized collaborative education platform. Then, a master node is generated from all participating nodes of the collaborative education platform. The master node generates a genesis block, thus completing the construction and initialization process of the collaborative education platform.

[0097] Here’s how:

[0098] 101) Determine the x and student Stu x The collaborative education platform CE Participating roles, each participating role is a participating node CE_i , participating nodes CE_i Including enterprise node Parties E_i and student nodes Parties S_i ;

[0099] 102) Collaborative Education Platform CE All participating nodes on Node CE_i Generate a private key CE_S and public key CE_P Asymmetric key Key CE_A And access strategy files, and let each participating role build a localized collaborative education platform Platform N ;

[0100] 103) The public key Key CE_P Distribute to all participating nodes CE_i ;

[0101] 104) All participating nodes CE_i Joint consultation and collaborative education platform Platform CE Consensus Node CON_i And generate the master node Node1, the master node Node1 generates the genesis block BlockCE_1 , and the public key Key CE_P Write to the genesis block CE_1 Centralize and broadcast, participating nodes CE_i Update the localized collaborative education platform after receiving the broadcast information N , thus completing the construction and initialization process of the collaborative education platform.

[0102] In step 102), the asymmetric key Key CE_A The generation method is as follows:

[0103] 10201) Select a large prime number P, which is usually 1024 bits or longer;

[0104] 10202) Randomly select an integer X as the private key Key CE_S , where 1<X<P-1,X∈R{2,3,…,P-2};

[0105] 10203) Using the private key CE_S , large prime number P and the primitive root G of large prime number P to calculate the public key Key CE_P , the formula is as follows:

[0106] Key CE_P =G X mod P

[0107] In step 104), the method for generating a master node is as follows:

[0108] 10401) Determine the collaborative education platform CE System parameters: time period T1, difficulty value D;

[0109] 10402) Calculate participating nodes CE_i Equity ratio P CE_i , the formula is as follows:

[0110]

[0111] Among them, R CE_i A is a quantitative score for participating node resource contribution. CE_i is the activity score, C CE_i For credibility score, are the weights of the above scores respectively, and z is the total number of participating nodes;

[0112] 10403) In each time period T1, each participating node Node CE_iThere is a certain probability of becoming the master node Node1; according to the above difficulty value D and equity ratio P CE_i , calculate the participating nodes Node CE_i The probability of becoming the master node Node1 is P i , the formula is as follows:

[0113]

[0114] 10404) In each time period T1, the system will CE_i The probability of becoming the master node Node1 is P i Make a random selection;

[0115] 10405) The selected master node Node1 is responsible for collecting the public key Key CE_P , package it into the genesis block CE_1 , and sign and broadcast; Participating Node CE_i Verify the genesis block CE_1 If the verification is successful, the genesis block CE_1 Synchronize to the localized collaborative education platform Platform N middle.

[0116] In step 10404), the method of randomly selecting according to the probability of each participating node becoming a master node is as follows:

[0117] 1040401) Generate a random number Random in the range [0,1] x ;

[0118] 1040402) According to each participating node CE_i The probability of becoming the master node Node1 is P i Calculate the cumulative sum of probabilities where k = 1, 2, ..., n;

[0119] 1040403) Find the one that satisfies Y k <Random x <Y k+1 Participating Node CE_i , then the participating node Node CE_i Become the master node Node1 in the current time period T1.

[0120] 2) Enterprise Co x Use the access policy file and key to generate the collaborative education request ciphertext and publish it to determine the student Stu x Select the number of people and upload to the smart contract;

[0121] 3)Student x Obtain the ciphertext of the collaborative education request, upload the personal access policy file, and then try to decrypt the ciphertext of the collaborative education request using your personal key. If the decryption is successful, jump to step 4); if the decryption fails, jump to step 5.

[0122] Here’s how:

[0123] 301)Student x Get the collaborative education request ciphertext from the smart contract;

[0124] 302)Student x Upload personal access policy file and query student Stu x The encrypted attribute information corresponding to the individual's private key;

[0125] 303) Try to use your personal private key to decrypt the encrypted attribute information you have queried. If the decryption is successful, the decrypted enterprise Co x The collaborative education request information is returned to the student for review, and the process goes to step 4).

[0126] 304) If decryption fails, jump to step 5).

[0127] In step 302), the student Stu x Upload personal access policy file and query student Stu x The method of encrypting attribute information corresponding to an individual's private key is as follows:

[0128] 30201)Student x Use your private key to digitally sign your personal access policy file and upload it, then use the collaborative education platform CE Perform node verification signature validation to confirm student Stu x The legitimacy of the identity and the authenticity of the operation;

[0129] 30202) Using TLS protocol, x Encrypt the personal access policy file to ensure the security of data during transmission and storage;

[0130] 30203) Broadcast the encrypted personal access policy file to the collaborative education platform Platform in the form of a transaction CE After verification, it will be permanently stored in the collaborative education platform Platform CE and collaborate with the education platform Platform CE The address is associated;

[0131] 30204) on the collaborative education platform Platform CE Establish an index relationship between private keys and encrypted attribute information, and realize structured storage of encrypted attribute information through smart contracts to facilitate subsequent rapid retrieval;

[0132] 30205)Student x Initiate a request to view the encrypted attribute information corresponding to an individual's private key. The private key is also used for digital signature. The encrypted attribute information request contains the private key identification information to be queried.

[0133] 30206) Collaborative Education Platform CE After receiving the above-mentioned encrypted attribute information request, the smart contract in the encrypted attribute information verification process verifies the validity of the signature and, based on the private key identification information in the encrypted attribute information request, queries the corresponding encrypted attribute information including the permission level, validity period and applicable scope from the constructed index relationship.

[0134] 4)Student x Decide whether to participate in Enterprise Co according to your own needs x If you decide to participate in the collaborative education and training program, use the generated personal access policy file and key to encrypt the confirmation information and submit it to the smart contract, then jump to step 6); if you do not participate, jump to step 5);

[0135] 5) Repeat steps 3) and 4) until no new collaborative education request ciphertext is released;

[0136] 6) Enterprise Co x Received student Stu x After confirming the participation information, the guarantee enterprise Co x Excellent students who are willing to participate can be selected b ;

[0137] Here’s how:

[0138] 601) Enterprise Co x Received student Stu x After confirming the participation information, collect the student x The relevant information of evaluation indicators including academic performance, competition awards and skill certificates is then organized into a feature vector F i , i=1,2,…,n, where n is the total number of students, each eigenvector F i ∈R d , d is the number of evaluation indicators;

[0139] 602) Determine each student Stu x Whether the student meets the requirements of the excellent student and is labeled Bi Indicates that B i ∈{+1, -1}, +1 means an excellent student, -1 means not;

[0140] 603)Student x Use your personal private key to find the feature vector F i and label B i Perform digital signature and then convert the signed feature vector F i and label B i Submit to the collaborative education platform Platform CE After the smart contract verifies the validity of the signature, it will i and label B i Storage forms an unalterable record;

[0141] 604) Training prediction student Stu x Behavioral model Model1;

[0142] 605) For new student data stu_new , use the above trained model Model1 to perform student Stu x Category prediction, and student category screening is performed through the function shown in the following formula:

[0143] g(F)=sign(δ *T F i +ε * )

[0144] If δ *T F i +ε * >0, then g(F)=+1, indicating that the student Stu x Stu is an excellent student b If δ *T F i +ε * <0, then g(F)=-1, indicating that the student Stu x Not an excellent student Stu b .

[0145] In step 604), the method for training the model for predicting student behavior is as follows:

[0146] 60401) Construct the objective function and find a hyperplane δ for the linearly separable case. T x+ε=0, so that the interval is maximized; solve the following optimization problem:

[0147]

[0148] sBi (δ T F i +ε)≥1, i=1, 2,…, n

[0149] Where δ is the weight vector, ε is the bias, ||δ|| 2 =δ T δ represents the squared norm of the weight vector δ;

[0150] 60402) Introducing the multiplier σ i ≥0, i=1, 2, …, g, constructor:

[0151]

[0152] 60403) Take partial derivatives with respect to the weight vector δ, the bias ε, and the multiplier σ, and set them to 0, and we get the dual problem:

[0153]

[0154] 60404) Solve the dual problem and obtain the optimal multiplier

[0155] 60405) According to the optimal multiplier σ * Calculate the weight vector δ * and the bias ε * :

[0156]

[0157] Choose a satisfying Sample (F i ,B i ), calculate the bias ε * :

[0158] ε * =B i -δ *T F i

[0159] 7) Select outstanding students b After that, it enters the personalized training process, realizes dynamic adjustment, and evaluates the training process.

[0160] Here’s how:

[0161] 701) From the collaborative education platform Platform CE Select outstanding students b The feature vector F in the confirmation participation information i and outstanding students Stu b Divide into x clusters, so that the outstanding students Stu in the clusterb The similarity is the highest, and the inter-cluster similarity is the lowest;

[0162] 702) Define the cluster center vector θ j (j=1,2,…,s), by minimizing the objective function Continuously iterate and update the cluster center until convergence, so as to achieve the goal of b Clustering;

[0163] 703) Enterprise Co x According to the outstanding student Stu b Based on the clustering results, basic training directions are assigned to different categories, and the weights of each training element are determined;

[0164] 704) From the collaborative education platform Platform CE Admiral Student Stu x Dynamically adjust the cultivation process;

[0165] 705) Enterprise Co x For students x Evaluate and summarize the training effect;

[0166] 706) Enterprise Co x Stu x The culture effect score S i Signed with the enterprise private key and uploaded to the collaborative education platform Platform CE , preserved permanently;

[0167] 707) Score S based on the above culture effect i Summary Enterprise Co x Collaborative teaching experience provides a reference for subsequent training work.

[0168] In step 703), the enterprise Co x According to the outstanding student Stu b Based on the clustering results, the method for specifying basic training directions for different categories and determining the weights of each training factor is as follows:

[0169] 70301) Constructing a judgment matrix Where m is the number of training elements including theoretical courses, practical projects and academic seminars, a ij is the importance of the i-th training factor relative to the j-th training factor;

[0170] 70302) Calculate the maximum eigenvalue α of the judgment matrix J max and the corresponding eigenvector F;

[0171] 70303) The feature vector F is normalized using L2 to obtain the weight vector ω = (ω1, ω2, ..., ω m ) T ;

[0172] 70304) For each student x , set the expected improvement target value of the training factors to g ij (j=1, 2, ..., m), and the personalized training priority score E is calculated based on the following formula in combination with the weights: i , and score E based on personalized training priority i Develop student x Detailed personalized training plan;

[0173]

[0174] In step 704), the collaborative education platform Platform CE Admiral Student Stu x The method for dynamically adjusting the cultivation process is as follows:

[0175] 70401) During the training process, regularly obtain CE Collect students Stu x Data including learning progress, assessment scores and project performance form a feedback data set;

[0176] 70402) will be students x The training state is taken as the environment state C, the adjustment measures of the training program are taken as actions A, and the student Stu x The improvement of learning results is used as reward R; the function P(C,A) is defined to represent the expected cumulative reward after performing action A in the environment state C;

[0177] 70403) Using the formula P(C,A)←P(C,A)+T s [R+F d max(P(C′,A′)-P(C,A))] updates the value of the function P, finds the optimal action strategy, and dynamically adjusts the training plan according to the optimal action strategy, where T s is the learning rate, F d is the discount factor.

[0178] In step 705), the enterprise Co x For students x The methods for evaluating the culture effect are as follows:

[0179] 70501) Establish an evaluation system that includes evaluation indicators such as knowledge mastery, skill improvement rate, and practical results, and set corresponding scoring standards for each evaluation indicator;

[0180] 70502) Let the evaluation index be A j (j=1,2,…,h), where h is the number of evaluation indicators and the weight is w j , use the weighted average method to calculate the student Stu x Comprehensive score of the training effect Among them A ij is the score of student i on evaluation indicator j.

Claims

1. A method for constructing a collaborative education platform based on attribute encryption and smart contracts, characterized by: The method for constructing a collaborative education platform based on attribute encryption and smart contracts includes the following steps performed in sequence: 1) Determine the participating roles of the collaborative education platform composed of enterprises and students. Each participating role serves as a participating node, generates a key and access policy file, and has each participating role build a localized collaborative education platform. Then, a master node is generated from all participating nodes of the collaborative education platform. The master node generates a genesis block, thus completing the construction and initialization process of the collaborative education platform. 2) The enterprise uses the access policy file and key to generate and publish the collaborative education request ciphertext, determine the number of students selected, and upload it to the smart contract; 3) The student obtains the ciphertext of the collaborative education request and uploads the personal access policy file. Then, the student attempts to decrypt the ciphertext of the collaborative education request using their personal key. If the decryption succeeds, the student proceeds to step 4). If the decryption fails, the student proceeds to step 5. 4) Students decide whether to participate in the company's collaborative education and training program based on their own needs. If they do, they use the generated personal access policy file and key to encrypt and confirm their participation information and submit it to the smart contract, then jump to step 6). If they do not participate, jump to step 5. 5) Repeat steps 3) and 4) until no new collaborative education request ciphertext is released; 6) After receiving the students’ confirmation of participation, the company will use a screening algorithm to select outstanding students who are willing to participate and are guaranteed to be selected by the company; 7) After selecting outstanding students, they will enter the personalized training process, implement dynamic adjustments, and evaluate the training process.

2. The method for constructing a collaborative education platform based on attribute encryption and smart contracts according to claim 1 is characterized by: In step 1), the participating roles of the collaborative education platform composed of enterprises and students are determined, each participating role serves as a participating node, and a key and access policy file are generated. Each participating role is then asked to build a localized collaborative education platform, and then a master node is generated from all participating nodes of the collaborative education platform. The master node generates a genesis block, thereby completing the construction and initialization process of the collaborative education platform as follows: 101) Determine the x and student Stu x The collaborative education platform CE Participating roles, each participating role is a participating node CE_i , participating nodes CE_i Including enterprise node Parties E_i and student nodes Parties S_i ; 102) Collaborative Education Platform CE All participating nodes on Node CE_i Generate a private key CE_S and public key CE_P Asymmetric key Key CE_A And access strategy files, and let each participating role build a localized collaborative education platform Platform N ; 103) The public key Key CE_P Distribute to all participating nodes CE_i ; 104) All participating nodes CE_i Joint consultation and collaborative education platform Platform CE Consensus Node CON_i And generate the master node Node1, the master node Node1 generates the genesis block Block CE_1 , and the public key Key CE_P Write to the genesis block CE_1 Centralize and broadcast, participating nodes CE_i Update the localized collaborative education platform after receiving the broadcast information N , thus completing the construction and initialization process of the collaborative education platform.

3. The method for constructing a collaborative education platform based on attribute encryption and smart contracts according to claim 2 is characterized by: In step 102), the asymmetric key Key CE_A The generation method is as follows: 10201) Select a large prime number P, which is usually 1024 bits or longer; 10202) Randomly select an integer X as the private key Key CE_S , where 1 <X<P-1,X∈R{2,3,…,P-2}; 10203) Using the private key CE_S , large prime number P and the primitive root G of large prime number P to calculate the public key Key CE_P , the formula is as follows: Key CE_P =G X mod P In step 104), the method for generating a master node is as follows: 10401) Determine the collaborative education platform CE System parameters: time period T1, difficulty value D; 10402) Calculate participating nodes CE_i Equity ratio P CE_i , the formula is as follows: Among them, R CE_i A is a quantitative score for participating node resource contribution. CE_i is the activity score, C CE_i For credibility score, are the weights of the above scores respectively, and z is the total number of participating nodes; 10403) According to the above difficulty value D and equity ratio P CE_i , calculate the participating nodes Node CE_i The probability of becoming the master node Node1 is P i , the formula is as follows: 10404) In each time period T1, the system will CE_i The probability of becoming the master node Node1 is P i Make a random selection; 10405) The selected master node Node1 is responsible for collecting the public key Key CE_P , package it into the genesis block CE_1 , and sign and broadcast; Participating Node CE_i Verify the genesis block CE_1 If the verification is successful, the genesis block CE_1 Synchronize to the localized collaborative education platform Platform N middle; In step 10404), the method of randomly selecting according to the probability of each participating node becoming a master node is as follows: 1040401) Generate a random number Random in the range [0,1] x ; 1040402) According to each participating node CE_i The probability of becoming the master node Node1 is P i Calculate the cumulative sum of probabilities where k = 1, 2, ..., n; 1040403) Find the one that satisfies Y k <Random x <Y k+1 Participating Node CE_i , then the participating node Node CE_i Become the master node Node1 in the current time period T1.

4. The method for constructing a collaborative education platform based on attribute encryption and smart contracts according to claim 1 is characterized by: In step 3), the student Stu x Obtain the ciphertext of the collaborative education request, upload the personal access policy file, and then try to decrypt the ciphertext of the collaborative education request using your personal key. If the decryption is successful, jump to step 4); if the decryption fails, jump to step 5) as follows: 301)Student x Get the collaborative education request ciphertext from the smart contract; 302)Student x Upload personal access policy file and query student Stu x The encrypted attribute information corresponding to the individual's private key; 303) Try to use your personal private key to decrypt the encrypted attribute information you have queried. If the decryption is successful, the decrypted enterprise Co x The collaborative education request information is returned to the student for review, and the process goes to step 4). 304) If decryption fails, jump to step 5); In step 302), the student Stu x Upload personal access policy file and query student Stu x The method of encrypting attribute information corresponding to an individual's private key is as follows: 30201)Student x Use your private key to digitally sign your personal access policy file and upload it, then use the collaborative education platform Platform CE Perform node verification signature validation to confirm student Stu x The legitimacy of the identity and the authenticity of the operation; 30202) Using TLS protocol, x Encrypt the personal access policy file to ensure the security of data during transmission and storage; 30203) Broadcast the encrypted personal access policy file to the collaborative education platform Platform in the form of a transaction CE After verification, it will be permanently stored in the collaborative education platform Platform CE and collaborate with the education platform Platform CE The address is associated; 30204) on the collaborative education platform Platform CE Establish an index relationship between private keys and encrypted attribute information, and realize structured storage of encrypted attribute information through smart contracts to facilitate subsequent rapid retrieval; 30205)Student x Initiate a request to view the encrypted attribute information corresponding to an individual's private key. The private key is also used for digital signature. The encrypted attribute information request contains the private key identification information to be queried. 30206) Collaborative Education Platform CE After receiving the above-mentioned encrypted attribute information request, the smart contract in the encrypted attribute information verification process verifies the validity of the signature and, based on the private key identification information in the encrypted attribute information request, queries the corresponding encrypted attribute information including the permission level, validity period and applicable scope from the constructed index relationship.

5. The method for constructing a collaborative education platform based on attribute encryption and smart contracts according to claim 1 is characterized in that: In step 6), the enterprise Co x Received student Stu x After confirming the participation information, the guarantee enterprise Co x Excellent students who are willing to participate can be selected b The method is as follows: 601) Enterprise Co x Received student Stu x After confirming the participation information, collect the student x The relevant information of evaluation indicators including academic performance, competition awards and skill certificates is then organized into a feature vector F i , i=1,2,…,n, where n is the total number of students, each eigenvector F i ∈R d , d is the number of evaluation indicators; 602) Determine each student Stu x Whether the student meets the requirements of the excellent student and is labeled B i Indicates that B i ∈{+1, -1}, +1 means an excellent student, -1 means not; 603)Student x Use your personal private key to find the feature vector F i and label B i Perform digital signature and then convert the signed feature vector F i and label B i Submit to the collaborative education platform Platform CE After the smart contract verifies the validity of the signature, it will i and label B i Storage forms an unalterable record; 604) Training prediction student Stu x Behavioral model Model1; 605) For new student data stu_new , use the above trained model Model1 to perform student Stu x Category prediction, and student category screening is performed through the function shown in the following formula: g(F)=sign(δ *T F i +ε * ) If δ *T F i +ε * >0, then g(F)=+1, indicating that the student Stu x Stu is an excellent student b If δ *T F i +ε * <0, then g(F)=-1, indicating that the student Stu x Not an excellent student Stu b ; In step 604), the method for training the model for predicting student behavior is as follows: 60401) Construct the objective function and find a hyperplane δ for the linearly separable case. T x+ε=0, which maximizes the interval; Solve the following optimization problem: stB i (d T F i +ε)≥1,i=1,2,…,n Where δ is the weight vector, ε is the bias, ||δ|| 2 =δ T δ represents the squared norm of the weight vector δ; 60402) Introducing the multiplier σ i ≥0, i=1,2,…,g, constructor: 60403) Take partial derivatives with respect to the weight vector δ, the bias ε, and the multiplier σ, and set them to 0, and we get the dual problem: 60404) Solve the dual problem and obtain the optimal multiplier 60405) According to the optimal multiplier σ * Calculate the weight vector δ * and the bias ε * : Choose a satisfying Sample (F i , B i ), calculate the bias ε * : e * =B i -d *T F i 。 6. The method for constructing a collaborative education platform based on attribute encryption and smart contracts according to claim 1 is characterized by: In step 7), the excellent students Stu b After that, the personalized training process is entered and dynamic adjustment is achieved. The method for evaluating the training process is as follows: 701) From the collaborative education platform Platform CE Select outstanding students b The feature vector F in the confirmation participation information i and outstanding students Stu b Divide into x clusters, so that the outstanding students Stu in the cluster b The similarity is the highest, and the inter-cluster similarity is the lowest; 702) Define the cluster center vector θ j (j=1, 2, ..., s), by minimizing the objective function Continuously iterate and update the cluster center until convergence, so as to achieve the goal of b Clustering; 703) Enterprise Co x According to the outstanding student Stu b Based on the clustering results, basic training directions are assigned to different categories, and the weights of each training element are determined; 704) From the collaborative education platform Platform CE Admiral Student Stu x Dynamically adjust the cultivation process; 705) Enterprise Co x For students x Evaluate and summarize the training effect; 706) Enterprise Co x Stu x The culture effect score S i Signed with the enterprise private key and uploaded to the collaborative education platform Platform CE , preserved permanently; 707) Score S based on the above culture effect i Summary Enterprise Co x Collaborative teaching experience provides reference for subsequent training work; In step 703), the enterprise Co x According to the outstanding student Stu b Based on the clustering results, the method for specifying basic training directions for different categories and determining the weights of each training factor is as follows: 70301) Constructing a judgment matrix Where m is the number of training elements including theoretical courses, practical projects and academic seminars, a ij is the importance of the i-th training factor relative to the j-th training factor; 70302) Calculate the maximum eigenvalue α of the judgment matrix J max and the corresponding eigenvector F; 70303) After L2 normalization of the feature vector F, the weight vector ω = (ω1, ω2, ..., ω m ) T ; 70304) For each student x , set the expected improvement target value of the training factors to g ij (j=1,2,…,m), and calculate the personalized training priority score E according to the following formula based on the weights i , and score E based on personalized training priority i Develop student x Detailed personalized training plan; In step 704), the collaborative education platform Platform CE Admiral Student Stu x The method for dynamically adjusting the cultivation process is as follows: 70401) During the training process, regularly obtain CE Collect students Stu x Data including learning progress, assessment scores and project performance form a feedback data set; 70402) will be students x The training state is taken as the environment state C, the adjustment measures of the training program are taken as actions A, and the student Stu x The improvement of learning results is used as reward R; the function P(C,A) is defined to represent the expected cumulative reward after performing action A in the environment state C; 70403) Using the formula P(C,A)←P(C,A)+T s [R+F d max(P(C′, A′)-P(C, A))] updates the value of the function P, finds the optimal action strategy, and dynamically adjusts the training plan according to the optimal action strategy, where T s is the learning rate, F d is the discount factor; In step 705), the enterprise Co x For students x The methods for evaluating the culture effect are as follows: 70501) Establish an evaluation system that includes evaluation indicators such as knowledge mastery, skill improvement rate, and practical results, and set corresponding scoring standards for each evaluation indicator; 70502) Let the evaluation index be A j (j=1, 2, ..., h), where h is the number of evaluation indicators and the weight is w j , use the weighted average method to calculate the student Stu x Comprehensive score of the training effect Among them A ij is the score of student i on evaluation indicator j.

Images