Online leasing method based on block chain

Abstract

The invention provides an online leasing method based on a block chain, and relates to the technical field of mobile phone online leasing, the online leasing method based on the block chain comprises the steps that the block chain is constructed, and user data of related platforms except a leasing platform are accessed, and the related platforms comprise an e-commerce platform, a financial platform and a credit rating platform; the smart contract searches user data related to the leasing user and accessed to the block chain by the related platform on the block chain; and the leasing platform calculates the credit rating of the leasing user according to the user data returned by the smart contract, and sets a corresponding leasing condition for the leasing user according to the credit rating. Therefore, the situation that only a few data sources are depended on during credit evaluation is avoided, different platforms and mechanisms can establish a data sharing mechanism on the block chain, more comprehensive data resources are provided for credit level calculation of leasing users, the accuracy of credit evaluation is improved, and potential losses caused by bad leasing to the leasing platform are reduced.

Description

Technical Field

[0001] This application relates to the technical field of online mobile phone rental, and particularly to an online rental method based on blockchain. Background Art

[0002] Online mobile phone rental is a new consumption model. Users rent mobile phone devices through an Internet platform, pay rent within a specified time, and choose to return the mobile phone, continue renting, or purchase the mobile phone according to the contract terms after the lease expires. Compared with the traditional mobile phone sales model, mobile phone rental converts the ownership of the mobile phone into the right of use, enabling users to enjoy the right of use of the mobile phone during the lease period after paying a certain rent.

[0003] The online mobile phone rental mainly includes the following processes: user registration and login, selection of the rented mobile phone, credit assessment, signing of the lease contract, device delivery and use, management during the lease period, return at the end of the lease, and subsequent services and feedback, etc. In the online mobile phone rental process, through credit assessment, the rental platform can identify users with a high risk of default to reduce the economic losses caused by user default. Currently, the common credit assessment method of the rental platform is to view the Sesame Credit score or credit information of other credit platforms (such as JD Xiaobai Credit) after obtaining the user's authorization, as a reference for credit assessment. For example, the invention application with the publication number CN118172070A discloses a risk control system for mobile phone rental services. This risk control system for mobile phone rental services evaluates the rental risk of users by calculating the risk index of users. The risk index is a representative parameter of the user's rental risk, used to represent the user's rental risk. This invention calculates the user's risk index based on the Sesame Credit score, the number of monthly performance fulfillment transactions, and the number of annual overdue records. The Sesame Credit score, the number of monthly performance fulfillment transactions, and the number of annual overdue records are indicators that continuously and objectively reflect the user's credit status over a long period. Through the above three indicators, the rental risk of users can be objectively reflected.

[0004] However, when the above-mentioned risk control system for mobile phone rental services conducts credit assessment, it only relies on a few data sources such as the Sesame Credit score and performance records from the Ant Financial platform and the rental platform. There are often data island problems with other platforms, and data is difficult to share and circulate. There is a lack of comprehensive consideration of more dimensional data of users, resulting in an incomplete credit assessment. For example, some users may provide false information to increase their Sesame Credit score. If the rental platform only relies on data sources such as the Sesame Credit score and performance records during credit assessment, it may not be able to effectively identify these fraud behaviors, thus bringing potential losses to the rental platform. Summary of the Invention

[0005] The purpose of this application is to provide an online leasing method based on blockchain, which is used to solve the problem that in the risk control system of the mobile phone leasing business in the related technology, when conducting credit assessment, the comprehensive consideration of more dimensional data of users is lacking, resulting in incomplete credit assessment.

[0006] An online leasing method based on blockchain provided by this application adopts the following technical solutions:

[0007] An online leasing method based on blockchain includes:

[0008] Construct a blockchain, formulate a unified data interface standard, access the user data of relevant platforms other than the leasing platform, and set the authorization permission for the leasing platform to access the user data of relevant platforms through a smart contract. The relevant platforms include e-commerce platforms, financial platforms, and credit rating platforms;

[0009] After the leasing user submits a leasing request, the leasing platform sends an access request for user data to the smart contract on the blockchain. After the request is verified, the smart contract searches on the blockchain for the user data related to the leasing user connected to the blockchain by the relevant platform, and returns the user data to the leasing platform;

[0010] The leasing platform calculates the credit rating of the leasing user based on the user data returned by the smart contract, and sets corresponding leasing conditions for the leasing user according to the credit rating.

[0011] Optionally, the leasing platform calculates the credit rating of the leasing user based on the user data returned by the smart contract, including: respectively calculating the first credit score S of the leasing user on the e-commerce platform 1 , the second credit score S of the leasing user on the financial platform 2 , and the third credit score S of the leasing user on the credit rating platform 3 , and fusing S 1 , S 2 , and S 3 to obtain a comprehensive credit score S, and mapping the comprehensive credit score S to the corresponding credit rating according to a preset threshold.

[0012] Optionally, the fusing of S 1 , S 2 , and S 3 to obtain a comprehensive credit score S includes: constructing a judgment matrix A=(a ij ) n×n , where a ij is the importance ratio of platform i relative to platform j; calculating the product M i of each row element of the judgment matrix A, calculating the cube root of M i Pair is normalized to obtain the weight vector ω = (ω 1 , ω 2 , ω 3 ), calculate the comprehensive credit score S,

[0013] Optionally, the calculation of the first credit score S of the rental user on the e-commerce platform 1 , includes: extracting e-commerce platform shopping data from the user data returned by the smart contract, and extracting credit features related to the rental user from the e-commerce platform shopping data, and converting the credit features related to the rental user into the first credit score S through the following formula 1 :

[0014] S 1 = ω 1 * N(f 1 ) + ω 2 * N(f 2 ) + ω 3 * N(f 3 ) + ω 4 * N(f 4 ),

[0015] where f 1 is the consumption frequency within a period of time, f 2 is the average consumption amount within a period of time, f 3 is the number of types of goods purchased within a period of time, f 4 is the retention rate of the purchased goods within a period of time, ω 1 , ω 2 , ω 3 and ω 4 are the weights of each feature respectively, and satisfy ω 1 + ω 2 + ω 3 + ω 4 = 1, N(f i ) is the index after normalizing the credit feature f i related to the rental user, and its normalization formula is:

[0016]

[0017] where min(f i ) is the minimum value selected after statistics on the credit feature f i of a certain number of users on the e-commerce platform, and max(f i ) is the maximum value selected after statistics on the credit feature f i of a certain number of users on the e-commerce platform.

[0018] Optionally, calculating the second credit score S of the leasing user on the financial platform 2 , including: collecting the financial data X of the training sample users on the financial platform, preprocessing the financial data X, and normalizing the feature variables in the financial data X to the interval [0, 1]; selecting the Gaussian kernel function as the kernel function, and determining the optimal penalty parameter and kernel function parameter through cross-validation; classifying the credit ratings of the training sample users into different categories y, and using the data samples (X, y) as the training sample set to train the support vector machine model to obtain the decision function f(x), extracting the financial data X related to the leasing user from the user data returned by the smart contract new , inputting the financial data X new into the decision function to calculate f(X new ); calculating the second credit score S of the leasing user on the financial platform according to f(X new ), S 2 = k × f(X 2 ) + c, where k is the conversion coefficient and c is the offset; new )+c, where k is the conversion coefficient and c is the offset;

[0019] The formula of the decision function f(x) is as follows:

[0020]

[0021] where x is the input feature vector, n is the number of training samples, α i is the Lagrange multiplier, y i is the credit rating category of the training sample, K(x i , x) is the Gaussian kernel function, x i is the feature vector of the i-th training sample, and b is the bias term.

[0022] Optionally, calculating the third credit score S of the leasing user on the credit rating platform 3 , including: obtaining the scoring data source of the original credit score S c of the leasing user on the credit rating platform, searching for verification data related to the scoring data source in the financial data X new related to the leasing user, and verifying the original credit score S c through the verification data to obtain the scoring authenticity verification value V, and calculating the third credit score S 3 according to the scoring authenticity verification value V, S 3 = V * S c ;

[0023] The calculation formula of the scoring authenticity verification value V is as follows:

[0024]

[0025] Among them, n is the number of different data items in the verification data, and w i is the weight of the i-th data item, which is set according to its importance for authenticity judgment, and m i is the matching degree between the scoring data source and the i-th data item in the verification data, and its value range is between [0, 1]. 1 indicates a perfect match, and 0 indicates a complete mismatch.

[0026] Optionally, setting corresponding rental conditions for rental users according to the credit rating includes: the rental platform sets a corresponding rental deposit D for the rental user according to the credit rating, and the calculation formula of the rental deposit D is as follows:

[0027]

[0028] Among them, D b is the basic deposit amount, N d is the historical default times of the rental user, S i is the income stability coefficient, and its value range is between [0, 1]. The closer S i is to 1, the more stable the income is, N r is the historical rental times of the rental user, and α, β, and γ are weight coefficients. D s is the basic deposit discount rate, and R is the credit rating, and its value range is between [0, 1].

[0029] Optionally, the rental platform rewards relevant platforms according to the user data situation of relevant platforms accessing the blockchain, and calculates the reward amount through the following formula:

[0030] I = α×(e βD -1) + γ×Q 2 + δ×ln(F + 1) + ε×(1 - e -λT ),

[0031] Among them, I is the reward amount value, D is the amount of user data of relevant platforms accessing the blockchain, Q is the quality score of user data of relevant platforms accessing the blockchain, and its value range is between [0, 1]. F is the data sharing frequency of relevant platforms within a certain period of time, T is the duration for which relevant platforms continuously access data to the blockchain, and α, β, γ, δ, ε, and λ are adjustment parameters.

[0032] Optionally, the user data of relevant platforms other than the access rental platform includes: encrypting the user data of relevant platforms and storing the encrypted data locally or in a distributed storage system, and storing the hash value of the encrypted data on the blockchain.

[0033] Optionally, the smart contract looks up relevant platform access to user data related to the rental user on the blockchain and returns the user data to the rental platform, including: the smart contract looks up the hash value of user data related to the rental user on the blockchain, obtains the encrypted data from a local or distributed storage system, recalculates the hash value and compares it with the hash value stored on the blockchain to verify the integrity of the data, then decrypts the encrypted data, and returns the decrypted user data to the rental platform.

[0034] In summary, the present application at least includes the following beneficial technical effects: by constructing a blockchain, formulating a unified data interface standard, and accessing user data of relevant platforms other than the rental platform, the relevant platforms include three platforms: an e-commerce platform, a financial platform, and a credit rating platform. After the rental user submits a rental request, the smart contract looks up the user data related to the rental user that the relevant platform accesses to the blockchain on the blockchain. The rental platform calculates the credit rating of the rental user based on the user data returned by the smart contract, and sets corresponding rental conditions for the rental user according to the credit rating. Thus, when conducting credit assessment, it avoids relying solely on a few data sources from platforms such as the Ant Financial platform or the rental platform, breaks data islands, enables different platforms and institutions to establish a data sharing mechanism on the blockchain, realizes the interconnection and interoperability of user data, integrates multi-source data, provides more comprehensive and rich data resources for calculating the credit rating of rental users, improves the accuracy of credit assessment, increases the recognition rate of fraud behavior of rental users, and reduces potential losses brought by bad rentals to the rental platform. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 It is a schematic flowchart of an online rental method based on blockchain in an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0036] The following is a further detailed description of the present application in conjunction with the attached Figure 1 drawings.

[0037] An embodiment of the present application discloses an online rental method based on blockchain.

[0038] An online rental method based on blockchain includes the following steps:

[0039] S1. Construct a blockchain, and the blockchain type can be selected as Ethereum. Formulate a unified data interface standard. For example, a unified RESTful API interface can be used to facilitate different platforms to access and operate data, and a JSON format can be used to define the data structure stored on the blockchain. This JSON structure defines the basic information of user data, including user ID, data type, data value, timestamp, and source platform. User data of different platforms can be stored and parsed according to this structure.

[0040] Access the user data of relevant platforms other than the rental platform. During this process, encrypt the user data of relevant platforms and store the encrypted data locally or in a distributed storage system, and store the hash value of the encrypted data on the blockchain. The relevant platforms include, but are not limited to, e-commerce platforms, financial platforms, and credit rating platforms.

[0041] Set the authorization permission for the rental platform to access the user data of relevant platforms through a smart contract. You can use Ethereum development tools (such as Remix, Truffle, or Hardhat) to deploy the smart contract to the Ethereum blockchain. The relevant platform can grant the rental platform the permission to access its own platform data by taking the address of the rental platform, the address of its own platform, and the authorization expiration time as parameters.

[0042] To enrich the data resources on the blockchain, promote data sharing and collaboration, attract more platforms to join, and form an active blockchain ecosystem and community. The rental platform can reward relevant platforms according to the user data situation of relevant platforms accessing the blockchain, and calculate the reward amount through the following formula:

[0043] I = α×(e βD -1)+γ×Q 2 +δ×ln(F + 1)+ε×(1 - e -λT )

[0044] Where, I is the value of the reward amount, D is the amount of user data of relevant platforms accessing the blockchain. For the part of the amount of user data D accessing the blockchain, use the exponential function e βD -1 to incentivize the data volume. When the data volume increases, the incentive value will increase exponentially, thereby motivating relevant platforms to actively increase the amount of data accessing the blockchain; Q is the quality score of the user data of relevant platforms accessing the blockchain, and the value range is between [0, 1]. For the part of the user data quality score Q, use Q 2 as an incentive factor to emphasize the importance of data quality, and the square term can make high-quality data (close to 1) obtain higher incentives, while punishing low-quality data (close to 0), thereby prompting relevant platforms to improve data quality; F is the data sharing frequency of relevant platforms within a certain period of time. For the part of the data sharing frequency F, use the natural logarithm function ln(F + 1) to incentivize the data sharing frequency. When F is small, the growth is relatively slow, but as F increases, the incentive growth will gradually accelerate, thus avoiding over-incentivizing relevant platforms with low sharing frequencies, while encouraging relevant platforms to increase the sharing frequency; T is the duration for which relevant platforms continuously access data to the blockchain. For the part of the duration T of continuously accessing data to the blockchain, use 1 - e -λTThe function is such that as the duration T increases, the incentive gradually approaches ε, enabling relevant platforms to be motivated to stably connect data to the blockchain in the long term. Relevant platforms that contribute data in the long term will receive higher incentives; α, β, γ, δ, ε, and λ are adjustment parameters used to adjust the influence degree of each factor on the incentive value.

[0045] S2. After the rental user submits a rental request, the rental platform sends an access request for the user data to the smart contract on the blockchain. After the request is verified, the smart contract searches on the blockchain for the user data related to the rental user that is connected to the blockchain and returns the user data to the rental platform. More specifically, the smart contract searches on the blockchain for the hash value of the user data related to the rental user, obtains the encrypted data from a local or distributed storage system, recalculates the hash value and compares it with the hash value stored on the blockchain to verify the integrity of the data, then decrypts the encrypted data and returns the decrypted user data to the rental platform.

[0046] By recalculating the hash value and comparing it with the hash value stored on the blockchain to verify the integrity of the data, the immutability of the blockchain can be utilized to effectively ensure the integrity and security of user data, provide reliable guarantee for the storage, use, and auditing of data, prevent the leakage or tampering of user data without being detected, and provide important support for the security and trust mechanism of the credit rating assessment of rental users.

[0047] S3. The rental platform calculates the credit rating of the rental user based on the user data returned by the smart contract. The rental platform calculates the credit rating of the rental user based on the user data returned by the smart contract, including the following steps: respectively calculate the first credit score S of the rental user on the e-commerce platform 1 、the second credit score S of the rental user on the financial platform 2 and the third credit score S of the rental user on the credit rating platform 3 , and fuse S 1 、S 2 and S 3 to obtain a comprehensive credit score S. The comprehensive credit score S can comprehensively consider the credit information of the rental user in multiple dimensions, improve the accuracy of the credit score, and map the comprehensive credit score S to the corresponding credit rating according to a pre-set threshold.

[0048] The step of fusing S 1 、S 2 and S 3 to obtain a comprehensive credit score S includes the following steps: construct a judgment matrix A=(a ij ) n×n , where a ij is the importance ratio of platform i relative to platform j. For example, a ij= 1 indicates that platform i and platform j are equally important, a ij = 3 indicates that platform i is slightly more important than platform j, a ij = 5 indicates that platform i is significantly more important than platform j, a ij The value can be determined through expert judgment or other evaluation methods; calculate the product M of each row element of the judgment matrix A i , Calculate M i The cube root of For Perform normalization to obtain the weight vector ω = (ω 1 , ω 2 , ω 3 ), Calculate the comprehensive credit score S,

[0049] In an optional embodiment, the first credit score S of the rental user on the e-commerce platform can be calculated through the following method steps 1 : Extract the e-commerce platform shopping data from the user data returned by the smart contract, and extract the credit features related to the rental user from the e-commerce platform shopping data. Convert the credit features related to the rental user into the first credit score S through the following formula 1 :

[0050] S 1 = ω 1 * N(f 1 ) + ω 2 * N(f 2 ) + ω 3 * N(f 3 ) + ω 4 * N(f 4 ),

[0051] Among them, f 1 Is the consumption frequency within a period of time. The more shopping times, the more active the user is, and the more likely the user has consumption ability and creditworthiness. Therefore, the higher the consumption frequency f 1 , the higher the score of the first credit score S 1 , f 2 Is the average consumption amount within a period of time, f 3 Is the number of types of goods purchased within a period of time. A higher average consumption amount f 2 And a larger number of types of purchased goods f 3 May imply that the user has strong consumption ability. The score of the first credit score S 1 Is proportional to the average consumption amount f 2 And the number of types of purchased goods f 3 , f 4is the retention rate of purchased goods within a period of time, and the retention rate f 4 represents the proportion of goods retained after being purchased by users. Contrary to the return rate, the retention rate f 4 is higher, the first credit score S 1 has a higher score, ω 1 、ω 2 、ω 3 and ω 4 are the weights of each feature respectively, and satisfy ω 1 +ω 2 +ω 3 +ω 4 =1, N(f i ) is an index after normalizing the credit feature f i related to rental users. Its normalization formula is:

[0052]

[0053] where, min(f i ) is the minimum value selected after statistically analyzing the credit feature f i of a certain number of users on the e-commerce platform, and max(f i ) is the maximum value selected after statistically analyzing the credit feature f i of a certain number of users on the e-commerce platform.

[0054] In an optional embodiment, the second credit score S of rental users on the financial platform can be calculated through the following method steps 2 : Collect the financial data X of the training sample users on the financial platform, preprocess the financial data X, and normalize the feature variables in the financial data X to the interval [0, 1]. The feature variables in the financial data X include transaction records, credit records, asset status, and liability status, etc.; Select the Gaussian kernel function as the kernel function, and determine the optimal penalty parameter and kernel function parameter through cross-validation; Divide the credit grades of the training sample users into different categories y, and use the data samples (X, y) as the training sample set to train the support vector machine model to obtain the decision function f(x). Extract the financial data X new related to the rental users from the user data returned by the smart contract new , and input the financial data X new into the decision function to calculate f(X new ); Calculate the second credit score S 2 of the rental users on the financial platform according to f(X 2 ), S new =k×f(X

[0055] The formula of the decision function f(x) is as follows:

[0056]

[0057] where x is the input feature vector, n is the number of training samples, α i is the Lagrange multiplier, and y i is the credit rating category of the training sample, K(x i , x) is the Gaussian kernel function, and x i is the feature vector of the i-th training sample, and b is the bias term.

[0058] In an optional embodiment, the third credit score S of the rental user on the credit rating platform can be calculated through the following method steps 3 : Obtain the original credit score S of the rental user on the credit rating platform c and the source of the score data. Search for verification data related to the score data source in the financial data X new related to the rental user, and verify the original credit score S c using the verification data to obtain the score authenticity verification value V. Calculate the third credit score S 3 based on the score authenticity verification value V, where S 3 = V * S c ;

[0059] The calculation formula for the score authenticity verification value V is as follows:

[0060]

[0061] where n is the number of different data items in the verification data, w i is the weight of the i-th data item, which is set according to its importance for authenticity judgment, and m i is the matching degree between the score data source and the i-th data item in the verification data, with a value range between [0, 1], where 1 represents a perfect match and 0 represents no match. For example, for the income stream data in the score data source of the original credit score S c , assuming the monthly income stream provided in the score data source is B yuan and the monthly income stream in the financial data X new is A yuan, the calculation formula for the matching degree of the monthly income stream is:

[0062] After the rental platform calculates the credit rating of the rental user based on the user data returned by the smart contract, the rental platform sets corresponding rental conditions for the rental user according to the credit rating. The steps of setting corresponding rental conditions for the rental user according to the credit rating include: The rental platform sets a corresponding rental deposit D for the rental user according to the credit rating. The calculation formula for the rental deposit D is as follows:

[0063]

[0064] Among them, D b is the basic deposit amount, N d is the historical default times of the rental user, S i is the income stability coefficient, and its value range is between [0, 1]. The closer S i is to 1, the more stable the income is. N r is the historical rental times of the rental user, and α, β, and γ are weight coefficients. D s is the basic deposit discount rate, and R is the credit rating, and its value range is between [0, 1].

[0065] The embodiments of this specific implementation manner are all preferred embodiments of this application, and do not limit the protection scope of this application accordingly. The same components are denoted by the same reference numerals. Therefore, all equivalent changes made according to the structure, shape, and principle of this application shall be covered within the protection scope of this application.

Claims

1. An online leasing method based on blockchain, characterized in that: include: Build a blockchain, formulate a unified data interface standard, access user data of related platforms other than the leasing platform, and set up authorization permissions for the leasing platform to access user data of related platforms through smart contracts. The related platforms include e-commerce platforms, financial platforms, and credit rating platforms; After the rental user submits a rental request, the rental platform sends an access request for user data to the smart contract on the blockchain. After the request is verified, the smart contract searches for the relevant platform on the blockchain to access the user data related to the rental user in the blockchain, and returns the user data to the rental platform; The leasing platform calculates the credit rating of the leasing user based on the user data returned by the smart contract, and sets corresponding leasing conditions for the leasing user based on the credit rating.

2. According to claim 1, an online leasing method based on blockchain is characterized in that: The leasing platform calculates the credit rating of the leasing user based on the user data returned by the smart contract, including: respectively calculating the leasing user's first credit score S1 on the e-commerce platform, the second credit score S2 on the financial platform, and the third credit score S3 on the credit rating platform, and fusing S1, S2 and S3 to obtain a comprehensive credit score S, and mapping the comprehensive credit score S to the corresponding credit rating according to a pre-set threshold.

3. According to the blockchain-based online leasing method of claim 2, it is characterized in that: The fusion of S1, S2 and S3 to obtain the comprehensive credit score S includes: constructing a judgment matrix A = (a ij ) n×n , where a ij is the importance ratio of platform i to platform j; calculate the product M of each row element of the judgment matrix A i , Calculate M i The cube root of right Normalization is performed to obtain the weight vector ω=(ω1,ω2,ω3), Calculate the comprehensive credit score S, 4. According to claim 3, an online leasing method based on blockchain is characterized in that: The calculating of the first credit score S1 of the leased user on the e-commerce platform includes: extracting the e-commerce platform shopping data from the user data returned by the smart contract, extracting the credit features related to the leased user from the e-commerce platform shopping data, and converting the credit features related to the leased user into the first credit score S1 by the following formula: S1=ω1*N(f1)+ω2*N(f2)+ω3*N(f3)+ω4*N(f4), Among them, f1 is the consumption frequency within a period of time, f2 is the average consumption amount within a period of time, f3 is the number of types of goods purchased within a period of time, f4 is the retention rate of goods purchased within a period of time, ω1, ω2, ω3 and ω4 are the weights of each feature, and they satisfy ω1+ω2+ω3+ω4=1, N(f i ) is the credit feature f related to the rental user i The normalized index has the following normalization formula: Among them, min(f i ) is the credit characteristics f of a certain number of users on the e-commerce platform i The minimum value filtered out after statistics, max(f i ) is the credit characteristics f of a certain number of users on the e-commerce platform i The maximum value selected after statistics.

5. According to claim 3, an online leasing method based on blockchain is characterized in that: The calculation of the second credit score S2 of the leasing user on the financial platform includes: collecting financial data X of the training sample users on the financial platform, preprocessing the financial data X, and normalizing the characteristic variables in the financial data X to the interval [0,1]; selecting a Gaussian kernel function as the kernel function, and determining the optimal penalty parameter and kernel function parameter through cross-validation; dividing the credit ratings of the training sample users into different categories y, and training the support vector machine model using the data sample (X, y) as the training sample set to obtain a decision function f(x), and extracting the financial data X related to the leasing user from the user data returned by the smart contract new , the financial data X new Input the decision function to calculate f(X new ); According to f(X new ) Calculate the second credit score S2 of the rental user on the financial platform, S2 = k × f (X new )+c, where k is the conversion coefficient and c is the offset; The formula of the decision function f(x) is as follows: Among them, x is the input feature vector, n is the number of training samples, α i is the Lagrange multiplier, y i is the credit rating category of the training sample, K(x i ,x) is the Gaussian kernel function, x i is the feature vector of the i-th training sample, and b is the bias term.

6. According to claim 5, an online leasing method based on blockchain is characterized in that: The calculation of the third credit score S3 of the rental user on the credit rating platform includes: obtaining the original credit score S of the rental user on the credit rating platform c The source of the rating data is the financial data X related to the rental users. new Search for verification data related to the source of the scoring data, and use the verification data to verify the original credit score S c Verify and obtain the score authenticity verification value V. Calculate the third credit score S3 based on the score authenticity verification value V, S3 = V*S c ; The calculation formula of the score authenticity verification value V is as follows: Where n is the number of different data items in the verification data, w i is the weight of the ith data item, which is set according to its importance to the authenticity judgment, and m i It is the matching degree between the scoring data source and the i-th data item in the verification data. The value range is between [0,1], where 1 indicates a complete match and 0 indicates a complete mismatch.

7. The blockchain-based online leasing method according to claim 1, characterized in that: The setting of corresponding rental conditions for the rental user according to the credit rating includes: the rental platform setting a corresponding rental deposit D for the rental user according to the credit rating, and the calculation formula of the rental deposit D is as follows: Among them, D b is the basic deposit amount, N d is the number of historical defaults of the rental user, S i is the income stability coefficient, ranging from [0,1], S i The closer it is to 1, the more stable the income is. r is the historical rental times of the rental user, α, β and γ are weight coefficients, D s is the basic deposit discount rate, R is the credit rating, and its value range is [0,1].

8. The blockchain-based online leasing method according to claim 1, characterized in that: The leasing platform rewards the relevant platforms based on the user data of the relevant platforms connected to the blockchain, and calculates the reward amount using the following formula: I=α×(ie βD -1)+γ×Q 2 +δ×ln(F+1)+ε×(1-e -λT ), Among them, I is the reward amount, D is the amount of user data connected to the blockchain by the relevant platform, Q is the quality score of the user data connected to the blockchain by the relevant platform, and the value range is between [0,1]. F is the data sharing frequency of the relevant platform within a certain period of time, T is the duration of the relevant platform continuously connecting data to the blockchain, and α, β, γ, δ, ε and λ are adjustment parameters.

9. The blockchain-based online leasing method according to claim 1, characterized in that: The user data of the related platforms other than the access leasing platform includes: encrypting the user data of the related platforms, storing the encrypted data in a local or distributed storage system, and storing the hash value of the encrypted data on the blockchain.

10. The blockchain-based online leasing method according to claim 9, characterized in that: The smart contract searches for user data related to the rental user in the blockchain that is accessed by the relevant platform on the blockchain, and returns the user data to the rental platform, including: the smart contract searches for a hash value of the user data related to the rental user on the blockchain, obtains encrypted data from a local or distributed storage system, recalculates the hash value and compares it with the hash value stored on the blockchain, verifies the integrity of the data, then decrypts the encrypted data, and returns the decrypted user data to the rental platform.

Images