Real estate registration system and real estate registration method

Abstract

The invention relates to the technical field of real estate registration, and particularly discloses a real estate registration system and a real estate registration method. The method comprises the following steps: acquiring associated information based on real estate registration information, and verifying the real estate registration information based on the associated information; selecting a management node, an uplink node and a verification node; after the real estate registration information passes the verification, splitting the associated information into a plurality of sub-data, and generating a data block by the uplink node according to the real estate registration information and each sub-data; the verification node returns the verification result of the data block to the management node; if the management node determines that the data block is legal, real estate registration information is stored in an upper-layer block chain, and associated information is stored in one of bottom-layer block chains; the data block in the upper block chain has a block value pointing to the data block for storing the associated information. According to the method, on the basis of the block chain, the security and reliability of the data are ensured by constructing a multi-level node system.

Description

Technical Field

[0001] The present application relates to the technical field of real estate registration, and in particular to a real estate registration system and a real estate registration method. Background Art

[0002] Currently, by establishing a real estate big data platform, we can collect multi-dimensional data including real estate ownership, transaction records, tax information, historical changes, etc. During the real estate registration process, relevant departments can query, monitor and analyze the accuracy of real estate registration data in real time on the platform.

[0003] For example, the Chinese patent document with publication number CN118195320A discloses a real estate registration information sharing management system and method based on big data technology. This method uses deep learning artificial intelligence technology to perform semantic understanding and joint clustering analysis on the collection of real estate registration information to capture the semantic association characteristics of the entire sample domain of real estate registration information, and to mine the potential risk factors in the real estate registration information collection, thereby achieving risk assessment of real estate registration information, and then setting open permissions based on the risk assessment results. This can improve the utilization efficiency and value of real estate registration information.

[0004] However, in the real estate registration system, the authenticity and security of data are easily affected by human tampering, data loss or system failure. If the registration information is tampered with or entered incorrectly, it may lead to disputes over real estate ownership and even legal proceedings. Summary of the invention

[0005] To solve the above problems, the present application provides a real estate registration system and a real estate registration method.

[0006] In order to achieve the above-mentioned purpose of the invention, the present invention proposes a method for real estate registration, comprising: Acquire real estate registration information, acquire associated information based on the real estate registration information, and verify and supplement missing values ​​of the real estate registration information based on the associated information; Build a node library, the node library includes multiple basic nodes, select multiple basic nodes from the node library and use them as management nodes, on-chain nodes and verification nodes respectively; After the real estate registration information is verified, the associated information is split into multiple sub-data, and the on-chain node generates a data block according to the real estate registration information and each sub-data; Each of the verification nodes verifies the block value of the data block and returns the verification result to the management node; Constructing an upper blockchain and multiple underlying blockchains, if the management node determines that the data block is legal according to the verification result, the real estate registration information is stored in the upper blockchain, and the associated information is stored in one of the underlying blockchains; The data block in the upper-layer blockchain has a block value pointing to the data block storing the associated information.

[0007] Furthermore, verifying the real estate registration information includes the following steps: If the associated information is different from the real estate registration information or the real estate is in an abnormal state, the verification of the real estate registration information fails. Otherwise, the pattern analysis is continued based on the associated information to obtain the real estate registration pattern. If the real estate registration pattern is judged to be normal, the verification passes.

[0008] Further, performing the registration pattern analysis includes the following steps: Obtain the geographic location in the associated information, generate a detection area with the geographic location as the center, obtain historical registration information of all real estate in the detection area within a preset time period, construct a registration time series based on the historical registration information, perform trend analysis on the registration time series, obtain the trend component sequence therein, perform cluster analysis on the trend component sequence based on a clustering algorithm, obtain multiple clusters, perform anomaly detection on each of the clusters based on an outlier factor detection algorithm, identify abnormal transaction points in each cluster, construct an abnormal prediction model based on the associated information corresponding to the abnormal transaction points, identify the associated information this time based on the abnormal prediction model, and determine whether the real estate registration mode is normal.

[0009] Furthermore, splitting the associated information includes the following steps: The associated information is converted into a high-dimensional tensor, wherein the high-dimensional tensor includes multiple independent data, a timestamp is generated based on the current time, the high-dimensional tensor is converted into multi-dimensional key data using the timestamp and a preset private key using a hash algorithm, each sub-data in the multi-dimensional key data is randomly perturbed to generate a multi-dimensional comprehensive key, each sub-data in the multi-dimensional comprehensive key is integerized to obtain multiple mapping integers, a Latin square is constructed based on the mapping integers, and a row or a column in the Latin square is randomly selected as a splitting scheme for the associated information, wherein the splitting scheme includes the number of splits of the independent data.

[0010] Furthermore, verifying the data block includes the following steps: Based on preset rules, the verification nodes are divided into multiple consensus units, each of which includes at least two verification nodes. The first hash value is defined as the value generated by the on-chain node. Each verification node calculates the sub-data to generate a second hash value. The second hash value that appears most frequently in the same consensus unit is output as the consensus result of the consensus unit. The frequency of occurrence of various second hash values ​​output by each consensus unit is counted, and the second hash value with the largest frequency of occurrence is used as the third hash value. If the third hash value is the same as the first hash value, the verification of the data block is passed.

[0011] Furthermore, dividing the verification node comprises the following steps: Calculating a trust score of each of the verification nodes, setting multiple value ranges, dividing the verification nodes into multiple initial groups based on the value ranges where the trust scores are located, obtaining a first number of the verification nodes in each of the initial groups, defining the initial groups whose first number is less than a third threshold as a first group, and defining the initial groups whose first number is greater than a fourth threshold as a second group; A computing model is constructed based on a linear regression algorithm. The computing model predicts the current consensus score of each verification node in the second group based on its historical consensus record. The verification nodes in the second group are reordered based on the consensus score. The verification nodes at a preset position in the reordered second group are inserted into the first group to complete the division of the verification nodes.

[0012] Further, calculating the trust score includes the following steps: After the data block is uploaded to the chain, a consensus matrix is ​​constructed based on the verification results of the verification node, and a consistency analysis is performed on the consensus matrix to obtain a consistency value of each verification node. The consistency value is normalized to obtain a normalized value, which is updated based on the normalized value and used as the trust score of the current verification node.

[0013] Furthermore, the verification nodes are divided into high-trust nodes and low-trust nodes based on the size of the trust score. If the high-trust node is combined with the same high-trust node multiple times in a row, the trust score of the high-trust node is reduced. If the low-trust node is combined with the same low-trust node multiple times in a row, the trust score of the low-trust node is increased.

[0014] Furthermore, the related information includes personal information, real estate mortgage data, real estate seizure data, and real estate price prediction data.

[0015] The present application also provides a real estate registration system, which is used to implement the real estate registration method described above, and the system includes: A collection module, used to obtain real estate registration information, obtain related information based on the real estate registration information, and verify the real estate registration information and supplement missing values ​​based on the related information; The bottom layer module is used to build a node library, which includes multiple basic nodes. Multiple basic nodes are selected from the node library and used as management nodes, on-chain nodes and verification nodes respectively; The verification module, after the real estate registration information is verified, splits the associated information into multiple sub-data, the on-chain node generates a data block according to the real estate registration information and each of the sub-data, each of the verification nodes verifies the block value of the data block, and returns the verification result to the management node; The chain module constructs an upper-layer blockchain and multiple underlying blockchains. If the management node determines that the data block is legal based on the verification result, the real estate registration information is stored in the upper-layer blockchain, and the related information is stored in one of the underlying blockchains. The data block in the upper-layer blockchain has a block value pointing to the data block storing the related information. Beneficial Effects

[0016] The present invention ensures the security and reliability of data by constructing a multi-level node system based on the blockchain. Management nodes, on-chain nodes and verification nodes cooperate with each other to ensure the effective storage and legal verification of data. In particular, when data is split and chained, the associated information is divided into multiple sub-data for processing, so that each verification node can only access part of the information, further improving data privacy and security. The data blocks stored in the upper blockchain contain data block values ​​pointing to the associated information in the underlying blockchain, so that when retrieving real estate registration information, the corresponding associated data blocks can be quickly located, greatly improving the data retrieval speed and improving the user's query experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0018] Figure 1 A flowchart of the steps of a method for real estate registration for this application; Figure 2 This is a structural diagram of a real estate registration system for this application. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0020] It is understood that the terms "first", "second", etc. used in this application may be used herein to describe various elements, but unless otherwise specified, these elements are not limited by these terms. These terms are only used to distinguish a first element from another element. For example, without departing from the scope of this application, a first xx script may be referred to as a second xx script, and similarly, a second xx script may be referred to as a first xx script.

[0021] It should be noted that all data acquisition actions in this application are carried out in compliance with the relevant data protection laws and policies of the country where the application is located, and with the authorization given by the owner of the corresponding device.

[0022] like Figure 1 As shown, a method for real estate registration includes: S1: Obtain real estate registration information, obtain related information based on the real estate registration information, and verify the real estate registration information and supplement missing values ​​based on the related information.

[0023] In this embodiment, the associated information includes personal information, real estate mortgage data, real estate seizure data, and real estate price prediction data.

[0024] The real estate registration information obtained first is manually filled in information, including but not limited to the geographical location, area size, personal information, etc. of the real estate. Based on personal information and real estate information, the system obtains the ownership information, geographical location, transaction records, tax information, historical change records and other related information of the real estate from databases from different sources through the API interface, so as to perform secondary verification on the filled in information, and can also supplement the original information based on the searched information. For example, if the manually filled in information does not include the area size, the system can supplement the area size through the aggregation results. In addition, the transaction situation of the real estate can be analyzed through the historical transaction information in the related information to determine whether there are abnormal behaviors of the real estate, such as abnormal buying and selling, price manipulation, etc.

[0025] S2: Build a node library, which includes multiple basic nodes. Select multiple basic nodes from the node library and use them as management nodes, on-chain nodes and verification nodes respectively.

[0026] The basic nodes are specifically composed of computers. When data needs to be uploaded to the chain, multiple basic nodes are first randomly selected from the node library, and one basic node is designated as the management node, one basic node as the upload node, and multiple basic nodes as verification nodes. Among them, the upload node is responsible for converting the data that needs to be uploaded to the chain into data blocks, and after the data blocks are verified, the data blocks are stored in the corresponding blockchain. The verification node is responsible for verifying the data blocks generated by the upload node, checking whether the data complies with the rules and returning the verification results. The management node decides which data can be uploaded to the chain based on the verification results.

[0027] S3: After the real estate registration information is verified, the associated information is split into multiple sub-data, and the on-chain node generates a data block based on the real estate registration information and each of the sub-data.

[0028] S4: Each of the verification nodes verifies the block value of the data block and returns the verification result to the management node.

[0029] In order to ensure the security of the data, before verifying the related information, the related information is first split into multiple sub-data. For example, the tax information is split into two sub-data, so that each sub-data only contains a part of the original information. When the sub-data is sent to each verification node for verification, the verification node cannot know the original complete information, thereby ensuring the security of the data.

[0030] S5: Construct an upper-layer blockchain and multiple underlying blockchains. If the management node determines that the data block is legal based on the verification result, the real estate registration information is stored in the upper-layer blockchain, and the related information is stored in one of the underlying blockchains.

[0031] S6: The data block in the upper-layer blockchain has a block value pointing to the data block storing the associated information.

[0032] The upper blockchain is used to store the core data of real estate registration information, including geographic location, area size, right holder information, etc., while the underlying blockchain stores related information, such as tax information, historical transaction records, etc. The present invention constructs multiple underlying blockchains to store related information, which can avoid storing all related information on one blockchain, and in the later retrieval, the shorter blockchain has a higher information retrieval speed, thereby shortening the data search time.

[0033] The specific verification process is to send the sub-data to the chain node and the verification node. The chain node converts the sub-data into a hash value based on the hash algorithm (SHA-256), and then sends the corresponding hash value to each verification node. The verification node converts the sub-data into a hash value based on the same hash algorithm (SHA-256). If the calculation result is consistent with the calculation result of the chain node, the verification node verifies the block value, otherwise the verification fails. Finally, the management node determines whether the data block is legal based on the verification results of each verification node. If it is legal, the chain node is notified to upload the data block corresponding to the real estate registration information to the upper blockchain, and the data block of each sub-data under the corresponding associated information is uploaded to the underlying blockchain.

[0034] The data block in the upper blockchain has a block value pointing to the associated information data block. When retrieving information subsequently, the corresponding data block is first searched in the upper blockchain. The data block also indicates in which underlying blockchain each sub-information under the corresponding associated information is stored and what the hash value of the specific data block is. After retrieving the data block corresponding to the real estate information in the upper blockchain, the corresponding data block storing the sub-data can be quickly found, and then the sub-data can be restored to the original associated information, which not only ensures the data retrieval speed, but also enhances the security of data storage.

[0035] The present invention first obtains the basic information of the real estate, and verifies the information twice through the associated information, ensuring the accuracy and completeness of the registration information. The system can not only verify the manually entered information, but also automatically supplement the missing items based on the associated data, thereby improving the data quality. By constructing a data storage method that combines the upper-layer blockchain with multiple underlying blockchains, it not only avoids the bottleneck of single chain storage, but also improves the efficiency of data retrieval. In later data searches, shorter blockchains provide faster access speeds, optimize the query process, and shorten response time.

[0036] Based on the blockchain, the present invention ensures the security and reliability of data by constructing a multi-level node system. Management nodes, on-chain nodes and verification nodes cooperate with each other to ensure the effective storage and legal verification of data. In particular, when data is split and chained, the associated information is divided into multiple sub-data for processing, so that each verification node can only access part of the information, further improving data privacy and security. The data blocks stored in the upper blockchain contain data block values ​​pointing to the associated information in the underlying blockchain, so that when retrieving real estate registration information, the corresponding associated data blocks can be quickly located. Greatly improve the data retrieval speed and enhance the user's query experience.

[0037] In this embodiment, verifying the real estate registration information includes the following steps: If the associated information is different from the real estate registration information or the real estate is in an abnormal state, the verification of the real estate registration information fails. Otherwise, the pattern analysis is continued based on the associated information to obtain the real estate registration pattern. If the real estate registration pattern is judged to be normal, the verification passes.

[0038] If the associated information of the real estate is inconsistent with the registration information, such as the geographical location, area, right holder and other basic information, the real estate registration information is considered to be abnormal and the verification fails. If the real estate is in an abnormal state (sealed, mortgaged), the verification fails. If the real estate has the above situation, the transaction situation of the real estate will be further analyzed to obtain the registration mode of the real estate and further determine whether there is any abnormality in the real estate.

[0039] In this embodiment, performing registration mode analysis includes the following steps: Obtain the geographic location in the associated information, generate a detection area with the geographic location as the center, obtain historical registration information of all real estate in the detection area within a preset time period, construct a registration time series based on the historical registration information, perform trend analysis on the registration time series, obtain the trend component sequence therein, perform cluster analysis on the trend component sequence based on a clustering algorithm, obtain multiple clusters, perform anomaly detection on each of the clusters based on an outlier factor detection algorithm, identify abnormal transaction points in each cluster, construct an abnormal prediction model based on the associated information corresponding to the abnormal transaction points, identify the associated information this time based on the abnormal prediction model, and determine whether the real estate registration mode is normal.

[0040] First, the geographical location of the real estate is extracted from the associated information of the real estate. The geographical location may include the longitude, latitude, city, region and other information of the real estate. With the geographical location as the center, a detection area is generated. The size of the area can be adjusted according to specific needs. In this embodiment, it is an area with a radius of 10km centered on the real estate. After the detection area is generated, the system further filters out the historical registration information of all real estate in the area from the historical registration records within a preset time period (nearly six months). Including but not limited to the real estate ID, registration time, registration price, registration parties and other information at each registration. Based on the historical registration information obtained, the system will organize the data according to the registered timestamp information to construct a registration time series for each real estate in the detection area. For example, the registration records of a real estate in the past six months constitute the registration time series of the real estate.

[0041] Based on the registration time series, the time series data is analyzed through trend analysis technology to obtain the trend component. Trend analysis can use a variety of methods, such as moving average method, exponential weighted average method, etc. to remove seasonal and random fluctuations. For example, in a certain registration time series, the registration price in January is 1 million yuan, and in February it is 1.05 million yuan, with a small change, but the subsequent registration price changes are more drastic, which may affect the analysis. After smoothing by the 3-period moving average method, the fluctuation of the data becomes more stable. For example, the original price in June was 1.3 million yuan, but after smoothing, the 3-period moving average result was 1.3167 million yuan, which made the trend of price changes smoother and eliminated the interference caused by single-month fluctuations.

[0042] After the trend components are extracted, the system uses clustering algorithms to analyze these trend components. For example, K-means, DBSCAN, etc. are used to group the trend component sequences of multiple real estates into several clusters, each cluster representing a group of real estates with similar registration trends. The historical registration behaviors under these clusters have common characteristics. After completing the cluster analysis, the system then uses outlier factor detection algorithms, such as the LOF algorithm, to perform anomaly detection on each cluster. The outlier factor detection algorithm can identify those transaction points that are significantly different from most registration behaviors, that is, abnormal transaction points in each cluster. Abnormal transaction points may manifest as abnormal registration frequencies or other abnormal behaviors occurring within a specific time period.

[0043] Finally, the association information of the abnormal transaction points and the remaining transaction points is used as a training set to construct an abnormal prediction model for registration mode judgment, that is, the historical registration data in the association information is used as input features to determine the registration mode of the real estate. The abnormal prediction model can be constructed using random forests and support vector machines. If the registration mode is normal after analysis by the abnormal prediction model, the system will allow the real estate to continue to be registered and put on the chain. If the registration mode is abnormal, the system will issue an early warning and recommend further review of the registration behavior of the real estate.

[0044] In this embodiment, splitting the associated information includes the following steps: The associated information is converted into a high-dimensional tensor, wherein the high-dimensional tensor includes multiple independent data, a timestamp is generated based on the current time, the high-dimensional tensor is converted into multi-dimensional key data using the timestamp and a preset private key using a hash algorithm, each sub-data in the multi-dimensional key data is randomly perturbed to generate a multi-dimensional comprehensive key, each sub-data in the multi-dimensional comprehensive key is integerized to obtain multiple mapping integers, a Latin square is constructed based on the mapping integers, and a row or a column in the Latin square is randomly selected as a splitting scheme for the associated information, wherein the splitting scheme includes the number of splits of the independent data.

[0045] Assume that the associated information retains the real estate ID, right holder information, and tax information, and then convert the above types into a 3-dimensional tensor, where dimension 1 is the real estate ID, dimension 2 is the right holder information, and dimension 3 is the tax information. The data in each dimension is the independent data mentioned above. By converting the associated information into a high-dimensional tensor, it is convenient for subsequent data processing. Use the hash algorithm and the preset private key to convert each independent data into a hash value. The conversion formula is Hash=H(T,sec,Data). In the formula, pk is the converted hash value, H() represents the hash function, such as the SHA-256 algorithm, T is the timestamp, such as 202311301511, sec is the preset private key, which can be a string of numbers, and Data is the independent data that needs to be converted, such as the binary representation of the right holder information.

[0046] After the conversion, three 256-bit hexadecimal hash values ​​can be obtained. The three hash values ​​combine the multidimensional key data, expressed as (Hash1, Hash2, Hash3). Then, each item in the multidimensional key data is randomly perturbed to obtain a multidimensional comprehensive key, expressed as (HashA, HashB, HashC). The random perturbation method is, for example, to randomly change the value of a certain item in the hash value and swap the positions of multiple hash values. Finally, the hexadecimal hash value is converted to decimal, and then the first value of each integer is taken as the mapping integer. For example, the integers after the hexadecimal hash value conversion are 106, 31, and 24, respectively. The first word numbers of each integer, 1, 3, and 2, are taken as mapping integers. This can avoid excessive splitting of associated data, which affects the subsequent calculation efficiency. Since there are 3 mapping integers, a 3*3 Latin square is constructed here. The Latin square is a square matrix with non-repeating integers in each row and column. For example, the first row of the square is 1, 2, 3, the second row of the square is 3, 1, 2, and the third row of the square is 2, 3, 1. Then the first row is extracted as the splitting scheme. The integers contained in the first row are 1, 2, 3. The real estate ID is not split, the right holder information is split into two sub-data, and the tax information is split into three sub-data. In particular, if the Latin matrix cannot be constructed, the repeated values ​​in the mapping integers are added by 1 and the Latin square is reconstructed.

[0047] In this embodiment, verifying the data block includes the following steps: Based on preset rules, the verification nodes are divided into multiple consensus units, each of which includes at least two verification nodes. The first hash value is defined as the value generated by the on-chain node. Each verification node calculates the sub-data to generate a second hash value. The second hash value that appears most frequently in the same consensus unit is output as the consensus result of the consensus unit. The frequency of occurrence of various second hash values ​​output by each consensus unit is counted, and the second hash value with the largest frequency of occurrence is used as the third hash value. If the third hash value is the same as the first hash value, the verification of the data block is passed.

[0048] The specific process of dividing the verification nodes based on the preset rules will be introduced later. For example, there are 10 verification nodes, which are divided into 3 consensus units, where consensus unit 1 includes 3 verification nodes, consensus unit 2 includes 2 verification nodes, and consensus unit 3 includes 5 verification nodes. The verification nodes in each consensus unit calculate the corresponding second hash value based on the received sub-data. For consensus unit 3, 4 verification nodes output the second hash value H1 and one verification node outputs the second hash value H2, then the consensus result of the consensus unit is the second hash value H1. Afterwards, if consensus unit 1 outputs the second hash value H2 and consensus unit 2 outputs the second hash value H1, the second hash value H1 is used as the third hash value. If the third hash value H1 is the same as the first hash value, the verification of the data block is passed. If they are not the same, the verification of the data block is not passed.

[0049] In this embodiment, dividing the verification node includes the following steps: Calculate the trust score of each of the verification nodes, set multiple numerical ranges, divide the verification nodes into multiple initial groups based on the numerical ranges where the trust scores are located, obtain a first number of the verification nodes in each of the initial groups, define the initial group whose first number is less than a third threshold as a first group, and define the initial group whose first number is greater than a fourth threshold as a second group.

[0050] A computing model is constructed based on a linear regression algorithm. The computing model predicts the current consensus score of each verification node in the second group based on its historical consensus record. The verification nodes in the second group are reordered based on the consensus score. The verification nodes at a preset position in the reordered second group are inserted into the first group to complete the division of the verification nodes.

[0051] The method for calculating the trust score is introduced later. This embodiment sets three numerical ranges. Then, according to the specific trust score of each verification node, all verification nodes are divided into corresponding groups. For example, if the numerical range is 0-0.3 and the trust score of the verification node is 0.2, it will be divided into this group. The second threshold is set to 2, and the third threshold is set to 3. The initial group containing less than 2 verification nodes is defined as the first group, and the initial group containing more than 3 verification nodes is defined as the second group. The purpose of this step is to locate the first group with too few verification nodes and the second group with too many nodes, so as to facilitate the subsequent division of the verification nodes in the second group into the first group.

[0052] The calculation mode based on the linear regression model of this embodiment is a linear regression function, specifically G=β0+β1*m+β2*m2-β3*m3, where G is the consensus score, β0 is the intercept term, β1, β2, β3 are the corresponding regression coefficients, m1 is the number of historical consensuses, m2 is the number of successful consensuses, and m3 is the number of failed consensuses. After calculating the consensus score of each verification node, the verification nodes in the second group are sorted from small to large based on the size of the consensus score, and the verification nodes in the middle position are selected and inserted into the second group. The purpose is to select the consensus score to be inserted into the first group. This can avoid the verification nodes with a large number of successes or identification times from being inserted into the first group, so that the verification results of the first group have a positive or negative orientation.

[0053] In this embodiment, calculating the trust score includes the following steps: After the data block is uploaded to the chain, a consensus matrix is ​​constructed based on the verification results of the verification node, and a consistency analysis is performed on the consensus matrix to obtain a consistency value of each verification node. The consistency value is normalized to obtain a normalized value, which is updated based on the normalized value and used as the trust score of the current verification node.

[0054] After the data block is successfully uploaded to the chain, the verification nodes participating in the consensus and the verification results of each sub-data are obtained, where the verification is 1 if it passes and 0 if it fails. A consensus matrix is ​​constructed based on the verification nodes and the verification results. In the consensus matrix, the value of the i-th row and j-th column represents the verification result of the verification node i on the sub-data j. Then, a consistency analysis is performed based on the above consensus matrix. The purpose of the consistency analysis is to evaluate the consistency of the verification results of each node with the results of other nodes, that is, whether the node tends to be consistent with other verification nodes, or consistent with the majority of nodes.

[0055] This embodiment calculates the trust score based on voting consistency. If the verification result of a node is consistent with that of most other nodes, the trust score of the node is high; otherwise, it is low. The trust score can be calculated using the following formula: ,in, Represents the consistency value of verification node i, is the number of times verification node i and verification node j are consistent, and N is the number of times they participate in verification. For example, for 2 sub-data, each verification node will perform 2 verifications in total, then N is 2. In the first verification, verification node 1 is consistent with verification nodes 2 and 3, then and The value of is 1. In the second verification, verification node 1 and verification node 2 are consistent. The value of is 1, The value of is 0. After calculation, the consistency score of verification node 1 is (1 / 2)(1+1+1)=1.5.

[0056] Since the range of consistency values ​​may not be uniform, in order to ensure that the final trust score is more intuitive and fair, the consistency values ​​need to be normalized. The purpose of normalization is to adjust the range of all consistency values ​​to a standard interval (such as 0 to 1) to facilitate comparison and update of trust scores. This embodiment uses the minimum-maximum normalization method for normalization, and the final normalized value is the trust score.

[0057] In this embodiment, the verification nodes are divided into high-trust nodes and low-trust nodes based on the size of the trust score. If the high-trust node is combined with the same high-trust node multiple times in a row, the trust score of the high-trust node is reduced. If the low-trust node is combined with the same low-trust node multiple times in a row, the trust score of the low-trust node is increased.

[0058] First, a critical threshold is set, and verification nodes with a trust score greater than the critical threshold are classified as high-trust nodes, and verification nodes with a trust score less than or equal to the critical threshold are classified as high-trust nodes. If a high-trust node is combined with the same high-trust node multiple times into a group during multiple verifications, in order to increase the randomness of the combination, the trust score of one of the high-trust nodes is reduced so that it can be combined with other verification nodes next time. Similarly, if a low-trust node is combined with the same low-trust node multiple times into a group, the trust score of one of the low-trust nodes is increased.

[0059] like Figure 2 As shown, the present application also provides a real estate registration system, which is used to implement the real estate registration method described above, and the system includes: A collection module, used to obtain real estate registration information, obtain related information based on the real estate registration information, and verify the real estate registration information and supplement missing values ​​based on the related information; The bottom layer module is used to build a node library, which includes multiple basic nodes. Multiple basic nodes are selected from the node library and used as management nodes, on-chain nodes and verification nodes respectively; The verification module, after the real estate registration information is verified, splits the associated information into multiple sub-data, the on-chain node generates a data block according to the real estate registration information and each of the sub-data, each of the verification nodes verifies the block value of the data block, and returns the verification result to the management node; The chain module constructs an upper-layer blockchain and multiple underlying blockchains. If the management node determines that the data block is legal based on the verification result, the real estate registration information is stored in the upper-layer blockchain, and the related information is stored in one of the underlying blockchains. The data block in the upper-layer blockchain has a block value pointing to the data block storing the related information.

[0060] It should be understood that the various technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the various technical features in the above-mentioned embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0061] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for real estate registration, characterized in that: include: Acquire real estate registration information, acquire associated information based on the real estate registration information, and verify and supplement missing values ​​of the real estate registration information based on the associated information; Build a node library, the node library includes multiple basic nodes, select multiple basic nodes from the node library and use them as management nodes, on-chain nodes and verification nodes respectively; After the real estate registration information is verified, the associated information is split into multiple sub-data, and the on-chain node generates a data block according to the real estate registration information and each sub-data; Each of the verification nodes verifies the block value of the data block and returns the verification result to the management node; Constructing an upper blockchain and multiple underlying blockchains, if the management node determines that the data block is legal according to the verification result, the real estate registration information is stored in the upper blockchain, and the associated information is stored in one of the underlying blockchains; The data block in the upper-layer blockchain has a block value pointing to the data block storing the associated information.

2. The method according to claim 1, characterized in that Verifying the real estate registration information includes the following steps: If the associated information is different from the real estate registration information or the real estate is in an abnormal state, the verification of the real estate registration information fails. Otherwise, the pattern analysis is continued based on the associated information to obtain the real estate registration pattern. If the real estate registration pattern is judged to be normal, the verification passes.

3. The method according to claim 2, characterized in that Conducting a registration pattern analysis involves the following steps: Obtain the geographic location in the associated information, generate a detection area with the geographic location as the center, obtain historical registration information of all real estate in the detection area within a preset time period, construct a registration time series based on the historical registration information, perform trend analysis on the registration time series, obtain the trend component sequence therein, perform cluster analysis on the trend component sequence based on a clustering algorithm, obtain multiple clusters, perform anomaly detection on each of the clusters based on an outlier factor detection algorithm, identify abnormal transaction points in each cluster, construct an abnormal prediction model based on the associated information corresponding to the abnormal transaction points, identify the associated information this time based on the abnormal prediction model, and determine whether the real estate registration mode is normal.

4. The method according to claim 1, characterized in that: Splitting the associated information includes the following steps: The associated information is converted into a high-dimensional tensor, wherein the high-dimensional tensor includes multiple independent data, a timestamp is generated based on the current time, the high-dimensional tensor is converted into multi-dimensional key data using the timestamp and a preset private key using a hash algorithm, each sub-data in the multi-dimensional key data is randomly perturbed to generate a multi-dimensional comprehensive key, each sub-data in the multi-dimensional comprehensive key is integerized to obtain multiple mapping integers, a Latin square is constructed based on the mapping integers, and a row or a column in the Latin square is randomly selected as a splitting scheme for the associated information, wherein the splitting scheme includes the number of splits of the independent data.

5. The method according to claim 1, characterized in that Verifying the data block includes the following steps: Based on preset rules, the verification nodes are divided into multiple consensus units, each of which includes at least two verification nodes. The first hash value is defined as the value generated by the on-chain node. Each verification node calculates the sub-data to generate a second hash value. The second hash value that appears most frequently in the same consensus unit is output as the consensus result of the consensus unit. The frequency of occurrence of various second hash values ​​output by each consensus unit is counted, and the second hash value with the largest frequency of occurrence is used as the third hash value. If the third hash value is the same as the first hash value, the verification of the data block is passed.

6. The method according to claim 5, characterized in that Dividing the verification node comprises the following steps: Calculating a trust score of each of the verification nodes, setting multiple value ranges, dividing the verification nodes into multiple initial groups based on the value ranges where the trust scores are located, obtaining a first number of the verification nodes in each of the initial groups, defining the initial groups whose first number is less than a third threshold as a first group, and defining the initial groups whose first number is greater than a fourth threshold as a second group; A computing model is constructed based on a linear regression algorithm. The computing model predicts the current consensus score of each verification node in the second group based on its historical consensus record. The verification nodes in the second group are reordered based on the consensus score. The verification nodes at a preset position in the reordered second group are inserted into the first group to complete the division of the verification nodes.

7. The method according to claim 6, characterized in that Calculating the trust score includes the following steps: After the data block is uploaded to the chain, a consensus matrix is ​​constructed based on the verification results of the verification node, and a consistency analysis is performed on the consensus matrix to obtain a consistency value of each verification node. The consistency value is normalized to obtain a normalized value, which is updated based on the normalized value and used as the trust score of the current verification node.

8. The method according to claim 6, characterized in that The verification nodes are divided into high-trust nodes and low-trust nodes based on the size of the trust score. If the high-trust node is combined with the same high-trust node multiple times in a row, the trust score of the high-trust node is reduced. If the low-trust node is combined with the same low-trust node multiple times in a row, the trust score of the low-trust node is increased.

9. The method according to claim 1, characterized in that: The related information includes personal information, real estate mortgage data, real estate seizure data, and real estate price prediction data.

10. A real estate registration system, used to implement the real estate registration method according to any one of claims 1 to 9, characterized in that: A collection module, used to obtain real estate registration information, obtain related information based on the real estate registration information, and verify the real estate registration information and supplement missing values ​​based on the related information; The bottom layer module is used to build a node library, which includes multiple basic nodes. Multiple basic nodes are selected from the node library and used as management nodes, on-chain nodes and verification nodes respectively; The verification module, after the real estate registration information is verified, splits the associated information into multiple sub-data, the on-chain node generates a data block according to the real estate registration information and each of the sub-data, each of the verification nodes verifies the block value of the data block, and returns the verification result to the management node; The chain module constructs an upper-layer blockchain and multiple underlying blockchains. If the management node determines that the data block is legal based on the verification result, the real estate registration information is stored in the upper-layer blockchain, and the related information is stored in one of the underlying blockchains. The data block in the upper-layer blockchain has a block value pointing to the data block storing the related information.

Images