Real estate data integration method

Abstract

The invention relates to the technical field of information statistics, and discloses a real estate data integration method, which comprises the following steps: S1, data collection: collecting real estate data from multiple channels; s2, data processing: performing data processing on all the collected data by using an AI technology, unifying the data format, and finally performing data verification through a block chain; s3, data association: constructing a knowledge graph, and associating real estate information with traffic, education and environment data; s4, establishing a generative adversarial network, and training a generator and a discriminator; s5, database establishment: integrating and storing the calculated integrated real estate data through a cloud computing platform and a distributed database; s6, updating real-time data; and S7, data maintenance. And high-quality synthetic real estate data is generated by using the generative adversarial network through the synthetic data generation module, the scarce part of the data is supplemented, the richness and diversity of a data set are improved, and the generalization ability and prediction precision of the model are improved.

Description

Technical Field

[0001] This invention relates to the field of information statistics technology, specifically a method for integrating real estate data. Background Technology

[0002] In modern society, the real estate industry is a vital component of the national economy, and its healthy development directly impacts economic stability and people's quality of life. The management and analysis of real estate data are crucial foundations for industry operations and decision-making. However, traditional real estate data management methods suffer from numerous technical shortcomings, severely affecting the efficiency and accuracy of data utilization.

[0003] The shortcomings of existing technologies include:

[0004] Single and outdated data sources: Traditional real estate data management methods primarily rely on a single data source, such as government databases or real estate platforms. While these data sources possess a certain degree of authority and accuracy, their low update frequency makes it difficult to provide real-time market information, resulting in insufficient data timeliness. Furthermore, a single data source cannot comprehensively cover all real estate-related information, leading to insufficient data comprehensiveness and impacting the accuracy of market analysis and decision-making.

[0005] Low data processing efficiency and poor quality: Traditional data processing methods rely heavily on manual operations, including missing value handling, data deduplication, and outlier detection. These methods are not only time-consuming and labor-intensive but also prone to human error, making it difficult to guarantee data quality. The lack of unified format standards for data from different sources increases the complexity and inconsistency of data integration, further reducing data processing efficiency and accuracy.

[0006] Lack of data generation and supplementation mechanisms: Existing technologies typically lack effective solutions to the problem of data scarcity. Current methods are unable to generate high-quality synthetic data to supplement insufficient data, resulting in a lack of richness and diversity in datasets, which severely impacts the effectiveness of subsequent data analysis and machine learning model training.

[0007] Data privacy and security issues: Protecting user privacy and ensuring data security are increasingly important requirements in data management. Traditional methods lack effective privacy protection mechanisms, making user privacy easily compromised. At the same time, inadequate data backup mechanisms make data recovery difficult in the event of system failures, increasing the risk of data loss and corruption.

[0008] Therefore, those skilled in the art provide a method for integrating real estate data to solve the problems mentioned in the background art. Summary of the Invention

[0009] To address the shortcomings of existing technologies, this invention provides a method for integrating real estate data, which solves the problems of single data source, low processing efficiency, poor data quality, lack of data generation mechanism, and privacy and security issues in existing technologies.

[0010] To achieve the above objectives, the present invention provides the following technical solution: a method for integrating real estate data, comprising the following steps:

[0011] S1. Data Collection: Collect real estate data from multiple channels;

[0012] S2. Data Processing: Use AI technology to process all collected data, standardize its data format, and finally verify the data through blockchain.

[0013] S3. Data Association: Construct a knowledge graph to associate real estate information with transportation, education, and environmental data;

[0014] S4. Establish a generative adversarial network, train the generator and discriminator, and generate high-quality synthetic real estate data to supplement the scarce data.

[0015] S5. Database Establishment: The calculated integrated real estate data will be integrated and stored through a cloud computing platform and a distributed database.

[0016] S6. Real-time data update: Deploy a real-time data acquisition system to obtain the latest data through API interfaces, web crawlers and IoT devices, establish an automated data update mechanism, and embed a scheduling system to periodically trigger the data acquisition, cleaning, integration and storage process;

[0017] S7. Data Maintenance: Differential privacy technology is used to protect the privacy of the database, and a database backup mechanism is set up.

[0018] Preferably, the multiple channels in step S1 include one or more of government databases, real estate platforms, property management systems, and Internet of Things devices; the real estate data includes one or more of the following: geographic location, housing type, area, price, and transaction record information.

[0019] Preferably, the data processing in step S2 includes missing value processing, data deduplication, and outlier detection, wherein missing value processing is performed using a decision tree algorithm, data deduplication is performed using a Levenshtein distance algorithm, and outlier detection is performed using a Z-score algorithm.

[0020] The calculation formula for the decision tree algorithm is:

[0021]

[0022] Where D is the dataset, A is the partitioning attribute, Values(A) is all possible values ​​of attribute A, and D v Given A = v, the dataset is a subset of data. Entropy(D) is the entropy of dataset D.

[0023] The Levenshtein distance algorithm is calculated as follows:

[0024]

[0025] Where d(i,j) is the Levenshtein distance between the first i characters of string s1 and the first j characters of string s2, and [s1[i]≠s2[j]] means that d(i,j) is 1 when characters s1[i] and s2[j] are not equal, and d(i,j) is 0 when they are equal.

[0026] Preferably, the data verification step in step S2 includes:

[0027] S21. Data matching: Matching data based on the unique identifier of the real estate and the data content, and then merging data based on predefined weight merging rules;

[0028] S22. Data Verification: The merged data is uploaded to the blockchain, and the authenticity of the data is verified through the immutability and distributed ledger technology of the blockchain.

[0029] Preferably, the S3 knowledge graph construction step includes:

[0030] S31. Architecture Design: Define the structure of the knowledge graph, including the types of entities and relationships, and identify the core entities and relationships related to real estate;

[0031] S32. Relationship Building: By matching geographical locations, the merged real estate location data is associated with nearby transportation facilities, schools, and environmental data.

[0032] Preferably, the core entities in step S31 are real estate, owners, and transaction records, and the relationships are property rights, transactions, and leases.

[0033] Preferably, the goal of the generative adversarial network algorithm in step S4 is to maximize the classification accuracy of the discriminator and minimize the identifiability of the data generated by the generator, so that the generator can generate realistic data. The calculation steps are as follows:

[0034] The loss function of the discriminator is:

[0035]

[0036] in, To maximize the parameters of the discriminator, V(D) is the value function of the discriminator. The expected value of the actual data x. Indicates the expected value. To generate the expected value of the data G(z).

[0037] The loss function of the generator is:

[0038] in, To minimize the generator's parameters θ g V(D) is the value function of the generator. To generate the expected value of the data G(z).

[0039] 9. Preferably, the automated data update mechanism step in step S6 is as follows:

[0040] S61. The scheduler triggers the tasks in steps S1-S5 according to the preset time interval.

[0041] S62. Monitor the data update process in real time and detect abnormal situations.

[0042] A real estate data integration system and a method for real estate data integration include a data acquisition module, a data preprocessing module, a data fusion and matching module, a synthetic data generation module, a data storage module, a data update implementation module, and a data maintenance module.

[0043] Preferably, the data acquisition module and the data preprocessing module are connected via a network, the data preprocessing module and the data fusion and matching module are connected via a network, the data fusion and matching module and the synthetic data generation module are connected via a network, the data acquisition module and the data storage module are electrically connected, the real-time data update module is electrically connected to the data storage module and the data acquisition module, and the data maintenance module and the implementation data update module are connected via a network.

[0044] This invention provides a method for integrating real estate data. It has the following beneficial effects:

[0045] 1. This invention utilizes Generative Adversarial Networks (GANs) in a synthetic data generation module to generate high-quality synthetic real estate data, supplementing scarce data and enhancing the richness and diversity of the dataset. Compared to traditional methods, GANs, through adversarial training of the generator and discriminator, can generate highly realistic data, effectively solving the problems of data insufficiency and privacy. The generated synthetic data can not only be used to train machine learning models, improving their generalization ability and prediction accuracy, but also provide support in simulation experiments and market analysis.

[0046] 2. This invention achieves efficient acquisition, intelligent preprocessing, and real-time updating of multi-source data through the combination of a data acquisition module, a data preprocessing module, and a real-time data update module. Utilizing API interfaces, web crawlers, and IoT devices, the system can dynamically acquire the latest real estate data and periodically trigger data acquisition, cleaning, integration, and storage processes through an automated scheduling system, ensuring the real-time nature and accuracy of the data.

[0047] 3. The data preprocessing module and data fusion matching module of this invention use AI technology to handle missing values, deduplicate data, and detect outliers, unifying data formats and improving data quality. Through intelligent matching and merging, the system effectively solves the problems of conflict and redundancy in multi-source data, ensuring data consistency and integrity. Furthermore, data verification through blockchain technology further guarantees the authenticity and immutability of the data.

[0048] 4. The data maintenance module of this invention employs differential privacy technology to protect the database, ensuring that user privacy is not leaked during data analysis and querying. A robust database backup mechanism performs regular full and incremental backups to ensure timely recovery of data in case of loss, damage, or system failure, guaranteeing data integrity and high system availability. These measures collectively enhance the system's security and reliability, providing users with a highly trustworthy data management platform. Attached Figure Description

[0049] Figure 1 This is a schematic diagram of the method flow of the present invention;

[0050] Figure 2 This is a schematic diagram of the system framework structure of the present invention. Detailed Implementation

[0051] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0052] Example 1:

[0053] Please see the appendix Figure 1 This invention provides a method for integrating real estate data, comprising the following steps:

[0054] S1. Data Collection: Collect real estate data from multiple channels; multiple channels include one or more of the following: government databases, real estate platforms, property management systems, and IoT devices; real estate data includes one or more of the following: geographic location, housing type, area, price, and transaction records.

[0055] Specifically, using data from government databases can ensure the authority and accuracy of the data, helping to establish a highly trustworthy basic dataset. Real estate platforms update data frequently, providing the latest market dynamics and facilitating real-time monitoring of market changes. Data from property management systems can provide detailed information on property use and maintenance, helping to comprehensively understand the actual use and status of the property. Real-time environmental data provided by IoT devices helps to dynamically monitor the environmental conditions of the property, providing a scientific basis for property management and maintenance.

[0056] S2. Data Processing: AI technology is used to process all collected data, standardize its format, and finally verify the data via blockchain. The data verification steps include:

[0057] Data processing includes missing value handling, data deduplication, and outlier detection. Missing value handling uses a decision tree algorithm, data deduplication uses a Levenshtein distance algorithm, and outlier detection uses a Z-score algorithm.

[0058] The calculation formula for the decision tree algorithm is:

[0059]

[0060] Where D is the dataset, A is the partitioning attribute, Values(A) is all possible values ​​of attribute A, and D v Given A = v, the dataset is a subset of data. Entropy(D) is the entropy of dataset D.

[0061] The Levenshtein distance algorithm is calculated as follows:

[0062]

[0063] Where d(i,j) is the Levenshtein distance between the first i characters of string s1 and the first j characters of string s2, and [s1[i[≠s2[j]] means that d(i,j) is 1 when characters s1[i[ and s2[j] are not equal, and d(i,j) is 0 when they are equal.

[0064] S21. Data matching: Matching data based on the unique identifier of the real estate and the data content, and then merging data based on predefined weight merging rules;

[0065] S22. Data Verification: The merged data is uploaded to the blockchain, and the authenticity of the data is verified through the immutability and distributed ledger technology of the blockchain.

[0066] Specifically, the work steps are as follows:

[0067] Data processing:

[0068] Missing value handling: Use decision tree algorithms to calculate and fill in missing values ​​to ensure data integrity.

[0069] Data deduplication: The Levenshtein distance algorithm is used to calculate string similarity and remove duplicate data. Outlier detection: The Z-score algorithm is used to calculate the degree of anomaly for each data point, and outliers are detected and processed.

[0070] Data matching: Matching data based on the unique identifier of the real estate property with the data content ensures data uniqueness. Data merging is performed based on predefined weighted merging rules to handle data conflicts and redundancy.

[0071] Data verification: The merged data is uploaded to the blockchain, and its authenticity and immutability are verified using blockchain technology. Smart contracts are used to automatically execute data verification rules, ensuring data consistency and reliability.

[0072] S3. Data Association: Construct a knowledge graph to associate real estate information with transportation, education, and environmental data; the knowledge graph construction steps include:

[0073] S31. Architecture Design: Define the structure of the knowledge graph, including the types of entities and relationships, and identify the core entities and relationships related to real estate; the core entities are real estate, owners, and transaction records, and the relationships are property rights, transactions, and leases.

[0074] S32. Relationship Building: By matching geographical locations, the merged real estate location data is associated with nearby transportation facilities, schools, and environmental data.

[0075] S4. Establish a generative adversarial network (GAN), train the generator and discriminator to generate high-quality synthetic real estate data, supplementing the scarce data. The goal of the GAN algorithm is to maximize the classification accuracy of the discriminator and minimize the identifiability of the data generated by the generator, enabling the generator to produce realistic data. The calculation steps are as follows:

[0076] The loss function of the discriminator is:

[0077]

[0078] in, To maximize the parameters of the discriminator, V(D) is the value function of the discriminator. The expected value of the actual data x. Indicates the expected value. To generate the expected value of the data G(z).

[0079] The loss function of the generator is:

[0080] in, To minimize the generator's parameters θ g V(D) is the value function of the generator. To generate the expected value of the data G(z).

[0081] Specifically, the training steps for the Generative Adversarial Network (GAN) algorithm are as follows:

[0082] Initialize the parameters of the generator and discriminator.

[0083] Repeat the following steps until the model converges: Sample a batch of real data from the real data distribution. Sample a batch of random noise from the noise distribution and generate data using the generator. Update the discriminator's parameters to maximize the discriminator's value function. Sample a batch of random noise from the noise distribution and generate data using the generator. Update the generator's parameters to minimize the generator's value function.

[0084] Results: By generating high-quality synthetic data, scarce portions of real estate data can be effectively supplemented, increasing the richness and diversity of the dataset. The generated synthetic data can be used to train machine learning models, improving their generalization ability and prediction accuracy. When using real data raises privacy concerns, synthetic data can be used as an alternative, protecting user privacy. Synthetic data can be used for various simulation experiments and analyses, helping to understand market trends and behaviors and supporting decision-making.

[0085] S5. Database Establishment: The calculated integrated real estate data will be integrated and stored through a cloud computing platform and a distributed database.

[0086] Data partitioning: Real estate data is partitioned according to a specific partitioning strategy (such as by geographical location, property ownership number, etc.). Data in each partition is stored on different database nodes to achieve load balancing and improve query efficiency. Data partitioning can improve data access speed and system scalability, avoid single points of failure, and improve overall system performance.

[0087] Data replication: To improve data reliability and availability, distributed databases typically perform data replication, storing multiple copies of the data on different nodes. A data replication strategy is set to determine the number of copies of each data entry. When data is written, the copies are synchronized to the designated replication nodes. The integrity and consistency of the data copies are regularly checked and maintained. Data replication prevents data loss, improves system fault tolerance and availability, and ensures the system can continue to operate normally even if some nodes fail.

[0088] Data consistency: Ensuring the consistency of all replicas of data is crucial for distributed databases. Different consistency models (such as strong consistency and eventual consistency) can be chosen to achieve this. Distributed transactions or consistency protocols (such as Paxos and Raft) are used to guarantee data consistency during write and read operations. By ensuring data consistency, data conflicts and inconsistencies can be avoided, improving data reliability and accuracy, and ensuring the correctness of data operations.

[0089] S6. Real-time data update: Deploy a real-time data acquisition system to obtain the latest data through API interfaces, web crawlers and IoT devices, establish an automated data update mechanism, and embed a scheduling system to periodically trigger the data acquisition, cleaning, integration and storage process;

[0090] Specifically, the latest data is collected in real time through API interfaces, web crawlers, and IoT devices. API interfaces are used to obtain structured data from external data sources, web crawlers are used to crawl semi-structured or unstructured web page data, and IoT devices are used to obtain real-time environmental data. Through an automated scheduling system, the data collection, cleaning, integration, and storage processes are triggered periodically to ensure continuous data updates.

[0091] The steps of the automated data update mechanism are as follows:

[0092] S61. The scheduler triggers tasks S1-S5 according to a preset time interval. Time interval setting: Set an appropriate time interval according to business needs, such as hourly, daily, weekly, etc. The scheduler periodically triggers data acquisition (S1), data processing (S2), data matching (S21), data verification (S22), synthetic data generation (S4), and data storage (S5) tasks.

[0093] Task chain execution: Ensure that each step is executed in sequence, and record the execution status after each step is completed.

[0094] S62. Monitor the data update process in real time and detect abnormal situations.

[0095] Deploy a monitoring system using monitoring tools to monitor the execution of data update tasks in real time. Set up anomaly detection rules to monitor key indicators such as task execution time, data integrity, and data consistency. Configure an alarm system (such as email, SMS, and push notifications) to promptly notify relevant personnel when anomalies are detected. Record log information during the data update process, analyze the logs regularly, and optimize the data update process.

[0096] S7. Data Maintenance: Differential privacy technology is used to protect the privacy of the database, and a database backup mechanism is set up.

[0097] Specifically, differential privacy technology is used for privacy protection, and the database backup mechanism is designed to ensure data recovery in the event of data loss, corruption, or system failure, guaranteeing data integrity and availability. Database backups can be divided into full backups, incremental backups, and differential backups.

[0098] Comparative experiment of examples and control examples

[0099] To more clearly demonstrate the superiority of the method of the present invention, a comparative experiment was designed to integrate real estate data using the method of the present invention and the traditional method, and to compare the effects in various aspects.

[0100] Experimental Design

[0101] Experimental objective: To compare the effectiveness of the method of this invention with traditional methods in terms of data integrity, processing efficiency, data consistency, real-time data updates, and data privacy protection.

[0102] Experimental dataset: Select the same batch of real estate data, including data from government databases, real estate platforms, property management systems, and IoT devices.

[0103] Experimental steps

[0104] Example 1: The method of the present invention is not summarized.

[0105] Comparison example: Traditional method

[0106] The steps are as follows: Data collection: Data is collected from a single data source (such as a government database or a real estate platform), where the data source is limited and not updated in a timely manner.

[0107] Data processing: Manual data processing lacks a standardized format, resulting in poor data quality, high error rates, and low efficiency.

[0108] Data matching: The lack of unified matching rules makes it easy for conflicts and redundancy to occur during the data merging process, resulting in poor data consistency.

[0109] Data correlation: There is a lack of multi-dimensional data correlation, and the information analysis is limited to a single data source, resulting in insufficient information dimensions.

[0110] Data generation: The lack of data generation and replenishment mechanisms makes it difficult to address the scarcity of data.

[0111] Database setup: Using traditional relational databases results in poor scalability, low data access efficiency, and a lack of high availability and consistency guarantees.

[0112] Data updates: Manually updating data results in a low update frequency, making it difficult to ensure the real-time nature and accuracy of the data.

[0113] Data maintenance: Lack of privacy protection mechanisms makes data prone to leakage; lack of a sound backup mechanism results in a high risk of data loss and corruption.

[0114] Comparative experimental results

[0115]

[0116]

[0117] Summarize

[0118] Comparative experiments demonstrate that the method of this invention outperforms traditional methods in terms of data integrity, processing efficiency, data consistency, real-time data updates, data privacy protection, data access speed, system high availability, multi-dimensional data analysis, supplementation of scarce data, and data maintenance and backup. This invention provides higher efficiency, accuracy, security, and reliability, offering a solid guarantee for the integration and application of real estate data.

[0119] Example 2:

[0120] Please see the appendix Figure 2 This invention provides a real estate data integration system, including a data acquisition module, a data preprocessing module, a data fusion and matching module, a synthetic data generation module, a data storage module, a data update module, and a data maintenance module.

[0121] The data acquisition module and the data preprocessing module are connected by a network; the data preprocessing module and the data fusion and matching module are connected by a network; the data fusion and matching module and the synthetic data generation module are connected by a network; the data acquisition module and the data storage module are electrically connected; the real-time data update module is electrically connected to the data storage module and the data acquisition module; and the data maintenance module and the implementation data update module are connected by a network.

[0122] Specifically, the data acquisition module acquires multi-source data in real time from government databases, real estate platforms, property management systems, and IoT devices via network connection, ensuring data comprehensiveness and timeliness. The data preprocessing module uses AI technology to handle missing values, deduplicate data, and detect outliers, unifying data formats and improving data quality. The data fusion and matching module achieves intelligent matching and merging of multi-source data via network connection, ensuring data consistency and integrity. The synthetic data generation module utilizes Generative Adversarial Networks (GANs) to generate high-quality synthetic data, supplementing scarce data and enhancing the richness and diversity of the dataset. The data storage module ensures efficient data storage and management via electrical connection, leveraging cloud computing platforms and distributed databases to provide high availability and scalability. The real-time data update module maintains synchronization with the data storage and acquisition modules via electrical connection, and an automated scheduling system ensures continuous and real-time data updates. The data maintenance module implements differential privacy protection and backup mechanisms for data via network connection, ensuring data security, integrity, and availability. The overall system design is reasonable, with complete functions, achieving efficient, accurate, secure, and intelligent integration and management of real estate data.

[0123] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for real property data integration, comprising: Comprise the following steps: S1, data collection: collect real estate data from multiple channels; S2, data processing: use AI technology to process all collected data and unify its data format, and finally verify the data through blockchain; S3, data association: build a knowledge graph to associate real estate information with transportation, education, and environmental data; S4, establish a generative adversarial network to train the generator and discriminator to generate high-quality synthetic real estate data to supplement data scarcity; S5, database establishment: integrate and store the calculated integrated real estate data through cloud computing platforms and distributed databases; S6, real-time data updating: deploy a real-time data collection system to obtain the latest data through API interfaces, web crawlers, and Internet of Things devices, establish an automated data updating mechanism, and embed a scheduling system to trigger data collection, cleaning, integration, and storage processes regularly; S7, data maintenance: use differential privacy technology to protect the privacy of the database and set up a database backup mechanism.

2. The method of claim 1, wherein, The multiple channels in the S1 step include one or more of government databases, real estate platforms, property management systems, and Internet of Things devices; the real estate data includes one or more of geographic location, house type, area, price, and transaction record information.

3. The method of claim 1, wherein, The data processing in the S2 step includes missing value processing, data deduplication, and outlier detection, where the missing value processing uses a decision tree algorithm, the data deduplication uses a Levenshtein distance algorithm, and the outlier detection uses a Z-score algorithm. The decision tree algorithm calculation formula is: where D is the dataset, A is the split attribute, Values(A) is the set of all possible values of attribute A, D v is the subset of data under the condition A = v, and Entropy(D) is the entropy of dataset D; The Levenshtein distance algorithm calculation formula is: Where d(i,j) is the Levenshtein distance between the first i characters of string s1 and the first j characters of string s2, and [s1[i]≠s2[j]] is 1 when the characters s1[i] and s2[j] are not equal and 0 when they are equal.

4. The method of claim 1, wherein, The data verification step in the S2 step includes: S21, data matching: match the unique identifier of the real estate with the data content, and then merge the data based on predefined weight merging rules; S22, data verification: chain the merged data, and verify its authenticity through the immutability of blockchain and distributed ledger technology.

5. The method of claim 1, wherein, The S3 knowledge graph construction step includes: S31, architecture design: define the structure of the knowledge graph, including the types of entities and relationships, and determine the core entities and relationships related to real estate; S32, relationship construction: through geographic location matching, associate the integrated real estate positioning data with nearby transportation facilities, schools, and environmental data.

6. The method of claim 5, wherein, The core entities in the S31 step are property, owner, and transaction record, and the relationships are property rights, transactions, and rentals.

7. The method of claim 1, wherein, The goal of the generative adversarial network algorithm in the S4 step is to maximize the classification accuracy of the discriminator and minimize the recognizability of the data generated by the generator, so that the generator can generate realistic data. The calculation steps are: The loss function of the discriminator is: wherein, V(D) is a value function of the discriminator D for maximizing parameters of the discriminator D, E(x) is an expected value of the real data x, E(x) is an expected value of the real data x, E(G(z)) is an expected value of the generated data G(z). The loss function of the generator is: wherein is the parameter of the generator g , V(D) is the value function of the generator, is the expected value of the generated data G(z).

8. The method of claim 1, wherein, The automated data updating mechanism step in the S6 step is: S61, the scheduler triggers S1-S5 steps task according to the preset time interval; S62, real-time monitoring data update process, detect abnormal conditions.

9. A real estate data integration system for use in a method of real estate data integration according to any one of claims 1-8, characterized by It comprises a data acquisition module, a data preprocessing module, a data fusion matching module, a synthetic data generation module, a data storage module, an implementation data update module and a data maintenance module.

10. The real property data integration system of claim 9, wherein, The data acquisition module and the data preprocessing module are network connected, the data preprocessing module and the data fusion matching module are network connected, the data fusion matching module and the synthetic data generation module are network connected, the data acquisition module and the data storage module are electrically connected, the real-time data update module and the data storage module and the data acquisition module are electrically connected, and the data maintenance module and the implementation data update module are network connected.

Images