Land transaction dynamic monitoring and analysis system based on multi-source data fusion

Abstract

The present application relates to the field of geographic information technology, in particular to a land transaction dynamic monitoring and analysis system based on multi-source data fusion, comprising a data integration and storage unit, a knowledge graph construction unit, a real-time verification unit, a trusted transaction verification unit, a dynamic monitoring and decision support unit. The present application integrates heterogeneous data through a multi-modal data lake architecture and unifies the time and space reference, uses knowledge graph technology to construct a transaction element correlation network to generate dynamic verification rules, combines with the Flink stream processing engine to realize real-time compliance filtering of the transaction process, uses the improved isolated forest algorithm and graph convolution network to complete the transaction price rationality verification and subject correlation analysis and stores the evidence based on the blockchain technology, so as to realize the intelligent management of the whole process of land transaction data, the accurate identification and prevention and control of risks, and the efficient and data-based support of regulatory decisions.

Description

Technical Field

[0001] This invention relates to the field of geographic information technology, specifically to a dynamic monitoring and analysis system for land transactions based on multi-source data fusion. Background Technology

[0002] Existing technologies suffer from problems such as fragmented and isolated data, delayed risk identification, and insufficient decision support in traditional land transaction supervision.

[0003] Based on this, the present invention provides a dynamic monitoring and analysis system for land transactions based on multi-source data fusion to solve the aforementioned technical problems. Summary of the Invention

[0004] The purpose of this invention is to provide a dynamic monitoring and analysis system for land transactions based on multi-source data fusion. This invention integrates heterogeneous data using a multimodal data lake architecture and unifies spatiotemporal benchmarks. It utilizes knowledge graph technology to construct a network of associations for transaction elements to generate dynamic verification rules. Combined with the Flink stream processing engine, it achieves real-time compliance filtering of the transaction process. By leveraging an improved isolated forest algorithm and graph convolutional networks, it completes the verification of the reasonableness of transaction prices and the analysis of the correlation between entities, and stores evidence based on blockchain technology. Finally, it quantifies market fluctuations and outputs risk signals through a heat index model and an idle early warning mechanism, thereby achieving intelligent governance of the entire process of land transaction data, accurate identification and prevention of risks, and efficient and data-driven support for regulatory decisions.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] This invention provides a dynamic monitoring and analysis system for land transactions based on multi-source data fusion, comprising a data integration and storage unit, a knowledge graph construction unit, a real-time verification unit, a trusted transaction verification unit, and a dynamic monitoring and decision support unit, wherein:

[0007] The data integration and storage unit is used to integrate natural resource data, transaction data, public opinion data, and macroeconomic indicators through a multimodal data lake architecture, and to perform unified spatiotemporal benchmark mapping of heterogeneous data using spatiotemporal coding technology.

[0008] The knowledge graph construction unit is used to construct a three-dimensional relational network that includes the equity penetration relationship of the transaction entities, land development constraints, and policy effectiveness time windows using knowledge graph technology, and dynamically generates real-time verification rules through semantic parsing of policy clauses.

[0009] The real-time verification unit: Based on the Flink stream processing engine, it establishes a transaction process rule base, executes the real-time verification rules generated by the knowledge graph construction unit, and performs compliance filtering on the transaction process;

[0010] The trusted transaction verification unit is used to verify the reasonableness of transaction prices by adopting an improved isolated forest algorithm, and to verify the correlation of transaction entities by combining graph convolutional networks. It performs secondary verification on transactions that pass the real-time verification unit and generates trusted evidence for compliant transactions based on blockchain technology.

[0011] The dynamic monitoring and decision support unit is used to quantify market fluctuations using the heat index model and output early warning signals for idle land.

[0012] The data integration and storage unit includes a multimodal data access module, a spatiotemporal coding mapping module, and a data lake management module, wherein:

[0013] The multimodal data access module is used to receive data from different sources and formats, including natural resources, transactions, public opinion, and economic indicators.

[0014] The spatiotemporal coding mapping module is used to map heterogeneous data onto a unified spatiotemporal reference using spatiotemporal coding technology.

[0015] The data lake management module is used to build and manage a multimodal data lake.

[0016] The spatiotemporal coding technology uses the GeoHash algorithm to encode the spatial location of the land parcel, and combines the ISO 8601 standard to format the transaction timestamp, unifying the spatiotemporal reference of the data, with an encoding precision of 6 GeoHash characters.

[0017] The knowledge graph construction unit includes an entity relationship extraction module, a policy parsing module, and a graph storage module, wherein:

[0018] The entity relationship extraction module is used to extract enterprise equity hierarchy relationships from business registration data using NLP technology.

[0019] The policy parsing module identifies constraint clauses in policy texts based on the BERT model and converts them into executable rules.

[0020] The graph storage module is used to store the "enterprise-land parcel-policy" association network using the Neo4j graph database, and supports multi-hop queries.

[0021] The policy parsing module identifies constraint clauses in policy texts based on the BERT model and converts them into executable rules. The specific operation is as follows:

[0022] A1: Text Vectorization: Converting policy text into a token sequence X = [t1, t2, ..., t] n The WordPiece word segmentation algorithm generates sub-word units, whose word vectors are represented as follows:

[0023] ei =E·t i +P i +S i

[0024] In the formula, E is the word embedding matrix, P i For position encoding, S i Segment and embed the sentence;

[0025] A2: BERT semantic encoding: Generates a context vector H = {h1,…,h} through a multi-layer Transformer encoder. n}, where the attention of the l-th layer is calculated as follows:

[0026]

[0027] In the formula, Q, K, and V are the query, key, and value matrices, respectively, and d k The dimension of the key vector;

[0028] A3: Rule Transformation: The identified constraint "If condition C then action A" is transformed into a Flink SQL rule, where the extraction probability of condition C satisfies:

[0029] p(C)=σ(W c ·h0+b c )

[0030] In the formula, σ is the Sigmoid function, and W c and b c These are the parameters for the classification layer.

[0031] The real-time verification unit includes a rule base management module, a real-time stream processing module, and a compliance filtering module, wherein:

[0032] The rule base management module is used to establish and maintain a transaction process rule base and store various verification rules.

[0033] The real-time stream processing module: based on the Flink stream processing engine, it processes transaction process data in real time;

[0034] The compliance filtering module is used to execute the verification rules generated by the knowledge graph, perform compliance filtering on the transaction process, generate log records for intercepted abnormal transactions, including transaction ID, violation type, triggering rule, and push warning information to the regulatory platform.

[0035] The trusted transaction verification unit includes a price reasonableness verification module, a correlation analysis module, a secondary verification module, and a blockchain evidence storage module, wherein:

[0036] The price reasonableness verification module is used to identify abnormal transactions that deviate from the normal price range by ≥2σ using a weighted isolated forest algorithm.

[0037] The correlation analysis module is used to calculate the similarity of bid prices between companies using GCN, and a similarity > 0.9 is marked as suspected bid rigging.

[0038] The secondary verification module is used to perform secondary verification on transactions that have passed real-time verification.

[0039] The blockchain evidence storage module is used to write the hash of the bid documents and key data of the margin certificate for compliant transactions into the Hyperledger Fabric chain node.

[0040] The correlation analysis module calculates the similarity of bid prices between companies using GCN. A similarity greater than 0.9 is flagged as suspected bid rigging. The specific steps are as follows:

[0041] B1. Constructing an enterprise relationship graph: Using the bidding enterprises as nodes and equity relationships and the degree of overlap of legal representatives as edges, construct an initial graph structure;

[0042] B2. Graph Convolution Feature Extraction: Node features are iteratively updated using the GCN model. The feature update formula for the k-th layer is:

[0043]

[0044] in, It is an adjacency matrix with self-connections. W is the degree matrix. ( k) are trainable weights;

[0045] B3. Similarity Matrix Calculation: For the final layer node features H ( K) Calculate the cosine similarity matrix S, where:

[0046]

[0047] B4. Determination of Bid-rigging: When S ij When the value is greater than 0.9, company i and company j are marked as suspected entities involved in bid rigging.

[0048] The blockchain evidence storage module writes the hash of the bid document and key data of the margin certificate for compliant transactions into the Hyperledger Fabric on-chain node. The specific operation is as follows:

[0049] C1: Data Preprocessing for Evidence Preservation: Extract key fields such as the SHA-256 hash value of the tender document, the margin certificate number, and the timestamp from the verified transaction data, and generate a JSON format evidence preservation request;

[0050] C2: Smart Contract Trigger: Call the Hyperledger Fabric chaincode to verify whether the evidence storage request meets the following conditions:

[0051] a) The transaction has passed the secondary verification by the trusted transaction verification unit;

[0052] b) The compliance status output by the real-time verification unit is "passed";

[0053] C3: Multi-node consensus: Based on the preset endorsement strategy, the evidence data is packaged to generate a new block;

[0054] C4: On-chain storage: Writes blocks to the LevelDB database of each node and returns a storage certificate containing the transaction ID, block height, and Merkle root hash.

[0055] The dynamic monitoring and decision support unit includes a market popularity calculation module, an idle resource early warning module, and a visualization module, wherein:

[0056] The market heat calculation module generates a regional heat index based on bidding participation rate × average premium rate, where bidding participation rate = number of actual bidding companies / number of potential bidding companies × 100%, and average premium rate = ∑(land transaction price - starting price) / ∑ starting price.

[0057] The idle early warning module is used to compare the development time nodes stipulated in the land transfer contract. If construction has not started within 6 months, a red warning is triggered, and if it has not started within 3 months, a yellow warning is triggered.

[0058] The visualization module is used to display the regional heat distribution and the location of idle land plots through a GIS map, and supports drill-down to view related enterprise information.

[0059] Compared with the prior art, the beneficial effects of the present invention are:

[0060] This invention integrates heterogeneous data through a multimodal data lake architecture and unifies spatiotemporal benchmarks. It utilizes knowledge graph technology to construct a network of associations for transaction elements to generate dynamic verification rules. Combined with the Flink stream processing engine, it achieves real-time compliance filtering of the transaction process. By leveraging an improved isolated forest algorithm and graph convolutional networks, it completes the verification of the reasonableness of transaction prices and the analysis of the correlation between entities, and stores the evidence based on blockchain technology. Finally, it quantifies market fluctuations and outputs risk signals through a heat index model and an idle early warning mechanism, thereby realizing intelligent governance of the entire process of land transaction data, accurate identification and prevention of risks, and efficient and data-driven support for regulatory decisions. Attached Figure Description

[0061] Figure 1 This is a system diagram of the dynamic monitoring and analysis system for land transactions based on multi-source data fusion, as described in this invention.

[0062] Figure 2 This is a flowchart illustrating the knowledge graph construction process in the dynamic monitoring and analysis system for land transactions based on multi-source data fusion, as described in this invention.

[0063] Figure 3 This is a flowchart of the real-time verification process in the dynamic monitoring and analysis system for land transactions based on multi-source data fusion, as described in this invention.

[0064] Explanation of icon numbers:

[0065] 100. Data Integration and Storage Unit; 101. Multimodal Data Access Module; 102. Spatiotemporal Coding Mapping Module; 103. Data Lake Management Module; 200. Knowledge Graph Construction Unit; 201. Entity Relationship Extraction Module; 202. Policy Analysis Module; 203. Knowledge Graph Storage Module; 300. Real-time Verification Unit; 301. Rule Base Management Module; 302. Real-time Stream Processing Module; 303. Compliance Filtering Module; 400. Trusted Transaction Verification Unit; 401. Price Reasonableness Verification Module; 402. Association Analysis Module; 403. Secondary Verification Module; 404. Blockchain Evidence Storage Module; 500. Dynamic Monitoring and Decision Support Unit; 501. Market Heat Calculation Module; 502. Idle Equipment Early Warning Module; 503. Visualization Module. Detailed Implementation

[0066] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. 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 of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0067] Example:

[0068] like Figures 1-3As shown, this embodiment provides a dynamic monitoring and analysis system for land transactions based on multi-source data fusion, including a data integration and storage unit 100, a knowledge graph construction unit 200, a real-time verification unit 300, a trusted transaction verification unit 400, and a dynamic monitoring and decision support unit 500. Specifically: the data integration and storage unit 100 integrates natural resource data, transaction data, public opinion data, and macroeconomic indicators through a multimodal data lake architecture, and uses spatiotemporal coding technology to perform unified spatiotemporal benchmark mapping of heterogeneous data; the knowledge graph construction unit 200 uses knowledge graph technology to construct a three-dimensional data graph containing the equity penetration relationship of transaction entities, land development constraints, and policy effectiveness time windows. The system consists of a network of interconnected entities and dynamically generates real-time verification rules through semantic parsing of policy clauses. The real-time verification unit 300, based on the Flink stream processing engine, establishes a transaction process rule base and executes the real-time verification rules generated by the knowledge graph construction unit 200 to filter transaction processes for compliance. The trusted transaction verification unit 400 verifies the reasonableness of transaction prices using an improved isolated forest algorithm, combines graph convolutional networks to verify the correlation between transaction entities, performs secondary verification on transactions that pass the real-time verification unit 300, and generates trusted evidence for compliant transactions based on blockchain technology. The dynamic monitoring and decision support unit 500 quantifies market fluctuations using a heat index model and outputs early warning signals for idle land.

[0069] It should be noted that the data integration and storage unit 100 realizes unified governance of multi-source heterogeneous data, the knowledge graph construction unit 200 establishes a deep correlation network of transaction elements, the real-time verification unit 300 conducts real-time compliance review of the transaction process based on dynamic rules, the trusted transaction verification unit 400 uses intelligent algorithms to complete the dual verification of the authenticity of the transaction and blockchain notarization, and finally the dynamic monitoring and decision support unit 500 realizes the quantitative assessment of market status and risk warning.

[0070] In this embodiment, it should also be noted that the data integration and storage unit 100 includes a multimodal data access module 101, a spatiotemporal encoding mapping module 102, and a data lake management module 103. Specifically: the multimodal data access module 101 receives data from different sources and formats, including natural resources, transactions, public opinion, and economic indicators; the spatiotemporal encoding mapping module 102 uses spatiotemporal encoding technology to map heterogeneous data onto a unified spatiotemporal reference; the spatiotemporal encoding technology uses the GeoHash algorithm to encode the spatial location of land parcels and combines it with the ISO 8601 standard to format transaction timestamps, unifying the spatiotemporal reference of the data, achieving an encoding precision of 6 GeoHash characters; and the data lake management module 103 is used to construct and manage the multimodal data lake.

[0071] It should be noted that the multimodal data access module 101 realizes the unified collection of multi-source heterogeneous data, and the spatiotemporal coding mapping module 102 completes the unified mapping of spatiotemporal reference using the GeoHash algorithm (6-bit precision) and the ISO 8601 standard. Finally, the data lake management module 103 constructs a standardized and traceable multimodal data lake.

[0072] Furthermore, it should be noted that the multimodal data access module 101 supports integration with land parcel map data (e.g., CAD format) from natural resources departments, listed transaction data (JSON / XML format) from trading platforms, web crawler data (text / HTML format), and macroeconomic indicators from the National Bureau of Statistics (API interface). The data lake management module 103 constructs a multimodal data lake based on Hadoop+Hive+HBase: ① Structured data (transaction records, economic indicators) is stored in the Hive data warehouse, partitioned by "time + region," supporting efficient SQL queries; ② Semi-structured data (policy texts, public opinion information) is stored in HBase, using "spatiotemporal encoding + data type" as the rowkey for fast retrieval; ③ Unstructured data (land parcel maps, remote sensing images) is stored in HDFS, with associated metadata (e.g., hash values, spatial ranges) linked to Hive tables. It also provides data lineage tracking, quality monitoring (verifying spatiotemporal encoding integrity and field null value rate), and lifecycle management (automatically archiving historical data older than 3 years) functions to ensure data quality and availability.

[0073] In this embodiment, it should also be noted that the knowledge graph construction unit 200 includes an entity relationship extraction module 201, a policy parsing module 202, and a graph storage module 203, wherein: the entity relationship extraction module 201 is used to extract the enterprise equity hierarchy relationship from business data using NLP technology; the policy parsing module 202 is used to identify the constraint clauses in the policy text based on the BERT model and convert them into executable rules; the specific operation is as follows: A1: Text vectorization: convert the policy text into a token sequence X = [t1, t2, ..., t n The WordPiece word segmentation algorithm generates sub-word units, whose word vectors are represented as follows:

[0074] e i =E·t i +P i +S i

[0075] In the formula, E is the word embedding matrix, P i For position encoding, S i Sentence segmentation and embedding; A2: BERT semantic encoding: generating context vectors H = {h1,…,h} through a multi-layer Transformer encoder. n}, where the attention of the l-th layer is calculated as follows:

[0076]

[0077] In the formula, Q, K, and V are the query, key, and value matrices, respectively, and d k A3: Rule Transformation: The identified constraint "If condition C then action A" is transformed into a Flink SQL rule, where the extraction probability of condition C satisfies:

[0078] p(C)=σ(W c ·h0+b c )

[0079] In the formula, σ is the Sigmoid function, and W c and b c For classification layer parameters. Graph storage module 203: Used to store the "enterprise-land parcel-policy" association network using the Neo4j graph database, supporting multi-hop queries.

[0080] It should be noted that the entity relationship extraction module 201 uses NLP technology to construct a corporate equity relationship network. The policy parsing module 202 uses the BERT model (which includes a three-layer processing architecture of WordPiece segmentation, Transformer encoding, and rule probability transformation) to realize the intelligent parsing and rule transformation of policy clauses. Finally, the graph storage module 203 constructs a three-dimensional relationship network of "enterprise-land-policy" in the Neo4j graph database.

[0081] Furthermore, it should be noted that the entity relationship extraction module 201, based on business registration data (company annual reports, equity change records), uses the BERT-NER model to identify entities such as company names, legal representatives, and land parcel codes (accuracy ≥ 95%); it extracts relationships such as "shareholding ratio" and "change of legal representative" through dependency parsing (e.g., extracting the triple (XX Company, holding shares, YY Company, 30%) from "XX Company holds 30% equity in YY Company"); it uses depth-first search to traverse the equity hierarchy, generating a penetrating relationship chain of "ultimate beneficiary-subsidiary-grandchild company", supporting the parsing of more than 5 levels of nested equity structures and accurately identifying related transaction entities. The graph storage module 203 supports multi-hop queries through Cypher statements (e.g., querying "land transactions of subsidiaries of credit-A enterprises subject to 2024 policy constraints"), with a response time ≤ 500ms (million-node scale), providing efficient support for complex relationship analysis.

[0082] In this embodiment, it should also be noted that the real-time verification unit 300 includes a rule base management module 301, a real-time stream processing module 302, and a compliance filtering module 303. Specifically: the rule base management module 301 is used to establish and maintain a transaction process rule base, storing various verification rules; the real-time stream processing module 302, based on the Flink stream processing engine, processes transaction process data in real time; and the compliance filtering module 303 is used to execute verification rules generated from the knowledge graph, perform compliance filtering on the transaction process, generate log records for intercepted abnormal transactions, including the transaction ID, violation type, and triggering rule, and push warning information to the regulatory platform.

[0083] It should be noted that the rule base management module 301 dynamically maintains the verification rule base generated by the knowledge graph, relies on the Flink engine of the real-time stream processing module 302 to achieve millisecond-level processing of transaction data, and finally the compliance filtering module 303 performs multi-dimensional rule matching and real-time interception.

[0084] Furthermore, it should be noted that the rule base management module 301 stores static rules (policy constraints, transaction specifications) and dynamic rules (based on historical anomaly data, dynamically adjusted through reinforcement learning). It provides rule version control (recording iteration history), conflict detection (verifying rule compatibility), and canary release (new rules are first applied to 10% of the transaction flow) to ensure the stability and reliability of rule updates. The real-time stream processing module 302, based on the Flink stream processing framework, uses Kafka as a message queue to access transaction data and resolves cross-time zone and delayed data timing issues through event-time processing. Combined with window calculations (e.g., a 1-minute scrolling window to count the number of land bids) and state storage (recording the number of historical violations by enterprises), it provides real-time context for compliance verification, with a processing latency of ≤200ms. The compliance filtering module 303 calls the rule base SQL rules and the knowledge graph Cypher query to filter transaction data in real time. It generates logs for abnormal transactions (including transaction ID, violation type, and triggering rule), stores them in Elasticsearch, and pushes them to the regulatory platform via WebSocket, with an alert latency of ≤200ms, achieving "second-level detection and real-time handling" of violations.

[0085] In this embodiment, it should also be noted that the trusted transaction verification unit 400 includes a price reasonableness verification module 401, a correlation analysis module 402, a secondary verification module 403, and a blockchain evidence storage module 404. Specifically: the price reasonableness verification module 401 is used to identify abnormal transactions that deviate from the normal price range by ≥2σ using a weighted isolated forest algorithm; the correlation analysis module 402 is used to calculate the similarity of bid prices between enterprises using GCN, and when the similarity is >0.9, it is marked as suspected bid rigging; the specific operations are as follows: B1, Constructing an enterprise correlation graph: Using bidding enterprises as nodes and equity relationships and legal representative overlap as edges to construct an initial graph structure; B2, Graph convolution feature extraction: Iteratively updating node features through the GCN model, the feature update formula for the k-th layer is:

[0086]

[0087] in, It is an adjacency matrix with self-connections. W is the degree matrix. ( k) are trainable weights; B3, similarity matrix calculation: for the final layer node features H ( K) Calculate the cosine similarity matrix S, where:

[0088]

[0089] B4. Determination of Bid-rigging: When S ij When the value is greater than 0.9, companies i and j are marked as suspected bid-rigging entities. The secondary verification module 403 is used to perform secondary verification on transactions that have passed real-time verification; the blockchain evidence storage module 404 is used to write the key data of the bid document hash and margin certificate of compliant transactions into the Hyperledger Fabric on-chain nodes. The specific operations are as follows: C1: Evidence storage data preprocessing: Extract the key fields of the bid document SHA-256 hash value, margin certificate number, and timestamp from the verified transaction data, and generate a JSON format evidence storage request; C2: Smart contract triggering: Call the Hyperledger Fabric chaincode to verify whether the evidence storage request meets the following conditions: a) The transaction has passed the secondary verification of the trusted transaction verification unit 400; b) The compliance status output by the real-time verification unit 300 is "passed"; C3: Multi-node consensus: According to the preset endorsement strategy, package the evidence storage data to generate a new block; C4: On-chain storage: Write the block to the LevelDB database of each node and return the evidence storage certificate containing the transaction ID, block height, and Merkle root hash.

[0090] It should be noted that the price reasonableness verification module 401 uses the weighted isolated forest algorithm to achieve intelligent screening of transaction prices. Combined with the GCN graph convolutional network (including enterprise association graph construction, multi-layer feature extraction and cosine similarity calculation) of the association analysis module 402, it deeply mines clues of bid rigging. The transaction is then double-verified by the secondary verification module 403. Finally, the hash value, voucher and other key data of compliant transactions are stored on the blockchain after being triggered by smart contracts and consensus among multiple nodes by the blockchain evidence storage module 404.

[0091] Furthermore, it should be noted that the price reasonableness verification module 401 employs a weighted isolated forest algorithm, selecting eight features including "regional benchmark land price, plot ratio, supporting infrastructure maturity, and historical premium rate." Core features (such as benchmark land price) are assigned high weights (e.g., 0.3) to reduce interference from secondary features. The path length from the sample to the root node of the isolated tree is calculated; if the path length is below the 50th percentile, it is marked as an anomaly. The anomaly identification accuracy is ≥92%, accurately identifying transactions where prices deviate from the reasonable market range. The secondary verification module 403 performs a "price verification → correlation verification → 5% manual review" process on transactions that pass real-time verification. Price verification calls the price reasonableness verification module 401, correlation verification calls the correlation analysis module 402, and manual review allows viewing anomaly details (such as price deviation curves and enterprise correlation paths) through a visual interface. The false positive rate is ≤3%, ensuring verification accuracy.

[0092] In this embodiment, it should also be noted that the dynamic monitoring and decision support unit 500 includes a market popularity calculation module 501, an idle early warning module 502, and a visualization module 503. Specifically: the market popularity calculation module 501 generates a regional popularity index based on bidding participation rate × average premium rate, where bidding participation rate = actual number of participating companies / potential number of participating companies × 100%.

[0093] Average premium rate = ∑(land transaction price - starting price) / ∑starting price; Idle land warning module 502: used to compare the development time nodes agreed in the land transfer contract. If construction has not started within 6 months, a red warning is triggered, and if it has not started within 3 months, a yellow warning is triggered; Visualization module 503: used to display the regional heat distribution and the location of idle land plots through a GIS map, and supports drill-down to view related enterprise information.

[0094] It should be noted that the market heat calculation module 501 uses core indicators such as bidding participation and premium rate to quantitatively assess the regional market situation. Combined with the time-series monitoring function of the idle land early warning module 502 (3-month yellow warning / 6-month red warning), it accurately identifies the risk of land idling. Finally, the GIS platform of the visualization module 503 realizes the multi-dimensional data linkage display of "heat distribution - idle land plots - enterprise association".

[0095] Furthermore, it should be noted that the idle land early warning module 502 compares the development time stipulated in the land transfer contract and determines the idle status by calculating the time difference: a yellow warning is triggered if the development time is overdue by 3 months (a rectification plan is pushed out), and a red warning is triggered if the development time is overdue by 6 months (an idle land investigation is initiated). The module integrates the issuance time of the construction project planning permit and construction filing information as the "basis for commencement of construction" to avoid misjudgment and improve the accuracy of early warnings. The visualization module 503 displays the regional heat distribution (heat map) and the location of idle land plots (marked with red dots) through a GIS map, supporting filtering by administrative region and time. Clicking on a land plot / enterprise allows drilling down to related transaction records (transaction price, bidding companies), policy constraints, and blockchain evidence details; it provides dashboards such as market heat trend charts and idle land disposal progress tables, supports PDF export and email subscription, and assists regulatory departments in "data-driven and proactive" decision-making.

[0096] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0097] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A dynamic monitoring and analysis system for land transactions based on multi-source data fusion, characterized in that: It includes a data integration and storage unit (100), a knowledge graph construction unit (200), a real-time verification unit (300), a trusted transaction verification unit (400), and a dynamic monitoring and decision support unit (500), wherein: The data integration and storage unit (100) is used to integrate natural resource data, transaction data, public opinion data, and macroeconomic indicators through a multimodal data lake architecture, and to perform unified spatiotemporal benchmark mapping of heterogeneous data using spatiotemporal coding technology. The knowledge graph construction unit (200) is used to construct a three-dimensional association network containing the equity penetration relationship of the transaction entity, land development constraints, and policy effectiveness time window using knowledge graph technology, and to dynamically generate real-time verification rules through semantic parsing of policy clauses. The real-time verification unit (300) is based on the Flink stream processing engine, establishes a transaction process rule base, executes the real-time verification rules generated by the knowledge graph construction unit (200), and performs compliance filtering on the transaction process. The trusted transaction verification unit (400) is used to verify the reasonableness of the transaction price by adopting the improved isolated forest algorithm, and to verify the correlation of the transaction subjects by combining the graph convolutional network. It performs secondary verification on the transactions that pass the real-time verification unit (300) and generates trusted evidence for compliant transactions based on blockchain technology. The dynamic monitoring and decision support unit (500) is used to quantify market fluctuations using the heat index model and output an early warning signal for idle land. The trusted transaction verification unit (400) includes a price reasonableness verification module (401), a correlation analysis module (402), a secondary verification module (403), and a blockchain evidence storage module (404), wherein: The price reasonableness verification module (401) is used to identify price deviations from the normal price range using a weighted isolated forest algorithm. Abnormal transactions; The association analysis module (402) is used to calculate the similarity of bid prices between enterprises using GCN. The time stamp is marked as suspected bid rigging; The secondary verification module (403) is used to perform secondary verification on transactions that have passed real-time verification. The blockchain evidence storage module (404) is used to write the hash of the tender documents and key data of the margin certificate for compliant transactions into the Hyperledger Fabric chain node. In the association analysis module (402), the similarity of bid prices between enterprises is calculated using GCN. The time stamp is marked as suspected bid rigging. The specific steps are as follows: B1. Constructing an enterprise relationship graph: Using the bidding enterprises as nodes and equity relationships and the degree of overlap of legal representatives as edges, construct an initial graph structure; B2. Graph Convolution Feature Extraction: Node features are iteratively updated using the GCN model. The feature update formula for the k-th layer is: in, It is an adjacency matrix with self-connections. For degree matrix, These are trainable weights; B3. Similarity Matrix Calculation: For the features of the final layer nodes Calculate the cosine similarity matrix ,in: B4. Determination of Bidding Rigging: When At that time, company i and company j were identified as suspected entities involved in bid rigging.

2. The land transaction dynamic monitoring and analysis system based on multi-source data fusion according to claim 1, characterized in that, The data integration and storage unit (100) includes a multimodal data access module (101), a spatiotemporal coding mapping module (102), and a data lake management module (103), wherein: The multimodal data access module (101) is used to receive data from different sources and formats, including natural resources, transactions, public opinion, and economic indicators. The spatiotemporal coding mapping module (102) is used to map heterogeneous data onto a unified spatiotemporal reference using spatiotemporal coding technology; The data lake management module (103) is used to build and manage a multimodal data lake.

3. The land transaction dynamic monitoring and analysis system based on multi-source data fusion according to claim 2, characterized in that, The spatiotemporal coding technology uses the GeoHash algorithm to encode the spatial location of the land parcel, and combines the ISO 8601 standard to format the transaction timestamp, unifying the spatiotemporal reference of the data, with an encoding precision of 6 GeoHash characters.

4. The land transaction dynamic monitoring and analysis system based on multi-source data fusion according to claim 1, characterized in that, The knowledge graph construction unit (200) includes an entity relation extraction module (201), a policy parsing module (202), and a graph storage module (203), wherein: The entity relationship extraction module (201) is used to extract the corporate equity hierarchy relationship from business data using NLP technology; The policy parsing module (202) identifies the constraint clauses in the policy text based on the BERT model and converts them into executable rules. The graph storage module (203) is used to store the "enterprise-land parcel-policy" association network using the Neo4j graph database, and supports multi-hop queries.

5. The land transaction dynamic monitoring and analysis system based on multi-source data fusion according to claim 4, characterized in that, The policy parsing module (202) identifies the constraint clauses in the policy text based on the BERT model and converts them into executable rules. The specific operation is as follows: A1: Text Vectorization: Converting policy text into a sequence of tokens Sub-word units are generated using the WordPiece word segmentation algorithm, and their word vectors are represented as follows: In the formula, For word embedding matrix, For position encoding, Segmenting and embedding sentences; A2: BERT semantic encoding: generating context vectors through a multi-layer Transformer encoder. , of which The attention calculation for each layer is as follows: In the formula, , , For querying key-value matrices, The dimension of the key vector; A3: Rule Transformation: Transform the identified constraint clause "If condition C then action A" into... The rule, where the extraction probability of condition C satisfies: In the formula, function, and These are the parameters for the classification layer.

6. The land transaction dynamic monitoring and analysis system based on multi-source data fusion according to claim 1, characterized in that, The real-time verification unit (300) includes a rule base management module (301), a real-time stream processing module (302), and a compliance filtering module (303), wherein: The rule base management module (301) is used to establish and maintain a transaction process rule base and store verification rules. The real-time stream processing module (302) is based on the Flink stream processing engine and processes transaction process data in real time. The compliance filtering module (303) is used to execute the verification rules generated by the knowledge graph, perform compliance filtering on the transaction process, generate log records for intercepted abnormal transactions, including transaction ID, violation type, triggering rule, and push warning information to the regulatory platform.

7. The land transaction dynamic monitoring and analysis system based on multi-source data fusion according to claim 1, characterized in that, The blockchain evidence storage module (404) writes the hash of the bid document and key data of the margin certificate for compliant transactions into the Hyperledger Fabric on-chain node. The specific operation is as follows: C1: Data Preprocessing for Evidence Preservation: Extract key fields such as the SHA-256 hash value of the tender document, the margin certificate number, and the timestamp from the verified transaction data, and generate a JSON format evidence preservation request; C2: Smart Contract Trigger: Call the Hyperledger Fabric chaincode to verify whether the evidence storage request meets the following conditions: a) The transaction has passed the secondary verification by the trusted transaction verification unit (400); b) The compliance status output by the real-time verification unit (300) is "passed"; C3: Multi-node consensus: Based on the preset endorsement strategy, the evidence data is packaged to generate a new block; C4: On-chain storage: Writes blocks to the LevelDB database of each node and returns a storage certificate containing the transaction ID, block height, and Merkle root hash.

8. The land transaction dynamic monitoring and analysis system based on multi-source data fusion according to claim 1, characterized in that, The dynamic monitoring and decision support unit (500) includes a market popularity calculation module (501), an idle early warning module (502), and a visualization module (503), wherein: The market popularity calculation module (501) is based on: Generate a regional heat index, in which , ; The idle early warning module (502) is used to compare the development time nodes stipulated in the land transfer contract. If construction has not started within 6 months, a red warning is triggered; if construction has not started within 3 months, a yellow warning is triggered. The visualization module (503) is used to display the regional heat distribution and the location of idle land plots through a GIS map, and supports drill-down to view related enterprise information.

Images