Intelligent Adaptation System for Cross-border Supervision of OTC Derivatives Based on Heterogeneous Data Fusion

Abstract

The present invention discloses an intelligent adaptation system for cross-border supervision of over-the-counter derivatives based on heterogeneous data fusion, which relates to the technical field of big data processing. A data collection module is set to obtain over-the-counter derivatives trading data and generate an original data set. A data structuring and processing module is set to perform semantic parsing on the original data set using a text model for financial texts to generate a transaction information table. A transaction type classification module is set to judge whether a cross-border transaction is involved according to the transaction type in the transaction information table. If a cross-border transaction is involved, a cross-border compliance assessment module is set to perform dynamic rule matching on the transaction information table to generate a compliance report; otherwise, a risk prediction module is set to perform risk prediction on the transaction information table using a pre-trained transaction graph neural network model to generate a risk warning signal and execute corresponding warning measures. The compliance cost is reduced and the risk control ability is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of big data processing, and specifically to an intelligent adaptation system for cross-border supervision of over-the-counter derivatives based on heterogeneous data fusion. Background Art

[0002] As an important part of the global financial system, the over-the-counter (OTC) derivatives market has a huge trading scale and high complexity. Different from exchange-traded standardized products, OTC derivatives are highly customized with flexible contract terms, which poses great challenges to supervision. Especially in cross-border trading scenarios, due to the involvement of laws, regulations, regulatory requirements, and compliance standards in different countries and regions, the supervision complexity increases exponentially.

[0003] In OTC derivatives trading, cross-border transactions need to meet the regulatory requirements of multiple jurisdictions simultaneously, involving multiple compliance considerations such as foreign exchange control, capital flow restrictions, and anti-money laundering regulations. At the same time, cross-border transactions may also trigger regulatory arbitrage, tax risks, and cross-border transmission of systemic risks, posing potential threats to financial stability.

[0004] Existing OTC derivatives supervision technologies have many deficiencies. Firstly, the problem of data silos is serious, and it is difficult to effectively integrate data among exchanges, banks, and regulatory agencies, resulting in a one-sided supervision perspective. Secondly, traditional rule matching methods lack flexibility and are difficult to cope with the highly customized characteristics of OTC derivative contracts.

[0005] Therefore, the present invention proposes an intelligent adaptation system for cross-border supervision of OTC derivatives based on heterogeneous data fusion. Summary of the Invention

[0006] The present invention aims to solve at least one of the technical problems existing in the prior art. For this purpose, the present invention proposes an intelligent adaptation system for cross-border supervision of OTC derivatives based on heterogeneous data fusion, which reduces the compliance cost and improves the risk control ability.

[0007] To achieve the above object, an intelligent adaptation system for cross-border supervision of OTC derivatives based on heterogeneous data fusion is proposed, including a data acquisition module, a data structuring processing module, a transaction type classification module, a cross-border compliance assessment module, a risk prediction module, and a supervision decision module; wherein, each module is electrically connected;

[0008] The data acquisition module is used to obtain OTC derivatives trading data, including structured data from exchanges, unstructured data from banks, and regulatory review information from regulatory agencies, generate an original data set, and send the original data set to the data structuring processing module;

[0009] A data structuring and processing module, which is used to perform semantic parsing on the original data set by using a text model for financial texts, generate a transaction information table, and send the transaction information table to a transaction type classification module;

[0010] A transaction type classification module, which determines whether a cross-border transaction is involved according to the transaction type in the transaction information table. If a cross-border transaction is involved, the transaction information table is sent to a cross-border compliance assessment module; otherwise, the transaction information table is sent to a risk prediction module;

[0011] A cross-border compliance assessment module, which performs dynamic rule matching on the transaction information table based on a pre-constructed rule knowledge graph to generate a compliance report;

[0012] A risk prediction module, which performs risk prediction on the transaction information table by using a pre-trained transaction graph neural network model, generates a risk warning signal, and executes corresponding warning measures.

[0013] The generation of the original data set includes the following steps:

[0014] Step 11: Collect exchange structured data through multi-source heterogeneous interfaces, including transaction records, product information, and market quotation data;

[0015] The collection of the exchange structured data includes the following sub-steps:

[0016] Step 111: Establish an API connection channel with the world's major over-the-counter derivatives exchanges;

[0017] Step 112: Extract transaction record data from the exchange system through the corresponding API interface according to a predefined data template;

[0018] Step 113: Obtain product information of over-the-counter derivatives from the exchange system through the corresponding API interface;

[0019] Step 114: Collect market quotation data from the exchange system in real time through the corresponding API interface;

[0020] Step 115: The transaction record data, product information of over-the-counter derivatives, and market quotation data constitute the exchange structured data;

[0021] Step 12: Use an adaptive crawler technology to obtain bank unstructured data, including transaction announcements, product manuals, and risk disclosure documents;

[0022] The obtaining of the bank unstructured data includes the following steps:

[0023] Step 121: Configure intelligent crawler tasks to directionally collect relevant documents of over-the-counter derivatives on the official websites of various financial institutions;

[0024] Step 122: Identify and extract product announcement information from the transaction announcements released by banks through intelligent crawler tasks;

[0025] Step 123: Obtain and parse the product manuals released by banks through intelligent crawler tasks, and extract product description information;

[0026] Step 124: Collect the risk disclosure documents released by banks through intelligent crawler tasks, and extract product risk information;

[0027] Step 125: The product announcement information, the product description information, and the product risk information constitute the unstructured data of the bank;

[0028] Step 13: Connect to the regulatory agency's protocol database based on the secure communication protocol to obtain regulatory filing and compliance review information;

[0029] Obtaining regulatory filing and compliance review information includes the following steps:

[0030] Step 131: Establish a secure data exchange channel with financial regulatory agencies in various countries;

[0031] Step 132: Through the secure data exchange channel, regularly obtain regulatory filing data for over-the-counter derivatives transactions according to the data exchange protocol preset by regulatory requirements;

[0032] Step 133: Through the secure data exchange channel, obtain the compliance review information released by the regulatory agency;

[0033] Step 134: The regulatory review information includes the regulatory filing data and the compliance review information;

[0034] Step 14: Classify the structured data of the exchange, the unstructured data of the bank, and the regulatory review information of the regulatory agency according to the product name or product code, and save them as the original dataset;

[0035] The steps for generating a transaction information table by semantically parsing the original dataset using a text model for financial texts are as follows:

[0036] Step 21: Construct a pre-trained language model dedicated to the financial field; the pre-trained language model dedicated to the financial field adopts a variant of the bidirectional encoder representation based on the Transformer architecture;

[0037] Step 22: Apply natural language processing techniques to the unstructured data of the bank to extract key transaction elements and clause information;

[0038] Step 23: Perform structured conversion on the regulatory review information to extract compliance requirements and regulatory decision-making bases;

[0039] Step 24: Based on the pre-trained language model, construct a multi-modal fusion model for financial texts, integrating the semantic information of structured data, key transaction elements and clause information, as well as compliance requirements and regulatory decision-making bases; the multi-modal fusion model adopts a cross-modal attention architecture based on Transformer;

[0040] Step 25: Perform deep information extraction on the fused semantic information, identify and extract the core elements and associated attributes of the transaction, and convert them into structured graph units;

[0041] Step 26: Semantically integrate the graph units to construct a transaction knowledge graph and form an association network among transaction entities;

[0042] Step 27: Based on the transaction knowledge graph, perform multi-dimensional feature extraction to generate a structured representation of the transaction;

[0043] Step 28: Based on the structured representation of the transaction, generate a standardized transaction information table, which includes key fields such as basic transaction information, participant information, product information, risk indicators, and compliance status; the generation of the transaction information table includes two stages: field design and data filling.

[0044] The method for determining whether a cross-border transaction is involved according to the transaction type in the transaction information table is as follows:

[0045] Read the field value of the transaction type of each product in the transaction information table. If the field value indicates a cross-border transaction, it is determined that a cross-border transaction is involved; otherwise, it is determined that no cross-border transaction is involved.

[0046] The steps for generating a compliance report by performing dynamic rule matching on the transaction information table based on the pre-constructed rule knowledge graph are as follows:

[0047] Step 31: Collect the regulatory rule texts of each financial regulatory agency in advance, and construct the rule knowledge graph based on the regulatory rule texts; the construction of the rule knowledge graph includes three stages: rule acquisition, rule parsing, and rule modeling;

[0048] Step 32: Extract rule nodes and their association relationships from the rule knowledge graph to construct a rule matching engine; the rule matching engine includes three core components: a rule parser, a rule executor, and a result collector, which are used to perform dynamic rule matching on the data in the transaction information table;

[0049] Step 33: Generate a structured compliance report through the rule matching engine.

[0050] Performing risk prediction on the transaction information table using a pre-trained transaction graph neural network model to generate risk warning signals and execute corresponding warning measures includes the following steps:

[0051] Step 41: Convert the transaction information table into graph-structured data as the transaction graph network;

[0052] Step 42: Pre-construct the transaction graph neural network model; the transaction graph neural network model includes three components: a graph convolutional encoder, a temporal attention layer, and a risk decoder;

[0053] The transaction graph neural network model adopts an architecture that combines a multi-layer graph convolutional network with a temporal attention mechanism. The transaction graph neural network model includes at least three components: a graph convolutional encoder, a temporal attention layer, and a risk decoder.

[0054] The graph convolutional encoder consists of three layers of graph convolutional networks, with each layer containing 64 convolutional kernels; the first layer of graph convolution aggregates information from the first-order neighborhood of nodes to capture direct transaction relationships; the second layer of graph convolution extends to the second-order neighborhood to capture indirect association patterns; the third layer of graph convolution further expands the receptive field to identify complex network structure features; after each layer of graph convolution, batch normalization and the ReLU activation function are sequentially passed through;

[0055] The temporal attention layer processes the time series features of transactions to capture the time evolution features of abnormal transaction patterns; the temporal attention layer adopts a self-attention mechanism to calculate the correlation weights of transaction features at different time points;

[0056] The risk decoder adopts a multi-task learning framework to simultaneously predict the risk scores in four dimensions: market risk, credit risk, operational risk, and compliance risk; the risk decoder first maps the node-level risk representation to a hidden feature space through two layers of fully connected networks, with the hidden layer dimension being 128, using the LeakyReLU activation function and Dropout regularization; then, for each risk type, a corresponding prediction head is designed, which includes a fully connected layer and a Softmax activation function, and outputs the five-level scoring probability distribution of this risk dimension.

[0057] Step 43: Train the transaction graph neural network model using the pre-collected historical transaction data;

[0058] Step 44: Input the transaction graph network into the pre-trained transaction graph neural network model to generate multi-dimensional risk scores;

[0059] Step 45: Integrate the multi-dimensional risk scores according to the preset risk weights to generate a comprehensive risk level; the generation of the comprehensive risk level adopts a weighted fusion method, and the weights are dynamically adjusted according to the importance and relevance of different risk dimensions.

[0060] Step 46: Generate a risk warning signal according to the comprehensive risk level and a preset warning threshold;

[0061] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0062] The present invention obtains over-the-counter derivatives trading data through a data acquisition module to generate an original data set, sets a data structuring and processing module to perform semantic parsing on the original data set using a text model for financial texts to generate a trading information table, sets a trading type classification module to determine whether a cross-border transaction is involved according to the trading type in the trading information table. If a cross-border transaction is involved, a cross-border compliance assessment module is set to perform dynamic rule matching on the trading information table to generate a compliance report; otherwise, a risk prediction module is set to perform risk prediction on the trading information table using a pre-trained trading graph neural network model to generate a risk warning signal and execute corresponding warning measures. By integrating heterogeneous data sources and combining knowledge graph and graph neural network technologies, the effective integration of data of exchanges, banks, and regulatory agencies is realized, providing a comprehensive regulatory perspective. Secondly, through the dynamic matching mechanism of the rule knowledge graph, the problem of lack of flexibility of traditional rule matching engines is solved, and it can accurately adapt to different types of over-the-counter derivatives contracts. Furthermore, the risk prediction model based on the graph neural network successfully captures the complex correlations in the trading network, realizes the effective identification and warning of systemic risks, greatly reduces the compliance cost, and improves the risk control ability. Description of the Drawings

[0063] Figure 1 It is a module connection relationship diagram of the over-the-counter derivatives cross-border regulatory intelligent adaptation system based on heterogeneous data fusion in Embodiment 1 of the present invention. Detailed Embodiments

[0064] The technical solutions of the present invention will be described clearly and completely below in conjunction with the embodiments. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0065] As Figure 1 shown, the over-the-counter derivatives cross-border regulatory intelligent adaptation system based on heterogeneous data fusion includes a data acquisition module, a data structuring and processing module, a trading type classification module, a cross-border compliance assessment module, a risk prediction module, and a regulatory decision-making module; wherein, each module is connected electrically.

[0066] A data acquisition module, which is used to obtain over-the-counter derivative trading data, including structured data from exchanges, unstructured data from banks, and regulatory review information from regulatory agencies, generate an original data set, and send the original data set to a data structuring processing module;

[0067] A data structuring processing module, which is used to perform semantic parsing on the original data set by using a text model for financial texts, generate a trading information table, and send the trading information table to a trading type classification module;

[0068] A trading type classification module, which determines whether a cross-border transaction is involved according to the trading type in the trading information table. If a cross-border transaction is involved, it sends the trading information table to a cross-border compliance assessment module; otherwise, it sends the trading information table to a risk prediction module;

[0069] A cross-border compliance assessment module, which performs dynamic rule matching on the trading information table based on a pre-constructed rule knowledge graph to generate a compliance report;

[0070] A risk prediction module, which performs risk prediction on the trading information table by using a pre-trained trading graph neural network model, generates a risk warning signal, and executes corresponding warning measures.

[0071] In an embodiment of the present invention, the generation of the original data set includes the following steps:

[0072] Step 11: Collect structured data from exchanges through a multi-source heterogeneous interface, including trading records, product information, and market quotation data;

[0073] Specifically, the collection of structured data from exchanges includes the following sub-steps:

[0074] Step 111: Establish an API connection channel with the world's major over-the-counter derivative exchanges;

[0075] Specifically, in a preferred embodiment of the present invention, the API connection channel adopts a hierarchical architecture design, including a communication layer, an authentication layer, a data exchange layer, and a monitoring layer. The communication layer realizes secure data transmission based on the HTTPS protocol, supports the TLS1.3 encryption standard, and ensures the confidentiality and integrity during the data transmission process. The authentication layer realizes an authentication mechanism based on OAuth 2.0, supports digital certificates and two-factor authentication, and prevents unauthorized access. The data exchange layer adopts a dual-mode design of REST and WebSocket. The REST interface is used to obtain batch historical data, and the WebSocket interface is used to push real-time trading data.

[0076] Step 112: Extract transaction record data from the exchange system through the corresponding API interface according to a predefined data template, including keyword fields such as information of both trading parties, trading time, product type, notional principal, currency, term, and settlement method;

[0077] Specifically, the predefined data template is defined using the JSON Schema specification and consists of two parts: core trading elements and extended attributes. The core trading elements include: a unique transaction identifier, a transaction execution timestamp, legal entity identification codes of both trading parties; the trading product type; the notional principal amount and currency, the transaction start date and maturity date, the settlement method, including options such as physical delivery, cash settlement, or net settlement. The extended attributes set specific fields according to different derivative types. For example, interest rate derivatives include benchmark interest rate, fixed interest rate, floating interest rate, etc.; foreign exchange derivatives include spot exchange rate, forward exchange rate, delivery method, etc.; commodity derivatives include underlying commodity, measurement unit, delivery location, etc.

[0078] Step 113: Obtain product information of over-the-counter derivatives from the exchange system through the corresponding API interface, including product code, contract specifications, pricing model, risk parameters, and market liquidity indicators;

[0079] Specifically, there are two modes for obtaining product information: active query and change subscription. In the active query mode, a full-volume product information synchronization is performed daily after the trading session ends to obtain the latest information of all active products. In the change subscription mode, product information change notifications are received in real time through a WebSocket connection, and updates are obtained immediately when a new product is launched, existing product parameters are adjusted, or a product is delisted.

[0080] The obtained product information includes the following key elements: a unique product identifier, using either ISIN or a custom coding system; product classification information, including main categories such as interest rate, credit, foreign exchange, commodities, etc. and sub-categories such as interest rate swaps, credit default swaps, foreign exchange options, etc.; contract specifications, including contract unit, minimum price change unit, trading time, and settlement rules; a description of the pricing model, including model type such as Black-Scholes, Hull-White, etc. and key parameters; risk parameters, including Greek letter values such as historical volatility, Delta, Gamma, Vega, Theta, etc.; market liquidity indicators, including average daily trading volume, bid-ask spread, and market depth, etc.

[0081] Step 114: Real-time collect market quotation data from the exchange system through the corresponding API interface, including quotes, transaction prices, bid-ask depth, implied volatility, and yield curve, etc.;

[0082] Specifically, the market quotation data collection adopts a multi-level cache and stream processing architecture to ensure the real-time and integrity of the data. A low-latency connection is established with the exchange quotation release system for key derivative varieties, such as foreign exchange options for major currency pairs, interest rate swaps for benchmark interest rates, etc.

[0083] In the specific implementation process of the present invention, the quotation data collection is divided into three frequency levels: the first level is the real-time quotation of core products, which adopts the push mode and is immediately pushed when the data at the exchange end changes, and the update frequency can reach 10 times per second; the second level is the quotation of regular products, which adopts the polling mode and is updated every 5 seconds; the third level is the quotation of low-liquidity products, which is updated once a minute. The collection frequency is dynamically adjusted according to the trading activity of the products, and the collection frequency is automatically increased when the market volatility intensifies.

[0084] The collected market quotation data includes: the latest transaction price and trading volume; the best bid and ask prices and the order quantities; the depth data of the bid and ask sides, recording the cumulative order quantities at different price levels; the implied volatility surface, covering combinations of different strike prices and expiration dates; the yield curve data, including the spot interest rate and forward interest rate at key maturity points; the basis data, reflecting the price difference relationship between the spot and derivatives; the market sentiment indicators, such as the panic index, etc.

[0085] Step 115: The transaction record data, the product information of over-the-counter derivatives, and the market quotation data constitute the exchange structured data;

[0086] Step 12: An adaptive crawler technology is adopted to obtain the unstructured data of banks, including transaction announcements, product manuals, and risk disclosure documents;

[0087] Specifically, the obtaining of the unstructured data of banks includes the following steps:

[0088] Step 121: Configure intelligent crawler tasks to directionally collect relevant documents of over-the-counter derivatives on the official websites of various financial institutions;

[0089] In the specific implementation process of the present invention, the intelligent crawler adopts a distributed architecture design, which at least includes five core components: a task scheduler, a URL manager, a page downloader, a content extractor, and a data storage. The task scheduler is responsible for the creation, allocation, and monitoring of crawler tasks, supporting priority-based task scheduling and an automatic retry mechanism for failed tasks. The URL manager maintains the queue of URLs to be crawled and the set of crawled URLs, realizing URL deduplication and priority sorting. The page downloader is responsible for obtaining the web page content, realizing customized request headers, proxy IP rotation, and access frequency control. The content extractor is responsible for extracting structured information from the web page, supporting multiple extraction methods such as XPath, CSS selectors, and regular expressions. The data storage is responsible for saving the extracted data to a distributed file system and a database;

[0090] Step 122: Through the intelligent crawler task, identify and extract the product announcement information in the transaction announcements released by the bank, including new product releases, transaction term changes, and major market event notifications;

[0091] In the specific implementation process of the present invention, the extraction of transaction announcements adopts a hybrid method combining rules and machine learning. The rule extraction part uses predefined document structure templates and keyword matching rules to quickly locate the page areas that may contain transaction announcements. The machine learning extraction part uses the BERT-BiLSTM-CRF sequence annotation model to perform refined entity recognition and relationship extraction on the text.

[0092] The input of the BERT-BiLSTM-CRF model is the web text paragraph. First, context-related word vector representations are generated through the pre-trained BERT model in the financial field, then the sequence features are captured through the bidirectional LSTM network, and finally the optimal annotation sequence prediction is achieved through the conditional random field layer.

[0093] During the transaction announcement extraction process, the main transaction announcements identified are new product release announcements, including product name, type, target market, and launch time; transaction term change notifications, including the comparison of terms before and after the change and the effective time; market interruption event announcements, including the interruption reason, impact scope, and recovery expectation; regulatory compliance updates, including the implementation of new rules and changes in compliance requirements; risk warning announcements, including risk warnings for specific markets or products;

[0094] Step 123: Through the intelligent crawler task, obtain and parse the product manuals released by the bank, and extract product description information such as product structure, pricing mechanism, risk-return characteristics, and applicable investor types;

[0095] Specifically, the extraction of the product description information includes three stages: layout analysis, content recognition, and information extraction;

[0096] In the layout analysis stage, a document layout analysis model based on Mask R-CNN is used to segment the product manual document into content areas such as titles, texts, tables, charts, and notes.

[0097] In the content recognition stage, specialized processing strategies are applied to different types of content areas. For example, for text areas, the pre-trained BERT model in the financial field is used for semantic understanding; for table areas, table structure recognition algorithms are used to extract row-column relationships and cell contents; for chart areas, image recognition techniques are used to extract key data points and trend information.

[0098] In the information extraction stage, based on a predefined product specification knowledge model, product specification information is extracted from the identified content. The extracted product specification information includes: basic product information, such as name, type, issuer, custodian; product structure, such as underlying assets, linking mechanism, payment structure; pricing mechanism, such as initial pricing, valuation method, price adjustment mechanism; risk-return characteristics, such as expected return rate, maximum loss, risk level; applicable investors, such as investor type, minimum investment amount, suitability requirements; legal terms, such as governing law, dispute resolution mechanism, early termination clause;

[0099] Step 124: Through an intelligent crawler task, collect risk disclosure documents issued by banks, and extract product risk information such as risk types, risk levels, risk mitigation measures, and investor protection clauses;

[0100] Specifically, the collection and analysis of risk disclosure documents adopt natural language processing technology to identify and quantify various risk factors. Its specific processing process includes four steps: risk paragraph identification, risk classification, and risk quantification.

[0101] In the risk paragraph identification stage, a sequence labeling model based on BiLSTM-CRF is adopted to identify the text paragraphs describing risks in the document.

[0102] In the risk classification stage, a hierarchical classification framework is adopted to classify the identified risk content into six categories: market risk, credit risk, liquidity risk, operational risk, legal risk, and compliance risk. Each major category has multiple sub-risk types. Classification uses a multi-label classification model based on XGBoost, supporting that one risk description can belong to multiple risk categories.

[0103] In the risk quantification stage, risk levels and quantification indicators are extracted through natural language processing technology. For content with clearly marked risk levels (such as "high risk", "medium risk", etc.), they are directly extracted and standardized; for descriptive risk content, through sentiment analysis and keyword weight calculation, the severity of the risk is quantified into a numerical level of 1-5. At the same time, quantification indicators related to risks are extracted, such as maximum loss ratio, value at risk, stress test results, etc. The risk category, as well as the risk level and quantification indicators, constitute the product risk information;

[0104] Step 125: The product announcement information, the product specification information, and the product risk information constitute the unstructured data of the bank;

[0105] Step 13: Based on a secure communication protocol, connect to the regulatory agency's protocol database to obtain regulatory filing and compliance review information;

[0106] Specifically, obtaining regulatory filing and compliance review information includes the following steps:

[0107] Step 131: Establish a secure data exchange channel with financial regulatory agencies in various countries;

[0108] In the specific implementation process, the secure data exchange channel uses dedicated network connections and VPN technology to establish point-to-point secure channels with financial regulatory agencies in various countries to obtain regulatory filing and compliance review information of each financial regulatory agency in real time;

[0109] Step 132: Through the secure data exchange channel, regularly obtain regulatory filing data of over-the-counter derivatives transactions according to the preset data exchange protocol required by the regulations;

[0110] Specifically, the obtained regulatory filing data includes: transaction report data, recording details of over-the-counter derivatives transactions reported to the regulatory agency; product registration data, including information on derivative products registered with the regulatory agency; market participant qualification data, including the regulatory qualifications and compliance status of trading entities; limit monitoring data, including position limits and usage of various derivatives; compliance inspection results, including findings from the regulatory agency's compliance inspections of market participants;

[0111] Step 133: Through the secure data exchange channel, obtain compliance review information issued by the regulatory agency, including compliance inspection results, violation penalty decisions, and regulatory policy interpretations;

[0112] In the specific implementation process of the present invention, the compliance review information is obtained by regularly querying structured data related to compliance reviews. The query parameters include inspection objects, such as institution names, registration numbers, inspection times, inspection scopes, problems found, penalty decisions, such as penalty types, penalty amounts, rectification requirements, legal bases, and regulatory interpretations, etc.

[0113] Step 134: The regulatory review information includes the regulatory filing data and the compliance review information;

[0114] Step 14: Classify the structured data of the exchange, the unstructured data of the bank, and the regulatory review information of the regulatory agency according to the name or code of the product, and save it as the original data set;

[0115] It can be understood that this classification can be achieved by using the name or code of the product as the primary key of the database table, and other information in the compliance review information obtained through the secure data exchange channel is used as non-primary attributes of the database table;

[0116] Furthermore, the steps for semantic parsing of the original data set using a text model for financial texts to generate a transaction information table include:

[0117] Step 21: Construct a pre-trained language model dedicated to the financial field;

[0118] Specifically, the pre-trained language model dedicated to the financial field adopts a variant of the bidirectional encoder representation based on the Transformer architecture and conducts domain adaptation pre-training for financial texts. This pre-trained language model mainly consists of three components: an encoder layer, an attention mechanism layer, and an output layer.

[0119] In the specific implementation process of the present invention, the encoder layer is stacked by 12 Transformer blocks, and each Transformer block contains a multi-head self-attention sub-layer and a feed-forward neural network sub-layer. The multi-head self-attention mechanism uses 8 attention heads, and the dimension of each attention head is 64, enabling the model to simultaneously focus on information at different positions. The feed-forward neural network adopts a two-layer structure, with a hidden layer dimension of 3072 and an activation function of GELU.

[0120] In the specific implementation process of the present invention, the pre-training of the pre-trained language model adopts a two-stage strategy: the first stage is to conduct pre-training on a general financial corpus, which includes financial news, research reports, regulatory documents, etc.; the second stage is to continue pre-training on over-the-counter derivative professional documents, including transaction documents, product manuals, regulatory rules, etc. The pre-training tasks include two objectives: masked language model and next sentence prediction (NSP).

[0121] Step 22: Apply natural language processing technology to bank unstructured data to extract key transaction elements and clause information;

[0122] Specifically, the extraction of key transaction elements and clause information adopts a method that combines named entity recognition and relation extraction.

[0123] Named entity recognition adopts a BiLSTM-CRF model to identify financial entities in the text, including transaction parties, product names, transaction amounts, transaction dates, maturity dates, interest rate types, settlement methods, etc. The BiLSTM-CRF model consists of three layers: the embedding layer converts the input text into a vector representation; the BiLSTM layer captures context information and outputs a feature sequence; the CRF layer considers the dependencies between labels and outputs the optimal label sequence.

[0124] Relation extraction adopts a graph convolutional network based on the attention mechanism to identify the relationships between entities, such as "transaction party - transaction product", "product - amount", "product - term", etc. The AGCN model first constructs a document-level entity relationship graph, then learns the representations of entity nodes through multi-layer graph convolutional operations, and finally aggregates node features through the attention mechanism to predict the relationship types between entities.

[0125] For more complex clause information, such as trigger conditions and early termination clauses, a method combining rules and machine learning is adopted for extraction. The rule part utilizes the structural features and keyword patterns of the clauses; the machine learning part uses a sequence-to-sequence model to convert the clause text into a structured representation.

[0126] Step 23: Structurally transform the regulatory review information to extract compliance requirements and the basis for regulatory decisions;

[0127] Specifically, the structural transformation of regulatory review information includes three steps: regulatory text classification, element extraction, and correlation analysis.

[0128] The regulatory text classification adopts a hierarchical classification framework, classifying the regulatory text into five categories: regulatory rules, compliance requirements, inspection findings, penalty decisions, and rectification suggestions. The classification model uses an XLNet-based text classifier, enhancing the understanding ability of long texts and complex contexts through the permutation language modeling pre-training method.

[0129] Element extraction uses a sequence labeling model based on BERT-CRF to extract key elements from the regulatory text, including regulatory clauses, violations, penalty types, penalty amounts, rectification deadlines, etc. The model input is the regulatory text paragraph, and the output is a sequence with BIO annotation, identifying the boundaries and types of various regulatory elements.

[0130] Correlation analysis establishes the logical relationship between regulatory elements, forming a complete regulatory chain of "violation - regulatory clause - penalty decision - rectification requirement". The correlation analysis uses a relationship reasoning model based on the graph attention network (GAT), taking the extracted regulatory elements as graph nodes, learning the correlation strength between nodes through a multi-layer graph attention mechanism, and outputting the relationship network of regulatory elements.

[0131] Step 24: Based on the pre-trained language model, construct a multi-modal fusion model for financial texts, integrating the semantic information of structured data, key transaction elements and clause information, as well as compliance requirements and the basis for regulatory decisions;

[0132] Specifically, the multi-modal fusion model adopts a cross-modal attention architecture based on Transformer to achieve semantic alignment and information fusion of different types of data.

[0133] The core components of the multi-modal fusion model include modality-specific encoders, cross-modal attention layers, and fusion representation layers. The modality-specific encoders design specialized encoding methods for different types of data: structured data generates an initial representation through field embedding and position encoding; text data generates context-related representations through pre-trained language models; tabular data generates representations considering row and column relationships through tabular Transformers.

[0134] The cross-modal attention layer adopts a bidirectional attention mechanism to calculate the correlation between data of different modalities. For example, it calculates the attention weights between the product codes in the transaction records and the text of the product manuals to achieve complementary enhancement of information. In the specific implementation process, for modalities A and B, first, it calculates the attention scores of each element in modality A for all elements in modality B, and then aggregates the information of modality B weighted based on these scores as the enhanced representation of modality A.

[0135] The fusion representation layer adopts a gating mechanism to adaptively control the fusion ratio of information of different modalities. The gating unit receives the representations of each modality and the cross-modal enhanced representation as inputs, and outputs a weight vector between 0 and 1 to control the fusion ratio of the original representation and the enhanced representation. The final fusion representation is further optimized through residual connection and layer normalization to generate a unified representation containing multi-modal information.

[0136] Step 25: Perform in-depth information extraction on the fused semantic information, identify and extract the core elements and associated attributes of the transaction, and convert them into structured graph units;

[0137] Specifically, the in-depth information extraction includes three sub-modules: entity recognition, relation extraction, and attribute filling. Entity recognition is based on the fusion representation, and uses a method that combines conditional random fields (CRF) and deep neural networks to identify key entities such as transaction parties, products, amounts, dates, etc. in the text. Relation extraction uses a relation classifier based on graph convolutional networks to identify semantic relations between entities, such as "transaction party - product", "product - amount", etc. Attribute filling supplements attribute information for the identified entities through a context-aware feature extractor, such as the registered location of the transaction party, the risk level of the product, etc. Organizing the identified entities, relations, and attributes into a semantic graph structure, it can be understood that each group of entities, relations, and attributes constitutes a group of graph units;

[0138] Step 26: Semantically integrate the graph units to construct a transaction knowledge graph and form an association network between transaction entities;

[0139] Specifically, the semantic integration uses a graph fusion algorithm to connect discrete graph units into a complete transaction knowledge graph. The graph fusion process includes two stages: entity alignment and relation mapping.

[0140] In the entity alignment stage, a multi-feature fusion entity matching algorithm is used to identify nodes in different data sources that refer to the same entity. The matching algorithm comprehensively considers the character similarity of entity names, the distribution similarity of attribute values, and the topological similarity of structural contexts, and calculates the comprehensive similarity between entities. For entity pairs whose similarity exceeds the preset threshold, they are marked as different representations of the same entity and merged in subsequent processing.

[0141] In the relationship mapping stage, based on a predefined relationship ontology model, relationship types from different sources are mapped to a unified semantic space. The relationship ontology model adopts a three-layer architecture: the top layer contains core relationship types, such as "participate in transactions", "include products", "apply rules", etc.; the middle layer contains domain-specific relationships, such as "serve as a counterparty", "provide guarantee", etc.; the bottom layer contains specific business relationships, which directly correspond to the relationship expressions in the original data. The mapping process uses a relationship classifier based on a graph neural network to map the original relationships to the standard relationship types in the ontology model.

[0142] Step 27: Based on the transaction knowledge graph, perform multi-dimensional feature extraction to generate a structured representation of the transaction;

[0143] Specifically, the multi-dimensional feature extraction includes three parts: entity feature extraction, relationship feature extraction, and path feature extraction.

[0144] In the entity feature extraction part, the transaction entity nodes in the knowledge graph are characterized. For transaction subject entities, the extracted features include subject types, such as banks, investment companies, hedge funds, etc., registered regions, asset sizes, credit ratings, historical transaction frequencies, and regulatory qualifications, etc.; for product entities, the extracted features include product categories, risk levels, liquidity indicators, price volatilities, and market shares, etc.; for transaction event entities, the extracted features include transaction amounts, transaction times, settlement methods, transaction purposes, and associated markets, etc. Feature extraction uses a representation learning method based on a graph attention network to fuse the attribute information and structural information of the entity into a fixed-dimensional entity vector representation.

[0145] In the relationship feature extraction part, the system analyzes the direct relationship features between entities. Relationship features include relationship types, relationship strengths, relationship durations, and relationship frequencies, etc. The relationship strength is calculated through the interaction frequency and transaction scale between associated entities; the relationship duration is determined based on the time span of historical transaction records; the relationship frequency reflects the number of interactions within a specific time window. The system uses a time-series aware relationship embedding model to encode the relationship features into a relationship vector representation.

[0146] In the path feature extraction part, the system analyzes the indirect association paths between entities. The system uses a meta-path analysis method to pre-define a series of semantically clear path patterns, such as "trading party - product - trading party" which reflects the products traded in common, "trading party - regulatory agency - trading party" which reflects being regulated by the same regulatory agency, etc. For each meta-path, the system calculates features such as the number of path instances, path diversity, and path importance. Path feature extraction uses a method that combines random walk and attention mechanism to generate a path feature representation considering path semantics.

[0147] The entity vector, relationship vector, and path feature representation serve as the structured representation of the transaction;

[0148] Step 28: Based on the structured representation of the transaction, generate a standardized transaction information table, which includes key fields such as basic transaction information, participant information, product information, risk indicators, and compliance status;

[0149] Specifically, the generation of the transaction information table includes two stages: field design and data filling.

[0150] In the field design stage, based on the internationally common over-the-counter derivatives transaction reporting standards and local regulatory requirements, design the table structure of the transaction information table. In the specific implementation process of the present invention, the transaction information table includes five major categories of fields: basic transaction information fields, including the transaction unique identifier, transaction type, such as whether it is a cross-border transaction, transaction execution time, transaction effective date, transaction expiration date, and transaction status, etc.; participant information fields, including the legal entity identification codes of both parties to the transaction, registration regions, institution types, and qualification status, etc.; product information fields, including product type, underlying asset, notional amount, currency, and settlement method, etc.; risk indicator fields, including market risk value, credit risk score, liquidity risk level, and operational risk indicators, etc.; compliance status fields, including regulatory reporting status, compliance inspection results, and risk warning signs, etc.

[0151] In the data filling stage, first obtain the values of each field in the transaction structured representation, including core elements such as transaction number, transaction time, transaction party information, transaction amount, transaction type, and transaction status. Then, according to the preset data filling rules, map the above field values to the corresponding positions in the transaction information table. For example, for the transaction party information, parse out the detailed information of the transaction initiator and recipient, including name, address, contact information, account information, etc., and fill this information into the corresponding data fields. For the transaction amount, convert it to the standard format, and calculate the equivalent local currency amount according to the exchange rate information at the time of the transaction, and fill it into the amount-related fields. For the transaction type, according to the predefined type mapping table, convert the original transaction type into the standard type code used internally by the system, and fill it into the type field. For the transaction status, through the status mapping table, convert the original status into the system standard status code, and fill it into the status field. In addition, according to the time information of the transaction, the settlement date, clearing date, and other time points of the transaction can be automatically calculated and filled into the corresponding time fields. For the goods or service information included in the transaction, parse it into detail items, each detail item includes information such as goods name, quantity, unit price, total price, etc., and fill it into the transaction detail-related fields. Finally, according to the filled data, automatically calculate the values of some derived fields, such as total transaction amount, taxes and fees, net amount, etc., and fill these calculation results into the corresponding fields;

[0152] Furthermore, the method for determining whether a cross-border transaction is involved according to the transaction type in the transaction information table is as follows:

[0153] Read the field values of the transaction types of each product in the transaction information table. If the field value indicates a cross-border transaction, it is determined that a cross-border transaction is involved; otherwise, it is determined that no cross-border transaction is involved.

[0154] Furthermore, the steps for dynamically matching rules for the transaction information table based on a pre-constructed rule knowledge graph to generate a compliance report are as follows:

[0155] Step 31: Collect the regulatory rule texts of each financial regulatory agency in advance, and construct the rule knowledge graph based on the regulatory rule texts.

[0156] Specifically, the construction of the rule knowledge graph includes three stages: rule acquisition, rule parsing, and rule modeling.

[0157] In the rule acquisition stage, the system collects regulatory rule texts from multiple sources, including international agreements such as the Basel Accord and the ISDA Master Agreement, national regulations such as the Securities Law and the Regulations on the Administration of Futures Trading, regulatory agency regulations such as the regulations issued by the central bank, the securities regulatory commission, and the foreign exchange administration, and industry self-regulatory rules. Multiple methods are used for rule acquisition: direct docking with structured rule libraries; regular crawling of regulatory agency official websites; API calls to regulatory databases; and OCR recognition and conversion of paper documents.

[0158] In the rule parsing stage, natural language processing is performed on the acquired rule texts to extract the core elements of the rules. The parsing process combines deep learning with rule templates. First, named entity recognition technology is applied to identify the subject entity, action entity, condition entity, and result entity in the rule text; then, relation extraction technology is applied to identify the semantic relationships between entities, such as "apply to", "prohibit", "require", etc.; finally, semantic role labeling technology is applied to identify the condition part, action part, and result part of the rule. The parsing result forms a structured rule representation, including rule ID, rule source, release time, effective time, scope of application, condition expression, and result expression.

[0159] In the rule modeling stage, the structured rule representation is converted into a knowledge graph model. The rule knowledge graph adopts a multi-level ontology model: the top-level ontology defines core concept categories, such as regulatory subjects, regulatory objects, regulatory behaviors, and regulatory results; the domain ontology defines concepts in specific regulatory fields, such as cross-border transactions, foreign exchange management, anti-illegal fund transfers, etc.; the application ontology defines concepts in specific business scenarios, such as over-the-counter derivatives, structured products, etc. Rules are represented as complex relationship patterns connecting multiple entities in the graph. Each rule contains condition nodes, action nodes, and result nodes, as well as the logical relationships between them. The relationships between rules are also explicitly modeled, including inclusion relationships, inheritance relationships, conflict relationships, and complementary relationships. This knowledge graph model is the rule knowledge graph;

[0160] Step 32: Extract rule nodes and their associated relationships from the rule knowledge graph to construct a rule matching engine;

[0161] Specifically, the rule matching engine includes three core components: a rule parser, a rule executor, and a result collector, which are used to perform dynamic rule matching on the data in the transaction information table.

[0162] Among them, the rule parser reads the rule nodes in the rule knowledge graph and converts each rule into an executable logical expression. For each rule, parse its trigger conditions, scope of application, execution actions, and priorities to form a structured rule object.

[0163] Traverse the transaction information table. For each transaction record, call the rule executor to perform rule matching. The rule executor filters out the set of rules relevant to the current transaction record according to the scope of application of the rules. The rule executor evaluates the trigger conditions of each rule in turn according to the priority order of the rules. When a transaction record meets the trigger conditions of a rule, the system marks the rule as "triggered" and records the specific reasons and relevant data items for the trigger.

[0164] For each triggered rule, the system executes the actions defined by the rule, including but not limited to: compliance marking, risk level assessment, abnormal behavior identification, and regulatory requirement verification. The execution results are recorded in the result collector.

[0165] The result collector summarizes the results of all rule executions to generate preliminary compliance analysis data. The system performs aggregation processing on the analysis data to identify compliance issues, risk points, and abnormal patterns in the transaction records.

[0166] Step 33: Generate a structured compliance report through the rule matching engine.

[0167] The compliance report at least includes the following main parts: compliance overview, risk assessment, details of abnormal transactions, compliance with regulatory requirements, and recommended measures. The compliance overview section presents the overall compliance status of transaction information through statistical analysis, including statistical data such as compliance rate, risk distribution, and main problem types. The risk assessment section details all identified risk points, including risk types, risk levels, risk descriptions, and indexes of relevant transaction records. The system hierarchically displays the risks according to their severity and impact scope; The details of abnormal transactions section lists all transaction records marked as abnormal, including transaction identifiers, abnormal types, triggered rule numbers, and abnormal descriptions. The system provides detailed information on abnormal transactions for further investigation and handling; The compliance with regulatory requirements section evaluates the degree of compliance of transaction information with various regulatory requirements, including lists of compliance items, non-compliance items, and items to be confirmed. According to the regulatory rule nodes in the rule knowledge graph, the compliance rate of each regulatory requirement is calculated; The recommended measures section provides targeted improvement suggestions and compliance measures based on the identified compliance issues and risk points. Relevant solution nodes are extracted from the rule knowledge graph to generate an actionable list of recommendations.

[0168] Further, the risk prediction of the transaction information table is performed using a pre-trained transaction graph neural network model to generate risk warning signals, and the corresponding warning measures are executed as follows:

[0169] Step 41: Convert the transaction information table into graph-structured data as the transaction graph network;

[0170] The conversion of the transaction information table into graph-structured data can be in the form of backtracking, by rereading the transaction knowledge graph corresponding to the transaction information table, and the graph structure of this transaction knowledge graph can be used as the transaction graph network;

[0171] Step 42: Pre-build the transaction graph neural network model; the transaction graph neural network model adopts an architecture that combines a multi-layer graph convolutional network with a temporal attention mechanism. The transaction graph neural network model includes three components: a graph convolutional encoder, a temporal attention layer, and a risk decoder;

[0172] In the specific implementation process of the present invention, the graph convolutional encoder consists of three layers of graph convolutional networks, each layer containing 64 convolutional kernels. The first layer of graph convolution aggregates information from the first-order neighborhood of nodes to capture direct transaction relationships; the second layer of graph convolution extends to the second-order neighborhood to capture indirect association patterns; the third layer of graph convolution further expands the receptive field to identify complex network structure features. After each layer of graph convolution, batch normalization and the ReLU activation function are sequentially passed through to prevent overfitting and enhance the non-linear expression ability.

[0173] The temporal attention layer processes the time series features of transactions and captures the time-evolution features of abnormal transaction patterns. The temporal attention layer adopts a self-attention mechanism to calculate the correlation weights of transaction features at different time points. In a specific implementation, the transaction sequence in the past 12 months is used as the input, and query, key, and value matrices are generated through linear projection. Attention scores are calculated and weighted aggregation is performed to obtain node representations considering time dependence.

[0174] The risk decoder adopts a multi-task learning framework to simultaneously predict the risk scores in four dimensions: market risk, credit risk, operational risk, and compliance risk. The risk decoder first maps the node-level risk representations to a hidden feature space through a two-layer fully connected network. The dimension of the hidden layer is 128, and the LeakyReLU activation function and Dropout regularization are used. Then, for each risk type, a dedicated prediction head is designed, which includes a fully connected layer and a Softmax activation function to output the five-level scoring probability distribution of this risk dimension. The market risk prediction head focuses on price volatility and liquidity indicators; the credit risk prediction head focuses on the default probability of the counterparty; the operational risk prediction head focuses on abnormal transaction processes and signs of system failures; the compliance risk prediction head focuses on violations of regulatory rules. The scores of each risk dimension adopt a grading system from 1 to 5, where level 1 indicates extremely low risk and level 5 indicates extremely high risk.

[0175] Step 43: Train the transaction graph neural network model through pre-collected historical transaction data;

[0176] Specifically, pre-collect the transaction history records in the past 12 months and use them as sample data for the transaction graph neural network model, and adopt a self-supervised learning method for pre-training during the training process of the transaction graph neural network model. The pre-training tasks include: node attribute prediction, randomly masking some attribute features of the nodes, and training the model to predict these masked attributes; edge prediction, randomly removing some edges in the graph, and training the model to predict the existence of these edges; subgraph structure prediction, training the model to identify common risk-related subgraph patterns. The pre-training uses the Adam optimizer, with an initial learning rate of 0.001 and a weight decay of 0.0001;

[0177] Step 44: Input the transaction graph network into the pre-trained transaction graph neural network model to generate multi-dimensional risk scores;

[0178] It can be understood that on the basis that the transaction graph neural network model has been trained, input the transaction graph network into this transaction graph neural network model, and the risk decoder of the transaction graph neural network model can directly output risk scores in four different dimensions;

[0179] Step 45: Integrate the multi-dimensional risk scores according to the preset risk weights to generate a comprehensive risk level;

[0180] Specifically, the generation of the comprehensive risk level adopts a weighted fusion method, and dynamically adjusts the weights according to the importance and relevance of different risk dimensions. First, calculate the weighted average of the scores of the four risk dimensions. The initial weights are set as 0.3 for market risk, 0.3 for credit risk, 0.2 for operational risk, and 0.2 for compliance risk.

[0181] In a further preferred embodiment, the synergistic effect of multi-dimensional risks can be further identified through a risk resonance detection algorithm. For example, when it is detected that two or more risk dimensions are simultaneously in a high-level state, the evaluation value of the comprehensive risk level can be increased. The risk resonance detection adopts a method combining rules and machine learning. The rule part defines the logical relationship between risk dimensions, and the machine learning part learns the risk collaboration pattern through historical cases. Finally, add the weighted average and the risk resonance adjustment value, and map the comprehensive risk level to different levels of risk levels. In the specific implementation process of the present invention, it can be set as extremely low risk (Level 1), low risk (Level 2), medium risk (Level 3), high risk (Level 4), and extremely high risk (Level 5).

[0182] Step 46: Generate a risk warning signal according to the comprehensive risk level and the preset warning threshold;

[0183] Specifically, the generation of the risk warning signal is based on a three-level warning mechanism. The system compares the comprehensive risk level with the preset warning threshold to generate a warning signal of the corresponding level. The warning levels include three types: reminder level, warning level, and emergency level. When the comprehensive risk level is Level 3 (medium risk), the system generates a reminder-level warning, suggesting that the supervisor pay attention to the relevant transactions; when the comprehensive risk level is Level 4 (high risk), the system generates a warning-level warning, suggesting that the supervisor conduct an in-depth review; when the comprehensive risk level is Level 5 (extremely high risk), the system generates an emergency-level warning, suggesting that immediate regulatory measures be taken.

[0184] The above embodiments are only used to illustrate the technical method of the present invention and not to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical method of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical method of the present invention.

Claims

1. An intelligent adaptation system for cross-border supervision of over-the-counter derivatives based on heterogeneous data fusion, characterized in that It includes a data collection module, a data structuring and processing module, a transaction type classification module, a cross-border compliance assessment module, a risk prediction module, and a regulatory decision-making module; among them, each module is electrically connected to each other; The data collection module is used to obtain over-the-counter derivatives trading data, including structured data from exchanges, unstructured data from banks, and regulatory review information from regulatory agencies, generate an original data set, and send the original data set to the data structuring and processing module; The data structuring and processing module uses a text model for financial texts to perform semantic parsing on the original data set, generate a transaction information table, and send the transaction information table to the transaction type classification module; among them, performing semantic parsing on the original data set using a text model for financial texts to generate a transaction information table includes the following steps: Step 21: Construct a pre-trained language model dedicated to the financial field; the pre-trained language model dedicated to the financial field uses a bidirectional encoder representation variant based on the Transformer architecture; Step 22: Apply natural language processing techniques to bank unstructured data to extract key transaction elements and clause information; Step 23: Perform structured conversion on regulatory review information to extract compliance requirements and regulatory decision-making bases; Step 24: Based on the pre-trained language model, construct a multi-modal fusion model for financial texts to integrate the semantic information of structured data, key transaction elements and clause information, and compliance requirements and regulatory decision-making bases; the multi-modal fusion model uses a cross-modal attention architecture based on Transformer; Step 25: Perform deep information extraction on the fused semantic information, identify and extract the core elements and associated attributes of the transaction, and convert them into structured graph units; Step 26: Semantically integrate the graph units to construct a transaction knowledge graph and form an association network between transaction entities; Step 27: Based on the transaction knowledge graph, perform multi-dimensional feature extraction to generate a structured representation of the transaction; Step 28: Based on the structured representation of the transaction, generate a standardized transaction information table; the generation of the transaction information table includes two stages: field design and data filling; The transaction type classification module determines whether a cross-border transaction is involved according to the transaction type in the transaction information table. If a cross-border transaction is involved, the transaction information table is sent to the cross-border compliance assessment module; otherwise, the transaction information table is sent to the risk prediction module; The cross-border compliance assessment module performs dynamic rule matching on the transaction information table based on a pre-constructed rule knowledge graph to generate a compliance report; The risk prediction module uses a pre-trained transaction graph neural network model to predict the risk of the transaction information table, generate a risk warning signal, and execute corresponding warning measures.

2. The cross-border regulatory intelligent adaptation system for over-the-counter derivatives based on heterogeneous data fusion according to claim 1, wherein The generation of the original data set includes the following steps: Step 11: Collect exchange structured data through a multi-source heterogeneous interface, including transaction records, product information, and market quotation data; Step 12: Use an adaptive crawler technology to obtain bank unstructured data, including transaction announcements, product manuals, and risk disclosure documents; Step 13: Connect to the regulatory agency's protocol database based on the secure communication protocol to obtain regulatory filing and compliance review information; Step 14: Classify the structured data of the exchange, the unstructured data of the bank, and the regulatory review information of the regulatory agency according to the name or code of the product, and save them as the original dataset.

3. The cross-border regulatory intelligent adaptation system for over-the-counter derivatives based on heterogeneous data fusion according to claim 2, wherein The collection of the structured data of the exchange includes the following sub-steps: Step 111: Establish an API connection channel with the world's major over-the-counter derivatives exchanges; Step 112: Extract transaction record data from the exchange system through the corresponding API interface according to the predefined data template; Step 113: Obtain product information of over-the-counter derivatives from the exchange system through the corresponding API interface; Step 114: Real-time collect market quotation data from the exchange system through the corresponding API interface; Step 115: The transaction record data, the product information of over-the-counter derivatives, and the market quotation data constitute the structured data of the exchange.

4. The cross-border regulatory intelligent adaptation system for over-the-counter derivatives based on heterogeneous data fusion according to claim 3, wherein The acquisition of the unstructured data of the bank includes the following steps: Step 121: Configure an intelligent crawler task to directionally collect relevant documents of over-the-counter derivatives on the official websites of various financial institutions; Step 122: Identify and extract product announcement information of the transaction announcements issued by the bank through the intelligent crawler task; Step 123: Obtain and parse the product manuals issued by the bank through the intelligent crawler task, and extract product description information; Step 124: Collect the risk disclosure documents issued by the bank through the intelligent crawler task, and extract product risk information; Step 125: The product announcement information, the product description information, and the product risk information constitute the unstructured data of the bank.

5. The cross-border regulatory intelligent adaptation system for over-the-counter derivatives based on heterogeneous data fusion according to claim 4, characterized in that, The acquisition of regulatory filing and compliance review information includes the following steps: Step 131: Establish a secure data exchange channel with the financial regulatory agencies of various countries; Step 132: Regularly obtain regulatory filing data of over-the-counter derivatives transactions through the secure data exchange channel according to the predefined data exchange protocol required by the supervision; Step 133: Obtain the compliance review information issued by the regulatory agency through the secure data exchange channel; Step 134: The regulatory review information includes the regulatory filing data and the compliance review information.

6. The intelligent adaptation system for cross-border supervision of over-the-counter derivatives based on heterogeneous data fusion according to claim 5, wherein The method for determining whether a cross-border transaction is involved according to the transaction type in the transaction information table is as follows: Read the field value of the transaction type of each product in the transaction information table. If the field value indicates a cross-border transaction, it is determined that a cross-border transaction is involved; otherwise, it is determined that no cross-border transaction is involved.

7. The cross-border regulatory intelligent adaptation system for over-the-counter derivatives based on heterogeneous data fusion according to claim 6, wherein The generation of a compliance report by dynamically matching rules for the transaction information table based on the pre-constructed rule knowledge graph includes the following steps: Step 31: Pre-collect the regulatory rule texts of each financial regulatory agency, and construct the rule knowledge graph based on the regulatory rule texts; the construction of the rule knowledge graph includes three stages: rule acquisition, rule parsing, and rule modeling; Step 32: Extract rule nodes and their association relationships from the rule knowledge graph to construct a rule matching engine; the rule matching engine includes three core components: a rule parser, a rule executor, and a result collector, which are used to perform dynamic rule matching on the data in the transaction information table; Step 33: Generate a structured compliance report through the rule matching engine.

8. The cross-border regulatory intelligent adaptation system for over-the-counter derivatives based on heterogeneous data fusion according to claim 7, wherein The risk prediction of the transaction information table is performed using a pre-trained transaction graph neural network model to generate risk warning signals. The implementation of corresponding warning measures includes the following steps: Step 41: Convert the transaction information table into graph-structured data as the transaction graph network. Step 42: Pre-construct the transaction graph neural network model. The transaction graph neural network model includes three components: a graph convolutional encoder, a temporal attention layer, and a risk decoder. Step 43: Train the transaction graph neural network model using pre-collected historical transaction data. Step 44: Input the transaction graph network into the pre-trained transaction graph neural network model to generate multi-dimensional risk scores. Step 45: Integrate the multi-dimensional risk scores according to the preset risk weights to generate a comprehensive risk level. The generation of the comprehensive risk level adopts a weighted fusion method, and the weights are dynamically adjusted according to the importance and relevance of different risk dimensions. Step 46: Generate risk warning signals based on the comprehensive risk level and the preset warning threshold.

9. The cross-border regulatory intelligent adaptation system for over-the-counter derivatives based on heterogeneous data fusion according to claim 8, characterized in that The graph convolutional encoder consists of three layers of graph convolutional networks, with each layer containing 64 convolutional kernels. The first layer of graph convolution aggregates information from the first-order neighborhood of nodes to capture direct transaction relationships. The second layer of graph convolution extends to the second-order neighborhood to capture indirect association patterns. The third layer of graph convolution further expands the receptive field to identify complex network structure features. After each layer of graph convolution, batch normalization and the ReLU activation function are sequentially applied. The temporal attention layer processes the time series features of transactions to capture the time evolution features of abnormal transaction patterns. The temporal attention layer adopts a self-attention mechanism to calculate the correlation weights of transaction features at different time points. The risk decoder adopts a multi-task learning framework to simultaneously predict the risk scores in four dimensions: market risk, credit risk, operational risk, and compliance risk. The risk decoder first maps the node-level risk representation to a hidden feature space through two layers of fully connected networks. The dimension of the hidden layer is 128, and the LeakyReLU activation function and Dropout regularization are used. Then, for each risk type, a corresponding prediction head is designed, which includes a fully connected layer and a Softmax activation function, and outputs the probability distribution of the five-level score for this risk dimension.

Images