45 results about "Risk model" patented technology

The invention discloses a financial deep counterfeiting real-time detection and defense method and system based on a multi-modal large model, and the method comprises the steps: obtaining multi-modal data in a financial transaction scene, and carrying out the desensitization of an edge end; performing dynamic time sequence alignment on the multi-modal data, and calculating a synchronous error of lip motion and voice by adopting a dynamic time warping algorithm; inputting the features into a dynamic risk modeling layer, and generating dynamic risk features in combination with the updated risk feature library; analyzing the features through a double-flow GAN detector, and outputting a forgery probability; and a detection result is input into a compliance verification layer, the supervision file is analyzed through a legal BERT, a structured rule is generated, and real-time transaction interception and block chain log recording are executed. The method protects user privacy and data security, combines the risk feature library updated in real time, and has high flexibility and adaptability. According to the design of the double-flow GAN detector, image and video stream information is fully utilized, and the accuracy and reliability of detection are further improved.

The invention relates to the technical field of invoice risk control management, and discloses an invoice risk control management method and system based on multi-source data linkage. According to the method, multi-source data of the whole invoice business process is collected in real time through a multi-source data fusion gateway. And performing streaming ETL processing and encryption integration on the multi-source data to generate an encrypted unified data view. And based on the view, constructing an association hypergraph by utilizing a dynamic graph association technology, analyzing an entity relationship by virtue of a graph neural network, and identifying false invoice making gang and repeated financing risks. And calculating a dynamic risk score by adopting a risk quantification and traceability technology, and outputting a risk level, a responsibility node identifier and a capital flow thermodynamic diagram. When the risk score exceeds a threshold value, risk gradient information is safely shared among the group headquarters, the molecular companies and the tax-bank institutions, and linkage processing actions such as invoice freezing, red character notification or credit adjustment are triggered. According to a manual reexamination result, the risk model is finely adjusted through a knowledge self-updating technology, and the rule base is hot-updated.

The invention discloses a bank credit evaluation system and method based on deep learning, and relates to the technical field of data analysis and evaluation, and the system comprises a data acquisition module, a credit evaluation module, a credit analysis module, and a decision output module. According to the method, customer groups are divided in a layered and detailed manner through data acquisition, a dynamic updating and exception rechecking mechanism is applied, a data set of key information is constructed, potential risks of daily demand expenditure of customers on credit are evaluated, and meanwhile income stability analysis is performed according to work to judge the stability of the work development trend of the customers; considering the influence of work on customer credit conditions, establishing a skill equivalent replacement risk model, simulating equivalent work replacement execution, performing customer economic risk evaluation, predicting credit risks generated by work, and integrating a historical credit data set and the work stability and development trend reflected by a work trend change coefficient to obtain a credit risk evaluation result. And a comprehensive customer credit assessment value is generated according to the fund demand and the risk condition embodied by the daily demand prediction risk value.

A fraud detection system preemptively identifies and mitigates fraudulent financial transactions in real-time across a diverse range of digital and physical transaction sources. Transaction data may be acquired from various platforms and categorized into rule sets specific to transaction types. Categorized data is aligned with existing risk profiles to construct a dynamic hysteresis model. This model, evaluated by a decision tree algorithm, identifies potential fraud by integrating immediate transaction details with a comprehensive historical data analysis, thus enabling advanced trend analysis and pattern recognition. Key features identified by the decision tree are used to form a heuristics model, which is then analyzed by a transformer network risk model. A feedback loop enhances the system's effectiveness by incorporating decision outcomes back into the model training server, thus refining the training dataset and continuously improving the accuracy of the risk model.

The invention discloses a camera gun video resource positioning method and related equipment. The method comprises the following steps: acquiring camera gun information of each financial network; obtaining structured risk model data for the daily recovery high-risk model troubleshooting index; determining a target financial network point corresponding to each piece of risk model data in the structured risk model data; according to the structured risk model data, the target financial network point corresponding to each piece of risk model data, the camera gun information of each financial network point and the current popularity value of each camera gun, determining a plurality of current retrieval camera guns; according to the current user operation log, updating the current popularity value corresponding to each camera gun; and according to the updated current popularity value corresponding to each camera gun, determining a plurality of updated current retrieval camera guns. The method can achieve the precise pushing of the video resources needed by the daily redisk of the website, improves the video verification efficiency, and can be widely applied to the technical field of artificial intelligence.

The invention discloses an enterprise financial risk real-time early warning and visualization method based on stream-oriented computing, and the method comprises the following steps: 1) multi-source heterogeneous data real-time collection and access: collecting stream-oriented data from a plurality of data sources in real time, the data sources comprising internal business data and external financial data; 2) stream data cleaning and standardization: preprocessing the data accessed in real time to ensure that the formats of the data from different sources are uniform and can be used for subsequent calculation; 3) real-time financial index calculation and risk model evaluation: 3.1) real-time index calculation; (3.2) carrying out real-time risk assessment; 4) risk early warning and graded pushing; 5) performing real-time visualization and interactive analysis; according to the method, efficient analysis, visual presentation and man-machine interaction of complex data are achieved in an automatic and intelligent mode, and therefore support is provided for decision making.

The embodiment of the invention relates to the technical field of artificial intelligence, and provides an apartment financial management intelligent auxiliary method and system based on artificial intelligence, and the method comprises the steps: obtaining apartment management data, hotel management data and transaction data disclosed by a market, and obtaining multi-source data; storing the multi-source data to a cloud financial management center and encrypting the multi-source data; outputting a prediction result of the multi-source data through a financial analysis prediction network; the financial analysis and prediction network comprises an income prediction model and a component analysis optimization model which run in parallel; constructing a visual report form and an analysis report based on the prediction result, and dynamically displaying the report form and the analysis report to the user; and inputting the multi-source data into the financial management model, performing analysis through an auditing rule in the financial management model to generate a financial auditing report, calling the financial risk model to monitor the financial auditing report, and synchronously pushing the financial auditing report to a manual auditing terminal for verification and approval processing. According to the method, the data utilization rate is increased, and the cost optimization accuracy, the financial management efficiency and the safety are improved.

The invention provides a risk list supervision system and method, the risk list supervision system comprises a database, a risk analysis unit, a risk list generation unit and a server, the database is used for storing account data of accounts and transaction cash flow data; a transaction risk model of the risk analysis unit is used for analyzing transaction cash flow data to generate at least one risk factor corresponding to the transaction cash flow data and calculating the weight of the at least one risk factor to generate a risk score, and when the risk score is larger than a risk standard threshold value, the risk list generation unit generates and sends a risk account list, and the risk account list is used for generating a risk account. The server marks the account corresponding to the account data in the risk account list as a temporary forbidden state; therefore, the risk list supervision system provided by the invention can identify the high-risk account timely, efficiently, quickly and automatically, so as to improve the efficiency and reduce the labor cost.

Methods and systems are presented for dynamically modifying a risk model based on detected shifts of input features. Input values corresponding to an input feature and associated with transactions may be obtained. A distribution of the input values is determined, and compared against a benchmark distribution. An anomaly is detected when a difference between the distribution of the input values and the benchmark distribution exceeds a threshold. Based on the detected anomaly, a risk model configured to perform risk predictions for incoming transaction requests may be modified. Input values corresponding to the input feature may be monitored to determine if the anomaly is sustained or receded. The modified risk model may be reverted back to the original risk model when the anomaly no longer exists.

A kind of multi-risk management method based on worst-case scenario conditional value at risk model, by setting the operation mode of virtual power plant in single electricity market, and after analyzing its bidding profit and constraint, the typical distribution scene of uncertain variable that will bring risk is constructed by using Monte Carlo simulation method and fast pre-generation elimination technology based on probability distance, then the conditional value at risk model under worst-case scenario is constructed, and the maximum conditional value at risk under worst-case scenario is taken as the goal, max-min problem is converted into semi-definite programming problem by using dual theory to solve, and the optimal bidding strategy of virtual power plant after management risk is solved.The present application can be used as a powerful tool for decision-making, provide theoretical support for virtual power plant to formulate scheme to improve the prediction accuracy of random variable, so as to minimize the market risk caused by prediction error, and make the transaction strategy closer to the optimal strategy under accurate situation.

The invention relates to the technical field of financial credit risk control, in particular to a risk control model construction method and system for identifying customer credit risks, and the method comprises the steps: constructing a cross-product dichotomy task data set; training a sample similarity screening model; outputting a probability score; performing equal-frequency division according to probability scores and calculating the concentration and multiple of new product positive samples in each subgroup; selecting the group with the highest multiple as a high-similarity candidate region, carrying out multi-round equal-frequency division, judging whether equal-frequency division is stopped or not by combining a convergence judgment mechanism, and determining to supplement positive samples; and integrating the new product positive sample and the supplementary positive sample to carry out weighted training. According to the method and the device, the sample similarity screening model is constructed, and multiple rounds of equal-frequency segmentation, KS index monitoring and a target sample multiple evaluation mechanism are combined, so that old product samples highly similar to target positive sample features are dynamically identified, high-quality expansion of an original positive sample set is realized, and the training effect and generalization ability of a subsequent risk model are improved.

The invention relates to the field of data risk assessment, and discloses a project information data assessment method and system based on big data, and the method comprises the steps: obtaining a calculation budget execution rate, a cost deviation and an unallocated cost, substituting the calculation budget execution rate, the cost deviation and the unallocated cost into a financial risk model to calculate a project financial risk, and then obtaining a milestone achievement rate, a progress deviation and floating time; substituting into a progress risk model, calculating a project progress risk, then obtaining a professional skill matching degree, average working experience, a personnel loss rate and average performance, substituting into a personnel risk model, selecting a project personnel risk, if financial, progress or personnel risk coefficients exceed a threshold value, triggering a high-risk alarm, and if all risk coefficients are lower than the threshold value, judging that the project is abnormal. And calculating a potential risk coefficient, if the potential risk coefficient exceeds a threshold value, prompting a potential risk, evaluating the risk from multiple dimensions of finance, progress and personnel, covering key risk points of a project, introducing the potential risk coefficient, and capturing a hidden risk.

This application discloses a risk model iteration method, apparatus, device, and product, relating to the field of model iteration technology. The method includes: collecting operational data of an initial risk model in a production environment; detecting the operational data using detection indicators; when the operational data reaches the iteration trigger condition, creating a model iteration task and loading a pre-configured business dataset; collaboratively optimizing the hyperparameters of the initial risk model using the business dataset to obtain a globally optimal feature subset and an optimal hyperparameter combination; and iterating the initial risk model using the globally optimal feature subset and the optimal hyperparameter combination to obtain a final risk model. This solves the problems of lacking automated closed-loop iteration, delayed response, and the need for professional personnel to operate the risk model, thus failing to quickly adapt to dynamic business changes and improving the efficiency of risk model iteration.

The invention relates to the field of artificial intelligence, and provides a distributed training method and device of a risk control model, equipment and a medium, which are applied to the financial field, and the method comprises the following steps: separately fragmenting risk control data of each mechanism node to obtain a plurality of pieces of target fragmented data; inputting the plurality of pieces of target fragment data and the global parameters into a risk control model to carry out distributed training and output first target parameters; aggregating, adjusting and collecting the first target parameter to obtain feedback data; and predicting the feedback data to obtain a prediction error value, updating the training parameters by using the prediction error value to obtain updated training parameters, and re-initializing the global parameters according to the updated training parameters to start a new round of distributed training. By implementing the embodiment of the invention, distributed training and parallel computing of each node of the risk model are realized, the model training time is shortened, the risk control model can quickly respond to the risk change of the current insurance market, the timeliness and accuracy of risk early warning are improved, and thus the efficiency is improved.

Risk element information indicating risk elements is acquired. An asset capable of becoming a fault due to a risk element indicated in the risk element information, and a fault probability that the asset becomes the fault are specified based on static configuration information. A risk model in which the asset capable of becoming the fault and the fault probability are associated is generated in advance. In response to a designated input, an one asset to be evaluated is specified as an evaluation target asset, based on the designated input. A risk model related to the evaluation target asset is specified. A risk evaluation value being an index indicating a risk of the evaluation target asset is calculated based on the fault probability of the evaluation target asset and the static configuration information. The risk evaluation value of the evaluation target asset is associated with the asset of the risk model.

The invention belongs to the technical field of artificial intelligence, and relates to a data generation method and device, computer equipment and a storage medium, and the method comprises the steps: obtaining disaster intensity data, and calculating corresponding disaster intensity change data based on the disaster intensity data; judging whether the disaster intensity change data meets a dynamic optimization condition or not; if yes, updating the initial loss rate function based on the disaster intensity data to obtain a target loss rate function; performing function updating processing on the risk model based on the target loss rate function to obtain a target risk model; solving the target risk model based on a quantum annealing strategy to obtain target reinsurance proportion data; generating a target reinsurance scheme based on the target reinsurance proportion data and outputting the target reinsurance scheme; in addition, the invention also relates to a block chain technology, and the target reinsurance scheme can be stored in a block chain. The method can be applied to an insurance risk assessment scene in the field of financial science and technology, and the generation efficiency and accuracy of the target reinsurance scheme are effectively improved.

The invention relates to the technical field of financial data analysis, in particular to a fund product analysis system based on big data, which comprises an event flow analysis module, a correlation analysis module, a dynamic optimization module, a strategy configuration module and a data synchronization verification module. The system collects financial market data in real time, generates an event feature matrix, combines a graph neural network and reinforcement learning technology, dynamically adjusts risk model parameters, generates a risk early warning and strategy optimization scheme, and monitors the consistency of a physical data source and a virtual model at the same time. According to the method, the accuracy and the real-time performance of fund product analysis can be improved, and the scientificity and the reliability of investment decision making are enhanced.

The embodiment of the invention provides a parameter adjustment method and device for a credit risk model, equipment and a storage medium, and belongs to the technical field of computers and the technical field of finance. The method comprises the following steps: acquiring a plurality of initial configuration parameters determined by a credit risk model in an initial coronary challenge; dividing the plurality of initial configuration parameters into a plurality of groups, and inputting the initial configuration parameters into a pre-trained multi-granularity hybrid expert model for processing to obtain new configuration parameters corresponding to the credit risk model; and updating the configuration parameters of the credit risk model according to the new configuration parameters to obtain a new credit risk model, and performing a new chafing challenge according to the new credit risk model. The method is used for improving the accuracy of the credit risk control model by flexibly adjusting the model parameters, and further enabling the credit risk model for carrying out the chay challenge to provide a more accurate credit risk identification result.

The invention relates to the technical field of big data, and provides a financial business risk control management system based on a big data model, and the system comprises the steps: accessing labor data in multiple platforms and financial data of different accounts, carrying out the cross-domain convergence and disambiguation of different accounts of a user, and forming a corresponding unified identity ID; the method comprises the following steps of: collecting multi-source time sequence events from each business system and risk control system by taking a unified identity (ID) as a main key, merging and connecting in series according to a time sequence, and generating a behavior time axis reflecting cross-platform working intensity and stability; carrying out statistics and operation on the labor behavior time axis and the fund behavior time axis, and generating a health degree index which simultaneously represents credit stability and fraud suspiciousness; and based on the health degree index and a preset label sample, jointly training a credit risk model and a fraud risk model, outputting a risk score and a strategy decision result at a business risk control node, and carrying out cooperative control on credit admission, quota management and fraud interception of the zero worker crowd.

A computer-implemented method for testing a risk model trained to output a risk determination for open banking transactions that includes: unit testing the risk model implemented in an application programming interface (API) platform; performing manual and automated feature validation of the risk model by comparing feature outputs from the risk model to expected feature output values; performing functional validation of the risk model according to an accuracy metric; benchmark testing the risk model, implemented in the API platform, using simulated scenario test data sets; load testing the risk model; after performing the foregoing, deploying the risk model as a production risk model with a web portal in a production environment; and dog-food testing the production risk model using a plurality of simulated scenario test data sets configured to simulate real production usage.

This invention relates to the field of financial transaction risk identification technology, and discloses a method for identifying risks in programmed abnormal securities transactions. The method includes: acquiring current transaction data for the current time period; extracting features from the current transaction data to obtain current transaction features corresponding to the current transaction data; constructing current graph structure data based on the current transaction features; inputting the current graph structure data into a first transaction risk identification model to obtain a first risk identification result; the first transaction risk model is obtained by training a graph neural network model based on graph structure sample data; the graph structure sample data is constructed based on transaction feature data of programmed transaction samples from historical time periods; the first risk identification result includes account information indicating transaction risks in the current time period. Through the above method, this invention effectively improves the accuracy of identifying programmed abnormal transactions.

The invention relates to the technical field of enterprise risk management, in particular to an enterprise intelligent financial risk prediction method fused with big data analysis, comprising: collecting structured financial data and unstructured text data of a target enterprise, the text data comprising a supply chain contract, a supervision punishment announcement and an industry policy document; generating a supply chain default risk signal and a policy sensitivity risk signal through entity relationship extraction; inputting the structured data and the risk signal into a time sequence diagram neural network, and outputting a dynamic risk probability value and a risk conduction node sequence; constructing an industrial chain risk conduction map according to the result, setting a dynamic risk probability value double threshold to trigger a grading early warning signal, and further pushing a key intervention path and risk control suggestion; the method has the multi-modal risk modeling capability and a map-level response mechanism, and can be widely applied to enterprise credit assessment, supply chain financial risk control and macroscopic supervision scenes.

The application relates to the technical field of invoice risk control management, and discloses an invoice risk control management method and system based on multi-source data linkage. The method realizes real-time collection of multi-source data of an invoice business whole process through a multi-source data fusion gateway. The multi-source data is subjected to flow ETL processing and encryption integration to generate an encrypted unified data view. Based on the view, a correlation hypergraph is constructed by using a dynamic graph correlation technology, and entity relationships are analyzed by means of a graph neural network to identify virtual invoice gang and repeated financing risks. Risk quantification and traceability technologies are adopted to calculate a dynamic risk score, and a risk level, a responsibility node identifier and a fund flow direction heat map are output. When the risk score exceeds a threshold value, risk gradient information is safely shared among a group headquarters, a branch company and a tax and bank institution to trigger linkage disposal actions such as invoice freezing, red letter notification or credit adjustment. According to an artificial review result, a risk model is fine-tuned and a rule library is hot-updated by using a knowledge self-updating technology.

A computer-implemented method for testing a risk model trained to output a risk determination for open banking transactions that includes: unit testing the risk model implemented in an application programming interface (API) platform; performing manual and automated feature validation of the risk model by comparing feature outputs from the risk model to expected feature output values; performing functional validation of the risk model according to an accuracy metric; benchmark testing the risk model, implemented in the API platform, using simulated scenario test data sets; load testing the risk model; after performing the foregoing, deploying the risk model as a production risk model with a web portal in a production environment; and dog-food testing the production risk model using a plurality of simulated scenario test data sets configured to simulate real production usage.

The invention discloses an intelligent power transaction management platform for medium-and-long-term spot green certificate fusion, relates to the technical field of power transaction management, and aims at realizing the purpose of realizing high-speed power transaction by synchronously collecting real-time data and generating a dynamic reference vector by adopting linear optimization as a unified benchmark price sequence of cross-time scale transaction. Calculating a cross fluctuation coupling index and an evidence power synchronization deviation index by using the vector, and generating a collaborative risk adjustment coefficient through a Gaussian-Laplacian mixed risk model to guide risk management; assembling a hedging packet based on the smart contract, atomizing and locking power and a corresponding green certificate, and ensuring compliance and traceability by means of a block chain; the hedging packet profit and loss and the deposit state are tracked in real time, and a self-adaptive threshold value is adopted to trigger addition or release of the deposit; iteratively correcting the reference vector by using settlement profit and loss data through the prediction model; price discovery, renewable energy source integration and market stability are remarkably improved, and economic and environmental double benefits are provided.

The present invention relates to a system and method for quantum-classical agentic orchestration for real-time financial risk modeling and architectural drift remediation in multi-cloud environments. The invention provides an integrated computing arrangement comprising at least one classical processing unit, a quantum interface controller, data acquisition and normalization processors, a risk modeling processor, an orchestration control unit, an architectural state monitoring unit, a drift detection processor, and a remediation execution unit configured to operate in coordination. The system continuously receives financial data streams and infrastructure telemetry from distributed cloud environments, processes the data to generate standardized analytical representations, and performs hybrid computational analysis by combining classical evaluation with quantum-assisted optimization for complex financial risk scenarios. The system further monitors configuration states, deployment topology, and resource allocation parameters to detect deviations from predefined architectural baselines.

The invention provides a financial risk control method and device, equipment, a storage medium and a program product, and relates to the field of financial science and technology or other related fields. The method comprises the steps of obtaining multi-modal data of a transaction, preprocessing the multi-modal data of the transaction to obtain feature data of the transaction, obtaining a preliminary screening risk point of the transaction when it is determined that the transaction meets a preset risk condition through a rule engine according to the feature data, and obtaining the preliminary screening risk point of the transaction according to the preliminary screening risk point. And performing risk depth analysis on the multi-modal data of the transaction through a pre-trained risk model to obtain a hidden risk point of the transaction, and generating a risk report of the transaction based on the preliminary screening risk point and the hidden risk point of the transaction. According to the method provided by the invention, the real-time performance and accuracy of financial risk identification are improved, and the reliability of risk control and the transaction security are enhanced.

This invention discloses a risk label processing method, apparatus, risk control method, device, and storage medium, comprising: responding to a risk scenario processing request and obtaining the feature matrix of the training samples corresponding to the risk model and their training sample labels; re-encoding the feature matrix of the training samples using labels to generate a labeled feature matrix; sorting the labeled feature matrix by label importance and extracting multiple important labels based on the sorting results; for each important label, obtaining real feature data of the important label over multiple historical periods, and using the risk model to obtain predicted feature data of the important label over multiple historical periods; and calculating the risk level of each important label based on the combined feature data of each important label using a maximum risk drawdown model. Based on the maximum risk drawdown algorithm, this invention can automatically select high-risk labeled customer groups and can periodically and automatically generate monitoring SQL for high-risk labeled customer groups for monitoring deployment.

The invention relates to the technical field of financial data management and control, in particular to a financial multi-source heterogeneous data real-time fusion and risk identification method, which comprises the steps of data acquisition, time alignment and metadata annotation, multi-modal preprocessing, real-time fusion mapping, risk modeling analysis, signal fusion decision, real-time early warning, credible evidence storage and online learning optimization. According to the invention, by introducing a distributed acquisition agent module and a nanosecond timestamp synchronization mechanism, the problems of time desynchrony and non-uniform formats among different data sources are solved; by combining metadata annotation and a micro-service preprocessing engine, structured, semi-structured and non-structured data can be standardized, the availability and semantic consistency of multi-modal data are remarkably improved, and a high-quality data basis is provided for subsequent risk identification.

The invention discloses a financial investment user operation optimization method based on big data, and relates to the technical field of data analysis, and the method comprises the following steps: setting a first constraint condition, constructing a risk model, setting an inspection index, inspecting the risk model according to the inspection index, and compensating the risk model according to an inspection result. And outputting a new investment type proportion according to the new risk model. Through accurate risk assessment and personalized investment suggestions, a user can obtain an investment scheme better meeting own requirements, big data and a machine learning technology can analyze market trends in real time, investors can be helped to make more scientific investment decisions, and by dynamically adjusting investment portfolios, financial institutions better cope with market fluctuations, and the investment portfolios can be dynamically adjusted. Potential risks are reduced, and the optimized portfolio can better balance risks and earnings, so that the rate of return on investment is improved.