CN114358607A - A kind of risk monitoring method and device - Google Patents

Abstract

The invention provides a risk monitoring method and device, relates to the technical field of artificial intelligence, and can be used in the financial field or other technical fields. The method comprises the following steps: and inputting the fusion data to a preset risk monitoring model, and taking an output result of the preset risk monitoring model as a risk monitoring result. The device performs the above method. The risk monitoring method and the risk monitoring device provided by the embodiment of the invention not only can reduce the labor cost and improve the efficiency, but also can overcome the defect of limited data dimension, and can comprehensively perform risk early warning aiming at the service characteristics.

Description

A kind of risk monitoring method and device

technical field

The invention relates to the technical field of artificial intelligence, in particular to a risk monitoring method and device.

Background technique

With the development of data technology, massive amounts of data are generated, and at the same time, there are data risks. At present, the technology of automatic monitoring of risk indicators is more and more widely used, and there are related technologies for risk early warning, but the shortcomings include: the data dimension of concern is limited, and risk early warning cannot be comprehensively carried out according to business characteristics; The labor cost is too high and the efficiency is low.

SUMMARY OF THE INVENTION

In view of the problems in the prior art, the embodiments of the present invention provide a risk monitoring method and device, which can at least partially solve the problems in the prior art.

On the one hand, the present invention proposes a risk monitoring method, comprising:

Input fusion data into a preset risk monitoring model; the preset risk monitoring model is obtained by training a neural network model according to the fusion sample data; the fusion data is fused with customer-dimensional comprehensive knowledge map data, product-dimensional comprehensive knowledge map data, and enterprise-level comprehensive knowledge map data. Dimensional comprehensive knowledge graph data; the customer-dimensional comprehensive knowledge graph data, the product-dimensional comprehensive knowledge graph data, and the enterprise-dimensional comprehensive knowledge graph data include respective corresponding information graph data, and risk monitoring corresponding to each information graph data map data;

The output result of the preset risk monitoring model is used as the risk monitoring result.

Wherein, the neural network model is obtained by combining BRNN and LSTM.

Wherein, the risk monitoring result corresponds to the customer dimension, the product dimension and the enterprise dimension respectively; correspondingly, after the step of using the output result of the preset risk monitoring model as the risk monitoring result, the risk monitoring method Also includes:

If at least one of the risk monitoring result in the customer dimension, the risk monitoring result in the product dimension, and the risk monitoring result in the enterprise dimension is at risk, a risk warning message corresponding to the at least one risky risk monitoring result is generated.

Wherein, obtaining the fusion data includes:

Build customer information map data, product information map data and enterprise information map data respectively;

According to the risk monitoring thresholds and corresponding data items of the data items corresponding to the customer information map data, the product information map data and the enterprise information map data, respectively construct the customer risk monitoring map data, the product risk monitoring map data and the corresponding data items. Enterprise risk monitoring map data;

respectively fuse the customer information map data and the customer risk monitoring map data, the product information map data and the product risk monitoring map data, and the enterprise information map data and the enterprise risk monitoring map data, The customer-dimension integrated knowledge map data, the product-dimension integrated knowledge map data, and the enterprise-dimension integrated knowledge map data are obtained.

Wherein, after the step of acquiring the fusion data, the risk monitoring method further includes:

The degree centrality calculation method is used to calculate the node weights in the customer-dimension integrated knowledge map data, the product-dimension integrated knowledge map data, and the enterprise-dimension integrated knowledge map data.

Wherein, after the utilization centrality calculation method calculates the node weights in the customer-dimension integrated knowledge map data, the product-dimension integrated knowledge map data, and the enterprise-dimension integrated knowledge map data, the risk monitoring Methods also include:

Assign the node weights to entities in the dimensional comprehensive knowledge graph data, the product dimensional comprehensive knowledge graph data, and the enterprise dimensional comprehensive knowledge graph data, and assign the node weights to the dimensions of the comprehensive knowledge graph data and product dimensions. Comprehensive knowledge graph data and enterprise dimension comprehensive knowledge graph data for vectorized representation;

The MainfoldE algorithm is used to perform operations on the vectorized customer dimension comprehensive knowledge graph data, product dimension comprehensive knowledge graph data and enterprise dimension comprehensive knowledge graph data.

Wherein, the risk monitoring method further includes:

Input the fusion data of the vectorized representation after operation processing into the preset risk monitoring model, and continue to perform the subsequent steps.

Wherein, obtaining the enterprise risk monitoring map data includes:

Obtain enterprise information, identify the text content of the enterprise information, and match the text content with the words in the preset positive public opinion database and the preset negative public opinion database respectively;

Determine the type of text content according to the matching result; the type of text content includes positive content or negative content;

Traverse all the text content, and obtain the number of text content corresponding to the positive content or the negative content according to the type of the text content of each text content;

If the quantity of text content corresponding to the positive content is less than the quantity of text content corresponding to the negative content, determine that the enterprise information is negative enterprise information;

Obtain enterprise interaction information corresponding to the negative enterprise information, and if the statistical value of at least one interaction index data in the enterprise interaction information is greater than a preset statistical data threshold, use the negative enterprise information as the enterprise risk monitoring Graph data.

In one aspect, the present invention provides a risk monitoring device, comprising:

The input unit is used to input the fusion data into the preset risk monitoring model; the preset risk monitoring model is obtained by training the neural network model according to the fusion sample data; the fusion data is fused with the customer-dimensional integrated knowledge map data and the product-dimensional integrated Knowledge graph data and enterprise-dimension integrated knowledge-graph data; the customer-dimension integrated knowledge-graph data, the product-dimension integrated knowledge-graph data, and the enterprise-dimension integrated knowledge-graph data include their corresponding information graph data, as well as the corresponding information graph data. Risk monitoring map data corresponding to the data;

The monitoring unit is configured to use the output result of the preset risk monitoring model as the risk monitoring result.

In another aspect, an embodiment of the present invention provides an electronic device, including: a processor, a memory, and a bus, wherein,

The processor and the memory communicate with each other through the bus;

The memory stores program instructions executable by the processor, and the processor invokes the program instructions to execute the following methods:

Input fusion data into a preset risk monitoring model; the preset risk monitoring model is obtained by training a neural network model according to the fusion sample data; the fusion data is fused with customer-dimensional comprehensive knowledge map data, product-dimensional comprehensive knowledge map data, and enterprise-level comprehensive knowledge map data. Dimensional comprehensive knowledge graph data; the customer-dimensional comprehensive knowledge graph data, the product-dimensional comprehensive knowledge graph data, and the enterprise-dimensional comprehensive knowledge graph data include respective corresponding information graph data, and risk monitoring corresponding to each information graph data map data;

The output result of the preset risk monitoring model is used as the risk monitoring result.

Embodiments of the present invention provide a non-transitory computer-readable storage medium, including:

The non-transitory computer-readable storage medium stores computer instructions that cause the computer to perform the following methods:

Input fusion data into a preset risk monitoring model; the preset risk monitoring model is obtained by training a neural network model according to the fusion sample data; the fusion data is fused with customer-dimensional comprehensive knowledge map data, product-dimensional comprehensive knowledge map data, and enterprise-level comprehensive knowledge map data. Dimensional comprehensive knowledge graph data; the customer-dimensional comprehensive knowledge graph data, the product-dimensional comprehensive knowledge graph data, and the enterprise-dimensional comprehensive knowledge graph data include respective information graph data, and risk monitoring corresponding to each information graph data map data;

The output result of the preset risk monitoring model is used as the risk monitoring result.

The risk monitoring method and device provided by the embodiments of the present invention input fusion data into a preset risk monitoring model, and use the output result of the preset risk monitoring model as the risk monitoring result, which can not only reduce labor costs, improve efficiency, but also overcome Due to the shortcomings of limited data dimensions, comprehensive risk early warning is carried out according to business characteristics.

Description of drawings

In order to explain the embodiments of the present invention or the technical solutions in the prior art more clearly, the following briefly introduces the accompanying drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only These are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can also be obtained according to these drawings without creative efforts. In the attached image:

FIG. 1 is a schematic flowchart of a risk monitoring method provided by an embodiment of the present invention.

FIG. 2 is a schematic structural diagram of a risk monitoring device according to an embodiment of the present invention.

FIG. 3 is a schematic diagram of a physical structure of an electronic device according to an embodiment of the present invention.

Detailed ways

In order to make the purposes, technical solutions and advantages of the embodiments of the present invention more clearly understood, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings. Here, the exemplary embodiments of the present invention and their descriptions are used to explain the present invention, but not to limit the present invention. It should be noted that, the embodiments in the present application and the features in the embodiments may be arbitrarily combined with each other if there is no conflict.

FIG. 1 is a schematic flowchart of a risk monitoring method provided by an embodiment of the present invention. As shown in FIG. 1 , the risk monitoring method provided by an embodiment of the present invention includes:

Step S1: input fusion data into a preset risk monitoring model; the preset risk monitoring model is obtained by training a neural network model according to the fusion sample data; the fusion data is fused with customer-dimensional comprehensive knowledge graph data and product-dimensional comprehensive knowledge graph data and enterprise-dimension integrated knowledge map data; the customer-dimension integrated knowledge map data, the product-dimension integrated knowledge map data, and the enterprise-dimension integrated knowledge map data include respective information map data corresponding to each information map data risk monitoring map data.

Step S2: Use the output result of the preset risk monitoring model as the risk monitoring result.

In the above step S1, the device inputs fusion data into a preset risk monitoring model; the preset risk monitoring model is obtained by training a neural network model according to the fusion sample data; the fusion data is fused with customer-dimensional comprehensive knowledge map data, product Dimensional integrated knowledge map data and enterprise-dimensional integrated knowledge map data; the customer-dimension integrated knowledge map data, the product-dimension integrated knowledge map data, and the enterprise-dimension integrated knowledge map data include their corresponding information map data, as well as the corresponding information map data. The risk monitoring map data corresponding to the information map data. The apparatus may be a computer device that executes the method. It should be noted that the customer, product, and enterprise-related data involved in the embodiment of the present invention are all authorized by the user.

Customer information graph data is knowledge graph data constructed based on customer information. Customer information can include:

The basic information of the customer, including: the scale of the customer's financial assets at the end of the previous month, the online verification certificate of the customer's identity, and the customer's credit rating, etc.

Basic information about family members associated with the customer, including: the scale of financial assets of the family members at the end of the previous month, online verification certificates and credit ratings, etc.

Information on companies held by the customer, including: the number of companies held, the industry to which the companies belong, shareholding ratio and shareholding amount, etc.

Product information graph data is knowledge graph data constructed based on product information. Product information can include:

The underlying investment information of trust products, including: underlying asset allocation ratio, investment product name, product code, product type and position ratio, etc.

Operational indicator data of trust product duration, including: unit net worth, historical net worth and range yield, etc.

The disclosure report information returned by the trust company includes: the text content of the disclosure report and the return date of the disclosure report, etc.

The enterprise information graph data is the knowledge graph data constructed according to the enterprise information, and the enterprise information can include enterprise news and so on.

The risk monitoring atlas data corresponding to each information atlas data can be understood as atlas data used for risk monitoring of each information atlas data.

The customer risk monitoring map data corresponding to the customer information map data is used for risk monitoring of the customer information map data.

The product risk monitoring map data corresponding to the product information map data is used for risk monitoring of the product information map data.

The enterprise risk monitoring graph data corresponding to the enterprise information graph data is used for risk monitoring of the enterprise information graph data.

The fusion of sample data can be understood as the following data pre-selected that can be used as the training of the neural network model:

Customer dimension comprehensive knowledge graph data, product dimension comprehensive knowledge graph data and enterprise dimension comprehensive knowledge graph data.

Neural network model (Neural Networks, NN) is a complex network system formed by a large number of simple processing units (called neurons) that are widely connected to each other. It reflects many basic features of human brain functions and is a highly complex nonlinear dynamic learning system. Neural networks have large-scale parallel, distributed storage and processing, self-organization, self-adaptation and self-learning capabilities, and are especially suitable for dealing with inaccurate and ambiguous information processing problems that need to consider many factors and conditions at the same time.

The training sample data adopts the customer dimension comprehensive knowledge graph data C1, the product dimension comprehensive knowledge graph data C2 and the enterprise dimension comprehensive knowledge graph data C3 constructed from the stock data. The training sample data includes risk data, risk edge data and risk-free data. The training method is a conventional training method in the field, and will not be repeated here.

In the above step S2, the device takes the output result of the preset risk monitoring model as the risk monitoring result.

In the risk monitoring method provided by the embodiment of the present invention, the fusion data is input into a preset risk monitoring model, and the output result of the preset risk monitoring model is used as the risk monitoring result, which can not only reduce labor costs, improve efficiency, but also overcome the data dimension There are limited shortcomings, and comprehensive risk warning is carried out according to business characteristics.

Further, the neural network model is obtained by combining BRNN and LSTM. BRNN bidirectional cyclic neural network (Bidirectional RNN), the main problem to be solved is that the elements of the previous sequence cannot perceive the problem of the output of the subsequent sequence.

Long short-term memory (LSTM) is a special RNN, mainly to solve the problem of gradient disappearance and gradient explosion during long sequence training.

Taking into account the relationship between the data and the timeliness of the data, the model adopts a bidirectional recurrent neural network (BRNN). At the same time, considering that the recurrent neural network has limited storage of context information, it will lose part of the time series information earlier in time. Therefore, a long short-term memory network layer LSTM is added to the model.

In summary, considering information contextual relevance, timeliness and information storage, this neural network model has five layers, including the input layer Input Layer, BRNN-Forward Layer, BRNN-Backward Layer, LSTM, and the output layer Output Layer.

The risk monitoring method provided by the embodiments of the present invention can overcome the shortcomings of a single model, thereby improving the model operation efficiency and the accuracy of the model output results.

Further, the risk monitoring result corresponds to the customer dimension, the product dimension and the enterprise dimension respectively; correspondingly, after the step of taking the output result of the preset risk monitoring model as the risk monitoring result, the risk monitoring Methods also include:

If at least one of the risk monitoring result in the customer dimension, the risk monitoring result in the product dimension, and the risk monitoring result in the enterprise dimension is at risk, a risk warning message corresponding to the at least one risky risk monitoring result is generated. For example, if there is a risk in the product dimension risk monitoring result, a risk warning message for the product dimension is generated.

The risk monitoring method provided by the embodiment of the present invention can obtain the risk monitoring results with risks in a more targeted and timely manner.

Further, obtaining the fusion data, including:

Build customer information map data, product information map data and enterprise information map data respectively; customer information map data is recorded as A1, and the construction is detailed as follows:

The basic information of the customer, the basic information data of the family members associated with the customer, and the information of the company held by the customer are regarded as structured and semi-structured data, and entities and attributes are extracted as the nodes of the knowledge graph through the knowledge graph extraction technology. The nodes include but not Limited to (customer name, customer age, customer credit rating, customer asset scale, number of customer-owned companies, and customer-owned company ratio, etc.), the relationship is used as the edge of the knowledge graph, and the customer information graph data A1 is constructed. For example, <customer A, holding company, number of holding companies>, <customer A, sister, family member B>, <family member B, credit, credit rating> can form a knowledge graph triple.

For the construction steps of the product information map data A2 and the enterprise information map data A3, reference may be made to the description of the above-mentioned construction steps of the customer information map data A1, and will not be repeated here.

According to the risk monitoring thresholds and corresponding data items of the data items corresponding to the customer information map data, the product information map data and the enterprise information map data, respectively construct the customer risk monitoring map data, the product risk monitoring map data and the corresponding data items. Enterprise risk monitoring map data; customer risk monitoring map data is recorded as B1, and the specific instructions for construction are as follows:

First, obtain the data item corresponding to the customer information map data. The data item can be understood as the data item used for risk monitoring, which can include:

For asset fluctuation data and share fluctuation data, the corresponding risk monitoring thresholds can be set to 30% and 10% respectively.

The data items corresponding to the customer information map data can also be data irrelevant to the risk monitoring threshold, such as the result of customer identification status, for example, the online verification result of the customer himself or related customers is abnormal, or the customer credit report results have bad records, Such results are included in the area of concern.

If the size of the client's financial assets decreases by more than 30% at the end of the month, the data item of the asset fluctuation data will be included in the area of concern.

If the shareholding ratio of the company held by the client decreases by more than 10%, the data item of the share fluctuation data will be included in the area of concern.

Build customer risk monitoring graph data B1, including:

The knowledge graph triples are formed by <asset fluctuation, threshold, 30%>, <threshold, drop by 30%, concern area>, <customer credit report, bad record, concern area>.

Further, obtaining the enterprise risk monitoring map data, including:

Obtain enterprise information, identify the text content of the enterprise information, and match the text content with the words in the preset positive public opinion database and the preset negative public opinion database respectively; the text content recognition technology is a mature technology in the field, no longer required. Repeat. The preset positive public opinion database refers to a preset positive word corpus, and the preset negative public opinion database refers to a preset negative word corpus.

Determine the type of text content according to the matching result; the type of text content includes positive content or negative content; if the matching result is a positive word corpus in the preset positive public opinion database, it means that the text content is positive content; if the matching result is in If the negative word corpus in the negative public opinion database is preset, it means that the text content is negative content.

Traverse all text content, and obtain the number of text content corresponding to the positive content or the negative content according to the type of text content of each text content; an example is as follows:

All the text contents are 5, which are denoted as a～e respectively, where a～b are positive contents and c～e are negative contents; the number of text contents corresponding to positive contents is 2, and the number of text contents corresponding to negative contents is 3.

If the quantity of text content corresponding to the positive content is less than the quantity of text content corresponding to the negative content, the company information is determined to be negative company information; referring to the above example, the company information is negative company information.

Obtain enterprise interaction information corresponding to the negative enterprise information, and if the statistical value of at least one interaction index data in the enterprise interaction information is greater than a preset statistical data threshold, use the negative enterprise information as the enterprise risk monitoring Graph data. The enterprise interaction information may include interactive indicator data such as the number of likes, the number of comments, and the number of reposts for the enterprise information.

Taking the number of likes as an example, if the corresponding preset statistical data threshold is set to 5,000 times, and if the number of likes is greater than 5,000 times, the negative enterprise information will be used as the enterprise risk monitoring map data.

Specifically, negative enterprise information can be put into the focus area as the data content in the construction of enterprise risk monitoring map data.

You can also use the following methods:

Divide the positive content into four grades from A to D according to the degree of positive influence from high to low, and divide the negative content into four grades from E to H in the order of the degree of negative influence from low to high, which will satisfy the preset statistical data. The negative enterprise information of the threshold condition is classified as G level or H level, and the negative enterprise information of G level or H level is put into the concern area, and then the concern area is used as the data content in the enterprise risk monitoring map data.

The risk monitoring method provided by the embodiment of the present invention can further improve the accuracy of the risk monitoring result by reasonably acquiring the enterprise risk monitoring map data.

Further, obtaining the product risk monitoring atlas data, including:

Divide investment products into equity, fixed income, commodities and financial derivatives, hybrids, etc., to obtain the proportion of various investment products and the volatility of product operation, and the corresponding risk monitoring thresholds can be set to 30% and 20% respectively. %. The calculation formula of the product operation volatility is a mature technology in the field, and will not be repeated here.

If the proportion of investment products exceeds 30%, the data item is included in the concern area.

If the product run volatility exceeds 10%, the data item is included in the area of interest.

The data item corresponding to the product information map data can also be data irrelevant to the risk monitoring threshold, for example:

Regularly crawl relevant disclosure reports or financial statements through the pre-set corporate official website address. Identify various financial metrics in a report or statement.

Build multivariate models and use Z-scores to predict the likelihood of business financial failure. Five variables can be used to reflect the solvency, profitability, and operational capability indicators of the company in each cycle, and comprehensive analysis can predict the possibility of financial failure or bankruptcy of the company.

Among them, the solvency index X1=(working capital/total assets)×100, X4=(total market value of common stock and preferred stock/total book value of liabilities)×100.

Enterprise profitability X2 = (retained earnings / total assets) × 100, X3 = (earnings before interest and taxes / total assets) × 100.

Enterprise operating capability X5 = sales revenue/total assets.

Z is the value of the discriminant function, Z=W1×X1+W2×X2+W3×X3+W4×X4+W5×X5, when the Z value is lower than the discriminant threshold, put Z into the attention area. Among them, the discriminant threshold can be set independently according to the actual situation, and can be selected as 1.81.

W1~W5 are the weights corresponding to X1~X5, which can be set independently according to the actual situation, and can be selected as 0.012, 0.014, 0.033, 0.006 and 0.999 respectively.

By reasonably obtaining product risk monitoring map data, the accuracy of risk monitoring results can be further improved.

For the construction steps of the product risk monitoring map data B2 and the enterprise risk monitoring map data B3, reference may be made to the above description of the construction of the customer risk monitoring map data B1, and will not be repeated here.

respectively fuse the customer information map data and the customer risk monitoring map data, the product information map data and the product risk monitoring map data, and the enterprise information map data and the enterprise risk monitoring map data, The customer-dimension integrated knowledge map data, the product-dimension integrated knowledge map data, and the enterprise-dimension integrated knowledge map data are obtained. The comprehensive knowledge graph data of the customer dimension is recorded as C1, and the specific construction instructions are as follows:

Match the entities of the customer information graph data A1 and the customer risk monitoring graph data B1, for example: point the client assets in the customer information graph data A1 and the asset fluctuations in the customer risk monitoring graph data B1 to the same entity, and use the entity-linked The technology carries out knowledge merging to realize the fusion of customer information map data A1 and customer risk monitoring map data B1, and obtain customer-dimensional comprehensive knowledge map data C1. For the construction steps of the product-dimension integrated knowledge graph data C2 and the enterprise-dimension integrated knowledge graph data C3, reference may be made to the above description of the construction of the customer-dimension integrated knowledge graph data C1, and will not be repeated here.

The risk monitoring method provided by the embodiment of the present invention helps to accurately and comprehensively conduct risk monitoring by comprehensively integrating data into customer-dimension comprehensive knowledge map data C1, product-dimension comprehensive knowledge map data C2, and enterprise-dimension comprehensive knowledge map data C3 monitor.

Further, after the step of acquiring the fusion data, the risk monitoring method further includes:

The degree centrality calculation method is used to calculate the node weights in the customer-dimension integrated knowledge map data, the product-dimension integrated knowledge map data, and the enterprise-dimension integrated knowledge map data.

Since each entity in the knowledge graph is connected in different ways with the previous entities, including one-to-many, many-to-many, and many-to-one, the importance of each node is also different. The degree of a node refers to the associated edges of a node. The more associated edges, the greater the weight of the node.

For example: shareholding enterprise entity node, customer entity in customer information map data A1, and enterprise share entity and enterprise share threshold entity in customer risk monitoring map data B1, compared with customer age only associated with customer entity, shareholding enterprise entity Nodes should be given greater weights.

The degree centrality calculation method is a mature method in the field, and will not be repeated here.

In the risk monitoring method provided by the embodiment of the present invention, the closeness of the association relationship between entities is reflected by accurately calculating the node weight, thereby helping to improve the accuracy of the output result of the preset risk monitoring model.

Further, after the utilization centrality calculation method calculates the node weights in the customer-dimension integrated knowledge map data, the product-dimension integrated knowledge map data, and the enterprise-dimension integrated knowledge map data, the risk Monitoring methods also include:

Assign the node weights to entities in the dimension integrated knowledge graph data, the product dimension integrated knowledge graph data, and the enterprise dimension integrated knowledge graph data, and assign the node weights to the dimensions integrated knowledge graph data and product dimensions. The integrated knowledge graph data and enterprise-dimensional integrated knowledge graph data are vectorized for representation; that is, to represent the knowledge graph as a vector, it is necessary to map the entities and relationships in the knowledge graph to spatial vectors through different mapping matrices for representation. Taking the triple <customer A, holding company, number of holding companies> as an example, the head entity "customer A" is trained by the word2vec model to obtain the vector h, the relationship/attribute "shareholding company" vector is r, and the tail entity " The “number of companies holding shares” vector is t, and the entity and relationship vectors can be mapped to space vectors h _T =M _rh ×h and t _T =M _rh ×t through the mapping function M, respectively.

The MainfoldE algorithm is used to perform operations on the vectorized customer dimension comprehensive knowledge graph data, product dimension comprehensive knowledge graph data and enterprise dimension comprehensive knowledge graph data. The MainfoldE algorithm is a mature algorithm in the field. It maps the relationship r to the process body, that is, h+r is a hypersphere with a center radius of r, rather than an accurate point approximating h+r, which is more conducive to processing knowledge graphs with complex relationships.

The risk monitoring method provided by the embodiments of the present invention is more conducive to processing knowledge graphs with complex relationships.

Further, the risk monitoring method also includes:

Input the fusion data of the vectorized representation after operation processing into the preset risk monitoring model, and continue to perform the subsequent steps. That is, in this step, the aforementioned fusion data is replaced by the fusion data inputted in vectorized representation after operation processing, and other descriptions will not be repeated.

It can be understood that, the fusion data represented by the vectorized representation after the operation processing is used as the input of the model, which can improve the operation efficiency of the preset risk monitoring model.

The risk monitoring method provided by the embodiment of the present invention can further improve the operation efficiency of the preset risk monitoring model.

It should be noted that the risk monitoring method provided by the embodiments of the present invention can be used in the financial field, and can also be used in any technical field except the financial field, and the application field of the risk monitoring method is not limited in the embodiments of the present invention.

FIG. 2 is a schematic structural diagram of a risk monitoring device provided by an embodiment of the present invention. As shown in FIG. 2 , the risk monitoring device provided by an embodiment of the present invention includes an input unit 201 and a monitoring unit 202, wherein:

The input unit 201 is used for inputting fusion data into a preset risk monitoring model; the preset risk monitoring model is obtained by training a neural network model according to the fusion sample data; the fusion data is fused with customer-dimension integrated knowledge map data and product-dimension synthesis Knowledge graph data and enterprise-dimension integrated knowledge-graph data; the customer-dimension integrated knowledge-graph data, the product-dimension integrated knowledge-graph data, and the enterprise-dimension integrated knowledge graph data include respective information graph data corresponding to each information graph. The risk monitoring map data corresponding to the data; the monitoring unit 202 is configured to use the output result of the preset risk monitoring model as the risk monitoring result.

Specifically, the input unit 201 in the device is used to input fusion data into a preset risk monitoring model; the preset risk monitoring model is obtained by training a neural network model according to the fusion sample data; the fusion data is fused with comprehensive knowledge of customer dimensions Graph data, product-dimensional comprehensive knowledge graph data, and enterprise-dimensional comprehensive knowledge graph data; the customer-dimensional comprehensive knowledge graph data, the product-dimensional comprehensive knowledge graph data, and the enterprise-dimensional comprehensive knowledge graph data include their corresponding information graph data , and the risk monitoring map data corresponding to each information map data; the monitoring unit 202 is configured to use the output result of the preset risk monitoring model as the risk monitoring result.

The risk monitoring device provided by the embodiment of the present invention inputs fusion data into a preset risk monitoring model, and uses the output result of the preset risk monitoring model as the risk monitoring result, which can not only reduce labor costs, improve efficiency, but also overcome the data dimension There are limited shortcomings, and comprehensive risk warning is carried out according to business characteristics.

The embodiments of the risk monitoring apparatus provided in the embodiments of the present invention may be specifically used to execute the processing procedures of the foregoing method embodiments, and the functions thereof will not be repeated here, and reference may be made to the detailed descriptions of the foregoing method embodiments.

FIG. 3 is a schematic diagram of the physical structure of an electronic device provided by an embodiment of the present invention. As shown in FIG. 3 , the electronic device includes: a processor (processor) 301, a memory (memory) 302, and a bus 303;

The processor 301 and the memory 302 communicate with each other through the bus 303;

The processor 301 is configured to call program instructions in the memory 302 to execute the methods provided by the above method embodiments, for example, including:

Input fusion data into a preset risk monitoring model; the preset risk monitoring model is obtained by training a neural network model according to the fusion sample data; the fusion data is fused with customer-dimensional comprehensive knowledge map data, product-dimensional comprehensive knowledge map data, and enterprise-level comprehensive knowledge map data. Dimensional comprehensive knowledge graph data; the customer-dimensional comprehensive knowledge graph data, the product-dimensional comprehensive knowledge graph data, and the enterprise-dimensional comprehensive knowledge graph data include respective corresponding information graph data, and risk monitoring corresponding to each information graph data map data;

The output result of the preset risk monitoring model is used as the risk monitoring result.

This embodiment discloses a computer program product, the computer program product includes a computer program stored on a non-transitory computer-readable storage medium, the computer program includes program instructions, and when the program instructions are executed by a computer, the computer program The methods provided by the above method embodiments can be performed, for example, including:

Input fusion data into a preset risk monitoring model; the preset risk monitoring model is obtained by training a neural network model according to the fusion sample data; the fusion data is fused with customer-dimensional comprehensive knowledge map data, product-dimensional comprehensive knowledge map data, and enterprise-level comprehensive knowledge map data. Dimensional comprehensive knowledge graph data; the customer-dimensional comprehensive knowledge graph data, the product-dimensional comprehensive knowledge graph data, and the enterprise-dimensional comprehensive knowledge graph data include respective corresponding information graph data, and risk monitoring corresponding to each information graph data map data;

The output result of the preset risk monitoring model is used as the risk monitoring result.

This embodiment provides a computer-readable storage medium, where the computer-readable storage medium stores a computer program, and the computer program enables the computer to execute the methods provided by the foregoing method embodiments, for example, including:

Input fusion data into a preset risk monitoring model; the preset risk monitoring model is obtained by training a neural network model according to the fusion sample data; the fusion data is fused with customer-dimensional comprehensive knowledge map data, product-dimensional comprehensive knowledge map data, and enterprise-level comprehensive knowledge map data. Dimensional comprehensive knowledge graph data; the customer-dimensional comprehensive knowledge graph data, the product-dimensional comprehensive knowledge graph data, and the enterprise-dimensional comprehensive knowledge graph data include respective corresponding information graph data, and risk monitoring corresponding to each information graph data map data;

The output result of the preset risk monitoring model is used as the risk monitoring result.

As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.

The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block in the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to the processor of a general purpose computer, special purpose computer, embedded processor or other programmable data processing device to produce a machine such that the instructions executed by the processor of the computer or other programmable data processing device produce Means for implementing the functions specified in a flow or flow of a flowchart and/or a block or blocks of a block diagram.

These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory result in an article of manufacture comprising instruction means, the instructions The apparatus implements the functions specified in the flow or flow of the flowcharts and/or the block or blocks of the block diagrams.

These computer program instructions can also be loaded on a computer or other programmable data processing device to cause a series of operational steps to be performed on the computer or other programmable device to produce a computer-implemented process such that The instructions provide steps for implementing the functions specified in the flow or blocks of the flowcharts and/or the block or blocks of the block diagrams.

In the description of this specification, reference to the terms "one embodiment", "one specific embodiment", "some embodiments", "for example", "example", "specific example", or "some examples", etc. Indicates that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

The specific embodiments described above further describe the purpose, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above-mentioned specific embodiments are only specific embodiments of the present invention, and are not intended to limit the scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims (11)

1. a risk monitoring method, is characterized in that, comprises: Input fusion data into a preset risk monitoring model; the preset risk monitoring model is obtained by training a neural network model according to the fusion sample data; the fusion data is fused with customer-dimensional comprehensive knowledge map data, product-dimensional comprehensive knowledge map data, and enterprise-level comprehensive knowledge map data. Dimensional comprehensive knowledge graph data; the customer-dimensional comprehensive knowledge graph data, the product-dimensional comprehensive knowledge graph data, and the enterprise-dimensional comprehensive knowledge graph data include respective corresponding information graph data, and risk monitoring corresponding to each information graph data map data; The output result of the preset risk monitoring model is used as the risk monitoring result. 2 . The risk monitoring method according to claim 1 , wherein the neural network model is obtained by combining BRNN and LSTM. 3 . 3. The risk monitoring method according to claim 1, wherein the risk monitoring result corresponds to the customer dimension, the product dimension and the enterprise dimension respectively; After the step of outputting the result as the risk monitoring result, the risk monitoring method further includes: If at least one of the risk monitoring result in the customer dimension, the risk monitoring result in the product dimension, and the risk monitoring result in the enterprise dimension is at risk, a risk warning message corresponding to the at least one risky risk monitoring result is generated. 4. The risk monitoring method according to claim 1, wherein obtaining the fusion data comprises: Build customer information map data, product information map data and enterprise information map data respectively; According to the risk monitoring thresholds and corresponding data items of the data items corresponding to the customer information map data, the product information map data and the enterprise information map data, respectively construct the customer risk monitoring map data, the product risk monitoring map data and the corresponding data items. Enterprise risk monitoring map data; respectively fuse the customer information map data and the customer risk monitoring map data, the product information map data and the product risk monitoring map data, and the enterprise information map data and the enterprise risk monitoring map data, The customer-dimension integrated knowledge map data, the product-dimension integrated knowledge map data, and the enterprise-dimension integrated knowledge map data are obtained. 5. The risk monitoring method according to claim 4, wherein after the step of acquiring the fusion data, the risk monitoring method further comprises: The degree centrality calculation method is used to calculate the node weights in the customer-dimension integrated knowledge map data, the product-dimension integrated knowledge map data, and the enterprise-dimension integrated knowledge map data. 6. The risk monitoring method according to claim 5, characterized in that, in the utilization centrality calculation method, the integrated knowledge graph data of the dimension, the integrated knowledge graph data of the product dimension and the integrated knowledge graph of the enterprise dimension are calculated in the utilization centrality calculation method. After the step of node weights in the knowledge graph data, the risk monitoring method further includes: Assign the node weights to entities in the dimensional comprehensive knowledge graph data, the product dimensional comprehensive knowledge graph data, and the enterprise dimensional comprehensive knowledge graph data, and assign the node weights to the dimensions of the comprehensive knowledge graph data and product dimensions. Comprehensive knowledge graph data and enterprise dimension comprehensive knowledge graph data for vectorized representation; The MainfoldE algorithm is used to perform operations on the vectorized customer dimension comprehensive knowledge graph data, product dimension comprehensive knowledge graph data and enterprise dimension comprehensive knowledge graph data. 7. The risk monitoring method according to claim 6, wherein the risk monitoring method further comprises: Input the fusion data of the vectorized representation after operation processing into the preset risk monitoring model, and continue to perform the subsequent steps. 8. The risk monitoring method according to claim 4, wherein obtaining the enterprise risk monitoring map data, comprising: Obtain enterprise information, identify the text content of the enterprise information, and match the text content with the words in the preset positive public opinion database and the preset negative public opinion database respectively; Determine the type of text content according to the matching result; the type of text content includes positive content or negative content; Traverse all the text content, and obtain the number of text content corresponding to the positive content or the negative content according to the type of the text content of each text content; If the quantity of text content corresponding to the positive content is less than the quantity of text content corresponding to the negative content, determine that the enterprise information is negative enterprise information; Obtain enterprise interaction information corresponding to the negative enterprise information, and if the statistical value of at least one interaction index data in the enterprise interaction information is greater than a preset statistical data threshold, use the negative enterprise information as the enterprise risk monitoring Graph data. 9. A risk monitoring device, characterized in that, comprising: The input unit is used to input the fusion data into the preset risk monitoring model; the preset risk monitoring model is obtained by training the neural network model according to the fusion sample data; the fusion data is fused with the customer-dimensional integrated knowledge map data and the product-dimensional integrated Knowledge graph data and enterprise-dimension integrated knowledge-graph data; the customer-dimension integrated knowledge-graph data, the product-dimension integrated knowledge-graph data, and the enterprise-dimension integrated knowledge graph data include respective information graph data corresponding to each information graph. Risk monitoring map data corresponding to the data; The monitoring unit is configured to use the output result of the preset risk monitoring model as the risk monitoring result. 10. An electronic device comprising a memory, a processor and a computer program stored on the memory and running on the processor, wherein the processor implements any one of claims 1 to 8 when the processor executes the computer program. A step of the method. 11. A computer-readable storage medium on which a computer program is stored, characterized in that, when the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.

Images