Behavior risk assessment method and device, computer equipment and storage medium

Abstract

The embodiment of the invention belongs to the field of artificial intelligence and intelligent risk assessment, and relates to a behavior risk assessment method, which comprises the steps of acquiring user behavior data and user browsing data, performing content violation judgment on the user browsing data to obtain a violation browsing area, superimposing and mapping the user behavior data to the user browsing data, and obtaining a violation browsing area. And obtaining a user attention distribution map, and carrying out association mining analysis on the user attention distribution map and the violation browsing area to obtain a violation risk report. The invention further provides a behavior risk assessment device, computer equipment and a storage medium. According to the method and the device, the violation information is accurately monitored and identified by combining the user behavior data and the browsing content, and the risk report is generated, so that the supervision efficiency is improved, the manual auditing requirement is reduced, and a financial institution can accurately monitor and manage a high-risk area.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence and intelligent risk assessment technology, and in particular to a behavior risk assessment method, apparatus, computer equipment and storage medium. Background Art

[0002] In the current financial market, many financial institutions engage in improper behavior in the promotion of loan products, such as hiding high interest rates and unclear rate descriptions, which infringes on consumer rights. In addition, due to the large amount of information in advertising and its rapid updates, traditional regulatory means are difficult to effectively identify and deal with these illegal promotions in real time, resulting in illegal information being easily spread among consumers.

[0003] Traditional solutions typically involve manual review and basic automated tools, such as using simple keyword searches to identify potential violations, with regulators and financial firms typically deploying teams to regularly review marketing materials to ensure they comply with regulations.

[0004] However, manual review is inefficient, costly, and difficult to cover all marketing materials, and it is easy to miss illegal content. Although keyword-based automation tools have improved the detection speed to a certain extent, this method lacks contextual understanding and will result in high false positive rates and missed detections. In addition, the above methods are difficult to assess users' actual attention to illegal content and the influence of that content. Summary of the invention

[0005] The purpose of the embodiments of the present application is to propose a behavior risk assessment method, apparatus, computer equipment and storage medium to solve the problems of low efficiency of traditional review methods and difficulty in assessing users' actual attention to illegal content and the impact of content.

[0006] In order to solve the above technical problems, the present application embodiment provides a behavior risk assessment method, which adopts the following technical solution:

[0007] Obtain user behavior data and user browsing data;

[0008] Conducting content violation judgment on the user's browsing data to obtain a violation browsing area;

[0009] Overlaying and mapping the user behavior data onto the user browsing data to obtain a user attention distribution map;

[0010] The user attention distribution map and the illegal browsing area are subjected to correlation mining analysis to obtain a violation risk report.

[0011] Furthermore, the step of determining whether the user's browsing data contains any content violations and obtaining the illegal browsing area specifically includes:

[0012] Divide the user browsing data into regions by using a page structure analysis algorithm to obtain a target browsing region;

[0013] The target browsing area is identified and processed by optical character recognition technology to obtain the illegal browsing area.

[0014] Furthermore, the step of identifying the target browsing area by optical character recognition technology to obtain the illegal browsing area specifically includes:

[0015] Performing text recognition processing on the target browsing area by optical character recognition technology to obtain target text data;

[0016] Perform keyword matching processing on the target text data to obtain a target illegal text;

[0017] Perform font detection processing on the target illegal text to determine the illegal browsing area.

[0018] Furthermore, the user behavior data includes user eye movement data and mouse movement data, and the step of superimposing and mapping the user behavior data onto the user browsing data to obtain a user attention distribution map specifically includes:

[0019] The user's eye movement data and the mouse movement data are superimposed on the user's browsing data by a heat map generation algorithm to obtain a target area distribution map;

[0020] The target area distribution map is associated with the illegal browsing area through a spatial matching algorithm to obtain the user attention distribution map.

[0021] Furthermore, the user attention distribution map includes a high attention area and a low attention area, and the step of performing correlation mining analysis on the user attention distribution map and the illegal browsing area to obtain a violation risk report specifically includes:

[0022] Analyzing the overlap between the high-attention area and the illegal browsing area through an association rule mining algorithm to obtain a first visual attraction index;

[0023] Analyzing the overlap between the low-attention area and the illegal browsing area through an association rule mining algorithm to obtain a second visual attraction index;

[0024] Performing weighted processing on the first visual attraction index and the second visual attraction index to obtain a target visual attraction index;

[0025] The illegal browsing area is matched with the target visual attraction index to obtain the illegal risk report.

[0026] Furthermore, after the step of matching the illegal browsing area by the target visual attraction index to obtain the illegal risk report, the method further includes:

[0027] Performing path mining on the illegal browsing area to obtain a risk propagation path;

[0028] Integrate the risk propagation path into the violation risk report, and update the violation risk report.

[0029] Furthermore, the step of performing path mining on the illegal browsing area to obtain the risk propagation path specifically includes:

[0030] Constructing a relationship network for the user behavior data and the user browsing data to obtain a user behavior relationship network;

[0031] The risk propagation path is obtained by performing path mining processing on the user behavior relationship network through a graph mining algorithm.

[0032] In order to solve the above technical problems, the embodiment of the present application also provides a behavior risk assessment device, which adopts the following technical solution:

[0033] The acquisition module is used to obtain user behavior data and user browsing data;

[0034] A judgment module, used to judge the violation of content on the user's browsing data and obtain a violation browsing area;

[0035] A mapping module, used to map the user behavior data onto the user browsing data to obtain a user attention distribution map;

[0036] The analysis module is used to perform correlation mining analysis on the user attention distribution map and the illegal browsing area to obtain a violation risk report.

[0037] In order to solve the above technical problems, the embodiment of the present application further provides a computer device, which adopts the following technical solution:

[0038] A computer device comprises a memory and a processor, wherein the memory stores computer-readable instructions, and the processor implements the steps of the behavior risk assessment method when executing the computer-readable instructions.

[0039] In order to solve the above technical problems, the embodiment of the present application further provides a computer-readable storage medium, which adopts the following technical solution:

[0040] A computer-readable storage medium stores computer-readable instructions, which implement the steps of the behavior risk assessment method when executed by a processor.

[0041] Compared with the prior art, the embodiments of the present application have the following beneficial effects:

[0042] The embodiment of the present application obtains user behavior data and user browsing data, makes content violation judgments on the user browsing data, obtains illegal browsing areas, overlays and maps the user behavior data to the user browsing data, obtains a user attention distribution map, performs correlation mining analysis on the user attention distribution map and the illegal browsing areas, obtains a violation risk report, automatically identifies and marks illegal content areas through comprehensive user behavior data and browsing content analysis, and generates a detailed violation risk report by associating the areas that users actually pay attention to, thereby improving the efficiency and accuracy of monitoring illegal propaganda, and accurately assessing the influence of illegal information on users, so that financial institutions can implement more targeted regulatory measures, effectively protect consumer rights and interests, and prevent the spread of improper marketing practices. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the scheme in the present application, a brief introduction is given below to the drawings required for use in the description of the embodiments of the present application. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0044] Figure 1 is an exemplary system architecture diagram to which the present application may be applied;

[0045] Figure 2 A flowchart of an embodiment of a behavior risk assessment method according to the present application;

[0046] Figure 3 is a structural schematic diagram of an embodiment of a behavior risk assessment device according to the present application;

[0047] Figure 4 It is a structural diagram of an embodiment of a computer device according to the present application. DETAILED DESCRIPTION

[0048] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by technicians in the technical field of the present application; the terms used in the specification of the application herein are only for the purpose of describing specific embodiments and are not intended to limit the present application; the terms "including" and "having" and any variations thereof in the specification and claims of the present application and the above-mentioned drawings are intended to cover non-exclusive inclusions. The terms "first", "second", etc. in the specification and claims of the present application or the above-mentioned drawings are used to distinguish different objects, not to describe a specific order.

[0049] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0050] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings.

[0051] like Figure 1 As shown, the system architecture 100 may include a terminal device 101, a network 102 and a server 103. The terminal device 101 may be a laptop 1011, a tablet computer 1012 or a mobile phone 1013. The network 102 is used to provide a medium for a communication link between the terminal device 101 and the server 103. The network 102 may include various connection types, such as wired, wireless communication links or optical fiber cables.

[0052] The user can use the terminal device 101 to interact with the server 103 through the network 102 to receive or send messages, etc. Various communication client applications can be installed on the terminal device 101, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social platform software, etc.

[0053] The terminal device 101 can be any electronic device with a display screen and supporting web browsing. In addition to a laptop computer 1011, a tablet computer 1012 or a mobile phone 1013, the terminal device 101 can also be an e-book reader, an MP3 player (Moving Picture Experts Group Audio Layer III), an MP4 (Moving Picture Experts Group Audio Layer IV), a laptop computer, a desktop computer, etc.

[0054] The server 103 may be a server that provides various services, such as a background server that provides support for web pages displayed on the terminal device 101 .

[0055] It should be noted that the behavior risk assessment method provided in the embodiment of the present application is generally executed by a server / terminal device, and accordingly, the behavior risk assessment device is generally arranged in the server / terminal device.

[0056] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0057] Continue to refer Figure 2 , showing a flow chart of an embodiment of a method for behavioral risk assessment according to the present application. The behavioral risk assessment method comprises the following steps:

[0058] Step S201, obtaining user behavior data and user browsing data.

[0059] In this embodiment, the above-mentioned behavior risk assessment method can be deployed in a behavior risk assessment platform. The above-mentioned behavior risk assessment platform can be constructed by a server or a server cluster. The above-mentioned server or server cluster can be any electronic device with image recognition, image processing, text recognition, text processing, data transmission, data storage and other functions. The behavior risk assessment method runs on the electronic device (for example, Figure 1 The server / terminal device shown in the figure) can receive the user behavior data and the user browsing data through a wired connection or a wireless connection. It should be noted that the above wireless connection method may include but is not limited to 3G / 4G / 5G connection, WiFi connection, Bluetooth connection, WiMAX connection, Zigbee connection, UWB (ultra wideband) connection, and other wireless connection methods currently known or to be developed in the future.

[0060] In this embodiment, the above-mentioned user behavior data may include but is not limited to eye tracking information, mouse activity data, keyboard input and scrolling behavior of the user when browsing the page. Specifically, the above-mentioned user behavior data can reveal the user's interaction method, focus and behavior pattern with the specific page content, and is used to evaluate the user's interest and reaction to the displayed content.

[0061] In this embodiment, the user browsing data may be the content data of the pages visited and browsed by the user on the network platform, including but not limited to text, images, videos, links and the structural layout information of the pages. Specifically, the user browsing data may be used to analyze the content attributes of the pages, including the layout and prominence of the information and whether it contains potential illegal or non-compliant information.

[0062] Step S202, determining whether the user's browsing data has any content violations, and obtaining a violation browsing area.

[0063] In this embodiment, the illegal browsing area may be a portion of the content that does not comply with regulations or standards in the page browsed by the user. Specifically, the illegal browsing area may include but is not limited to hidden or ambiguous rate information, misleading advertising slogans, unfair marketing methods, and any content that may violate relevant laws, industry norms or advertising standards.

[0064] Specifically, regarding the specific implementation process of judging the content violation of the user browsing data and obtaining the illegal browsing area, this application will provide further details in subsequent specific embodiments, and will not be elaborated in detail here.

[0065] Step S203, overlay and map the user behavior data onto the user browsing data to obtain a user attention distribution map.

[0066] In this embodiment, the user attention distribution map can be a spatial distribution of the user's line of sight and cursor's dwell time on the page, movement path, click frequency and other behaviors. The user attention distribution map can be used to identify the page parts that the user is most interested in and interacts with most frequently.

[0067] In this embodiment, the user attention distribution map may include but is not limited to hot spots, attention paths, and interaction-intensive areas. Specifically, the hot spots may be the part of the page where the user's eyes stay the longest or the mouse moves the most frequently, which is used to indicate the user's high attention to the area; the attention path may be the trajectory of the user's eyes or cursor movement, which is used to show the user's browsing order and mode from one page element to another; the interaction-intensive area may be the area on the page where mouse clicks, scrolling, or keyboard operations are concentrated.

[0068] Specifically, regarding the specific implementation process of mapping the user behavior data onto the user browsing data to obtain the user attention distribution map, this application will provide further detailed descriptions in subsequent specific embodiments, and will not be elaborated on here.

[0069] Step S204, performing correlation mining analysis on the user attention distribution map and the illegal browsing area to obtain a violation risk report.

[0070] In this embodiment, the above-mentioned violation risk report can be a graphic report used to evaluate and summarize the user's attention to and potential impact of illegal content. Specifically, the above-mentioned violation risk report can include but is not limited to the type of illegal content, location, user interaction intensity and possible impact on user behavior.

[0071] In one possible embodiment, the above-mentioned violation risk report can be used to provide accurate monitoring and assessment of violation content for financial regulators or related companies, point out which content may lead to the risk of misleading consumers or improper behavior, and recommend regulatory or corrective measures. Relevant institutions optimize their monitoring strategies and customer protection measures to ensure the transparency and fairness of advertising and information dissemination.

[0072] Specifically, regarding the specific implementation process of performing correlation mining analysis on the user attention distribution map and the illegal browsing area to obtain the violation risk report, this application will provide further detailed descriptions of this in subsequent specific embodiments and will not be elaborated in detail here.

[0073] The present application obtains user behavior data and user browsing data, makes content violation judgments on the user browsing data, obtains illegal browsing areas, overlays and maps the user behavior data to the user browsing data, obtains a user attention distribution map, performs correlation mining analysis on the user attention distribution map and the illegal browsing areas, obtains a violation risk report, automatically identifies and marks illegal content areas through comprehensive user behavior data and browsing content analysis, and generates a detailed violation risk report by associating the areas that users actually pay attention to, thereby improving the efficiency and accuracy of monitoring illegal propaganda and accurately evaluating the impact of illegal information on users, thereby enabling financial institutions to implement more targeted regulatory measures, effectively protect consumer rights and interests, and prevent the spread of improper marketing practices.

[0074] In some optional implementations, step S202 includes the following steps:

[0075] The user browsing data is divided into regions through the page structure analysis algorithm to obtain the target browsing area;

[0076] The target browsing area is identified and processed by optical character recognition technology to obtain the illegal browsing area.

[0077] In this embodiment, the target browsing area may include but is not limited to a text block area, an advertisement banner, a pop-up window, and a footer or header area in the user's browsing data. Specifically, the text block area may include but is not limited to financial information such as loan conditions, interest rate information, and fee descriptions; the advertisement banner may include but is not limited to an advertisement area for promoting financial products; the pop-up window may include but is not limited to a pop-up window with words such as special offers or promotional information; the footer or header area may be an area containing a legal disclaimer or other important compliance information.

[0078] In this embodiment, the page structure analysis algorithm may include, but is not limited to, web page content analysis algorithms such as DOM analysis, CSS box model analysis, and visual hierarchy analysis.

[0079] In this embodiment, the above optical character recognition technology can be implemented through third-party OCR service plug-ins such as Tesseract OCR, Google Vision API and Microsoft OCR.

[0080] In a possible embodiment, DOM analysis is used to perform structural analysis (i.e., the above-mentioned area division) on web page data browsed by users (i.e., the above-mentioned user browsing data). The above-mentioned DOM analysis can identify different HTML elements and their attributes, such as text blocks, pictures, links and their container elements, by deconstructing the document object model of the web page, so as to accurately locate the key target browsing area. For example, the text area contains financial information such as "loan conditions" and "interest rate information". Next, Tesseract OCR technology is applied to perform optical character recognition on these identified text blocks, and the text content in the area is extracted by Tesseract OCR, and further content review is performed to detect whether there is any illegal information, such as misleading statements or hidden rate information (i.e., the above-mentioned illegal browsing area).

[0081] This application provides accurate data support by automatically identifying and marking illegal browsing areas, and improves processing efficiency by combining DOM analysis and OCR technology, significantly enhancing the accuracy of identifying illegal content.

[0082] Specifically, regarding the specific implementation process of identifying and processing the user browsing area through optical character recognition technology to obtain the illegal browsing area, this application will provide further details in subsequent specific embodiments, and will not be elaborated on here.

[0083] In some optional implementations, the above step of "identifying the target browsing area by optical character recognition technology to obtain the illegal browsing area" includes the following steps:

[0084] Perform text recognition processing on the target browsing area through optical character recognition technology to obtain target text data;

[0085] Perform keyword matching processing on the target text data to obtain the target illegal text;

[0086] Perform font detection on the target illegal text to determine the illegal browsing area.

[0087] In this embodiment, the target violation text may be text data in the webpage content that is identified as having potential violation characteristics. Specifically, the target violation text may include but is not limited to hidden rate information, misleading financial statements, statements beyond legal and regulatory requirements, and seductive or fraudulent language.

[0088] In this embodiment, the above keyword matching process can be implemented by methods such as regular expression matching, keyword extraction and semantic analysis.

[0089] In this embodiment, the above-mentioned font detection process can be implemented by image processing algorithms such as edge detection and color analysis.

[0090] In a possible embodiment, firstly, through optical character recognition technology such as Tesseract OCR or Google Vision API, a text extraction is performed on the selected target browsing area of ​​the webpage (i.e., the above-mentioned target browsing area) to obtain target text data. Next, regular expressions or predefined keyword libraries are used for text screening to identify target illegal texts containing specific keywords (e.g., "annual interest rate", "unsecured", etc.), and the above-mentioned target illegal texts can be texts containing annualized interest rate words. Finally, font detection processing is performed through an image processing library such as OpenCV to analyze the font size, color contrast and background color difference of these texts to verify whether the visibility and significance of the text meet regulatory requirements, for example, whether the font size of the text containing the annualized interest rate words is less than a threshold value, if it is less than the threshold value, it is considered that the interest rate disclosure is not significant, it is improper marketing, and the illegal browsing area is determined; whether the contrast between the font color and background color of the text containing the annualized interest rate is less than a threshold value, if it is less than the threshold value, it is considered that the interest rate disclosure is not significant, it is improper marketing, and the illegal browsing area is determined.

[0091] This application automatically identifies potentially illegal financial texts and evaluates the visual prominence of these texts to ensure that illegal information is not easily overlooked by users. It can significantly improve regulatory efficiency and reduce labor costs. At the same time, it can enhance compliance monitoring of financial advertising and information dissemination, effectively prevent consumers from being misled, and ensure market fairness and transparency.

[0092] In some optional implementations, step S203 includes the following steps:

[0093] The user's eye movement data and mouse movement data are superimposed on the user's browsing data through a heat map generation algorithm to obtain a target area distribution map;

[0094] The target area distribution map is associated with the illegal browsing area through a spatial matching algorithm to obtain the user attention distribution map.

[0095] In this embodiment, the user behavior data may include the user eye movement data and the mouse movement data. Specifically, the user eye movement data may include but is not limited to the user's gaze point, line of sight movement path, blinking frequency, gaze duration, and pupil size change, and the mouse movement data may include but is not limited to the mouse movement speed, movement path, click events, hovering time, and mouse scrolling behavior. The user behavior data may be used to represent the user's attention distribution and depth of attention on the page, and the mouse movement data may represent the user's interaction method and interest points with the page elements.

[0096] In this embodiment, the implementation of the heat map generation algorithm may include but is not limited to density-based heat maps, time-weighted heat maps, and event-driven heat maps. Specifically, the density-based heat map can generate a heat map based on the spatial density of user behavior data, such as the degree of aggregation of mouse movement or eye movement, to display the hot spots of user behavior on the web page; the time-weighted heat map can use the user's stay time in a certain area as a weight to generate a heat map reflecting the user's attention level; the event-driven heat map can generate a heat map based on user interaction events such as clicks and scrolls, highlighting the areas where users interact most frequently.

[0097] In this embodiment, the spatial matching algorithm may include but is not limited to the Euclidean distance matching algorithm, the cosine similarity matching algorithm and the cross-correlation matching algorithm. Specifically, the Euclidean distance matching algorithm may use the Euclidean distance to calculate the distance between the point in the user attention distribution map and the illegal browsing area to determine the correlation; the cosine similarity matching algorithm may evaluate the spatial correlation by calculating the cosine similarity between the user attention distribution and the illegal content distribution vector; the cross-correlation matching algorithm may use the cross-correlation technology to analyze the correlation between the two regional distribution maps, and is suitable for processing data sets with time series or spatial changes.

[0098] In a possible embodiment, a heat map generation algorithm is used, specifically using a time-weighted or density-based method, to comprehensively analyze the user's eye movement data and mouse movement data and superimpose them on the user's browsing data. By counting the gaze dwell time and mouse activity frequency on each web page element, the above gaze dwell time and mouse activity frequency are converted into hotspot intensity values, thereby generating a target area distribution map reflecting the density of user behavior. A spatial matching algorithm, such as cosine similarity matching or cross-correlation matching, is applied to compare and associate the target area distribution map with the identified illegal browsing areas. By calculating the spatial correlation or vector similarity of the two area distributions, the correlation between them is established, thereby obtaining the final user attention distribution map.

[0099] This application reveals the degree of user attention to specific illegal content through the application of heat map generation algorithm and spatial matching algorithm, which can be used to help regulators and website operators identify and improve potential risk points, enhance the compliance of web content and the security of user experience.

[0100] In some optional implementations, step S204 includes the following steps:

[0101] The first visual attraction index is obtained by analyzing the overlap between the high-attention area and the illegal browsing area through the association rule mining algorithm;

[0102] The second visual attraction index is obtained by analyzing the overlap between low-attention areas and illegal browsing areas through association rule mining algorithm;

[0103] The first visual attraction index and the second visual attraction index are weighted to obtain a target visual attraction index;

[0104] The illegal browsing areas are matched by the target visual attraction index to obtain a violation risk report.

[0105] In this embodiment, the above association rule mining algorithm may include but is not limited to the Apriori algorithm and the FP-growth algorithm.

[0106] In this embodiment, the user attention distribution diagram includes high attention areas and low attention areas. Specifically, the high attention areas can be parts of the web page where users keep their eyes for a long time, click or hover their mouse frequently, such as important news headlines, eye-catching images or videos, key interactive elements (such as forms or search bars), and the low attention areas can be areas where users interact and stay less, such as the edge of the page, advertising borders, legal notices or footer links.

[0107] In this embodiment, the first visual attraction index can be a quantitative indicator of the degree of overlap between the high-attention area and the illegal browsing area, which is calculated based on the user's eye movement dwelling time or click frequency on the illegal content, indicating the attractiveness of the illegal content to the user; the second visual attraction index can be a quantitative indicator of the degree of overlap between the low-attention area and the illegal browsing area, reflecting the potential influence of the illegal content in the area where the user pays less attention.

[0108] In a possible embodiment, the user's eye movement data and mouse movement data are collected, and these data are matched with the position of the web page element browsed by the user (that is, the above-mentioned user browsing data), the data is formatted, and a transaction record containing user behavior (such as eye movement stop point, mouse click and hover position) and web page elements (such as text blocks, pictures, links) is created, and the FP-growth algorithm is applied to generate frequent item sets from the transaction records. The above-mentioned frequent item sets represent the combination related to the user behavior and the illegal browsing area. The association rules are extracted from the frequent item sets, and the rules for associating the high-attention area or the low-attention area with the illegal browsing area are constructed to obtain the first visual attraction index and the second visual attraction index. The formula of the above-mentioned first visual attraction index is as follows:

[0109]

[0110] Among them, the above I1 is the first visual attraction index, which is used to measure the overall influence of the degree of overlap between the high-attention area and the illegal browsing area. rules1 is the confidence of each association rule, which is a conditional probability, indicating the probability that the subsequent item (illegal browsing area) will also appear when the antecedent item (high attention area) appears. rules1 is the support of each association rule, which is a frequency value indicating the proportion of itemsets involved in a specific rule (i.e., the combination of high-attention areas and illegal browsing areas) in all transactions. The higher the support, the more frequently this combination appears in the data, indicating that it is generally more important. The above rules 1 is an association rule generated by the association rule mining algorithm, which can be the relationship between user behavior and web page content. For example, if a user often stays on a specific type of article or product page for a long time, a rule may be generated to show the user's interest in this type of content.

[0111] The formula for the second visual attraction index is as follows:

[0112]

[0113] The above I2 is the second visual attraction index, which is used to measure the overall influence of the overlap between the low-attention area and the illegal browsing area.rules2 is the confidence of each association rule, which is a conditional probability, indicating the probability that the consequent (illegal browsing area) will also appear when the antecedent (low attention area) appears. rules2 is the support of each association rule, which is a frequency value indicating the proportion of the item set involved in a specific rule (i.e., the combination of low-attention areas and illegal browsing areas) in all transactions. The higher the support, the more frequently this combination appears in the data, indicating that it is generally more important. The above rules2 are association rules generated by the association rule mining algorithm, which can be the relationship between user behavior and web page content. For example, if users often stay on a specific type of article or product page for a long time, a rule may be generated to show the user's interest in this type of content.

[0114] The first visual attraction index and the second visual attraction index are weighted to form a target visual attraction index, which reflects the impact of illegal content on the overall user experience. Finally, the target visual attraction index is used to accurately match illegal browsing areas and generate a detailed violation risk report, which indicates the specific location of illegal content and its potential impact on user behavior, providing a scientific basis for subsequent regulatory measures.

[0115] This application provides website managers with different levels of user behavior insights through the calculation of the first visual attraction index and the second visual attraction index. The target visual attraction index obtained by weighting can more comprehensively evaluate the overall risk level of illegal content, and can accurately quantify the actual attention of users to illegal content and the potential influence of the content, and summarize it into a violation risk report, which reflects the overall importance and urgency of illegal content in user experience, helps relevant departments to adjust strategies in a timely manner, take appropriate regulatory measures or content improvements to enhance the compliance and security of the website, prevent misleading users, protect user rights, improve the accuracy and response speed of illegal content monitoring, and optimize content supervision and user experience management.

[0116] In some optional implementations, after step S204, the following steps are included:

[0117] Perform path mining on illegal browsing areas to obtain risk propagation paths;

[0118] Integrate risk propagation paths into violation risk reports and update violation risk reports.

[0119] In this embodiment, the risk propagation path may include but is not limited to sharing behaviors between users, navigation paths within a page, and paths generated by a content recommendation engine. Specifically, the sharing behaviors between users may include, for example, users sharing links to illegal content through social media, email, or instant messaging; the navigation paths within a page may be the navigation sequence of users from accessing pages containing illegal content to accessing other pages, including paths that include clicking on advertisements, tracking links, or other guiding elements; the paths generated by the content recommendation engine may be paths that automatically recommend illegal content based on user behaviors and preferences.

[0120] In this embodiment, the path mining process may include but is not limited to graph mining technologies such as Neo4j or Gephi and sequence pattern mining algorithms such as PrefixSpan or SPMF.

[0121] In a possible embodiment, it first involves constructing a relationship graph between user behavior and web page content, using a graph database such as Neo4j to store the sequence data of users accessing illegal content and their interactive links with other users and content, and using graph mining techniques such as community detection algorithms and node influence analysis to identify and analyze user nodes and connection paths that play a core role in the spread of illegal content to determine the above-mentioned risk propagation path (the risk propagation path can be formed by tracking how users push content to other users in their network through clicks, sharing or recommendation mechanisms). Subsequently, the above-mentioned risk propagation path is integrated into the violation risk report, and data visualization tools such as Tableau or PowerBI are used to present the path diagram and key nodes, and the violation risk report is updated to include propagation dynamics analysis and potential impact assessment.

[0122] This application conducts path mining on illegal browsing areas and integrates risk propagation paths into violation risk reports, thereby achieving in-depth analysis and visual display of the propagation mechanism and impact range of illegal content. It can accurately identify the propagation path of illegal content, identify key propagation nodes and channels, and evaluate the influence of illegal content on user networks. In addition, the updated violation risk report provides data-driven decision support for regulators and website administrators, allowing for the implementation of more effective monitoring strategies and intervention measures based on actual data, thereby effectively preventing the further spread of illegal content and enhancing the health and security of the network environment.

[0123] Specifically, regarding the specific implementation process of performing path mining on illegal browsing areas to obtain risk propagation paths, this application will provide further detailed descriptions of this in subsequent specific embodiments, and will not be elaborated on here.

[0124] In some optional implementations, the step of “performing path mining on the illegal browsing area to obtain the risk propagation path” includes the following steps:

[0125] Construct a relationship network for user behavior data and user browsing data to obtain a user behavior relationship network;

[0126] The user behavior relationship network is processed through graph mining algorithm to obtain the risk propagation path.

[0127] In this embodiment, the user behavior relationship network can be a network structure composed of multiple nodes and edges, wherein the nodes represent users, web page elements (such as text blocks, pictures, links, etc.), pages and other related entities, and the edges represent the interactive relationship between users and these elements, such as clicks, eye tracking, page visit sequences, etc.

[0128] In this embodiment, the above-mentioned graph mining algorithms may include but are not limited to community detection algorithms (such as Louvain and Girvan-Newman), shortest path algorithms (such as Dijkstra and Floyd-Warshall) and node importance scores (such as PageRank and Betweenness Centrality).

[0129] This application builds a relationship network based on user behavior data and user browsing data, and uses graph mining algorithms to perform path mining processing, which can achieve an in-depth understanding of user interaction and content dissemination patterns, revealing how users interact with website content and how these interactions lead to specific behaviors or decisions.

[0130] It is understandable that in the specific implementation of this application, related data such as the above-mentioned violation risk reports, user behavior data, and user browsing data are involved. When the embodiments in this application are applied to specific products or technologies, user permission or consent is required, and the collection, use and processing of relevant data need to comply with relevant laws, regulations and standards of relevant countries and regions.

[0131] It should be emphasized that in order to further ensure the privacy and security of the above-mentioned violation risk reports, user behavior data, user browsing data and other information, the above-mentioned violation risk reports, user behavior data, user browsing data and other information can also be stored in a blockchain node.

[0132] The blockchain referred to in this application is a new application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanism, encryption algorithm, etc. Blockchain is essentially a decentralized database, a string of data blocks generated by cryptographic methods. Each data block contains a batch of network transaction information, which is used to verify the validity of its information (anti-counterfeiting) and generate the next block. Blockchain can include the underlying blockchain platform, platform product service layer, and application service layer.

[0133] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Among them, artificial intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0134] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through computer-readable instructions, and the computer-readable instructions can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the aforementioned storage medium can be a non-volatile storage medium such as a disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0135] It should be understood that, although the steps in the flowchart of the accompanying drawings are displayed in sequence as indicated by the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.

[0136] Further references Figure 3 , as a response to the above Figure 2 The present application provides an embodiment of a behavior risk assessment device. Figure 2 Corresponding to the method embodiment shown, the device can be specifically applied to various electronic devices.

[0137] like Figure 3 As shown, the behavior risk assessment device 300 described in this embodiment includes: an acquisition module 301, a judgment module 302, a mapping module 303 and an analysis module 304. Among them:

[0138] Acquisition module 301, used to acquire user behavior data and user browsing data;

[0139] The judgment module 302 is used to judge whether the user's browsing data has any content violations, and obtain a violation browsing area;

[0140] A mapping module 303 is used to map the user behavior data onto the user browsing data to obtain a user attention distribution map;

[0141] The analysis module 304 is used to perform correlation mining analysis on the user attention distribution map and the illegal browsing area to obtain a violation risk report.

[0142] In a possible embodiment, the determining module 302 includes:

[0143] A division submodule, used for dividing the user browsing data into regions through a page structure analysis algorithm to obtain a target browsing region;

[0144] The identification submodule is used to identify the target browsing area through optical character recognition technology to obtain the illegal browsing area.

[0145] In a possible embodiment, the identification submodule includes:

[0146] A recognition unit, used for performing text recognition processing on the target browsing area through optical character recognition technology to obtain target text data;

[0147] A matching unit, used for performing keyword matching processing on the target text data to obtain a target illegal text;

[0148] The detection unit is used to perform font detection processing on the target illegal text to determine the illegal browsing area.

[0149] In a possible embodiment, the mapping module 303 includes:

[0150] A superposition submodule, used for superimposing the user eye movement data and the mouse movement data on the user browsing data through a heat map generation algorithm to obtain a target area distribution map;

[0151] The association submodule is used to associate the target area distribution map with the illegal browsing area through a space matching algorithm to obtain the user attention distribution map.

[0152] In a possible embodiment, the analysis module 304 includes:

[0153] A first analysis submodule, configured to analyze the overlap between the high-attention area and the illegal browsing area through an association rule mining algorithm to obtain a first visual attraction index;

[0154] A second analysis submodule is used to analyze the overlap between the low-attention area and the illegal browsing area through an association rule mining algorithm to obtain a second visual attraction index;

[0155] a processing submodule, configured to perform weighted processing on the first visual attraction index and the second visual attraction index to obtain a target visual attraction index;

[0156] The matching submodule is used to match the illegal browsing area with the target visual attraction index to obtain the illegal risk report.

[0157] In a possible embodiment, after the analysis module 304, the method further includes:

[0158] A mining module, used to perform path mining on the illegal browsing area to obtain a risk propagation path;

[0159] An integration module is used to integrate the risk propagation path into the violation risk report and update the violation risk report.

[0160] In a possible embodiment, the mining module includes:

[0161] A construction submodule, used to construct a relationship network for the user behavior data and the user browsing data to obtain a user behavior relationship network;

[0162] The mining submodule is used to perform path mining processing on the user behavior relationship network through a graph mining algorithm to obtain the risk propagation path.

[0163] In this embodiment, by obtaining user behavior data and user browsing data, the user browsing data is judged for content violations to obtain a violation browsing area, the user behavior data is superimposed and mapped to the user browsing data to obtain a user attention distribution map, and the user attention distribution map is associated with the violation browsing area for mining analysis to obtain a violation risk report. By integrating user behavior data and browsing content analysis, the violation content area is automatically identified and marked, and by associating the area of ​​actual user attention, a detailed violation risk report is generated, which improves the efficiency and accuracy of monitoring illegal propaganda and accurately evaluates the impact of illegal information on users, thereby enabling financial institutions to implement more targeted regulatory measures, effectively protect consumer rights and interests, and prevent the spread of improper marketing practices.

[0164] In this embodiment, the operations performed by the above units or modules correspond one-to-one to the steps of the behavior risk assessment method in the above implementation mode, which will not be repeated here.

[0165] To solve the above technical problems, the present application also provides a computer device. Figure 4 , Figure 4 This is a basic structural block diagram of the computer device in this embodiment.

[0166] The computer device 4 includes a memory 41, a processor 42, and a network interface 43 that are interconnected through a system bus. It should be noted that the figure only shows a computer device 4 with components 41-43, but it should be understood that it is not required to implement all the components shown, and more or fewer components can be implemented instead. Among them, those skilled in the art can understand that the computer device here is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application specific integrated circuits (Application Specific Integrated Circuit, ASIC), programmable gate arrays (Field-Programmable Gate Array, FPGA), digital processors (Digital Signal Processor, DSP), embedded devices, etc.

[0167] The computer device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The computer device may interact with a user through a keyboard, a mouse, a remote controller, a touch pad, or a voice control device.

[0168] The memory 41 includes at least one type of readable storage medium, and the readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (for example, SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 41 can be an internal storage unit of the computer device 4, such as a hard disk or memory of the computer device 4. In other embodiments, the memory 41 can also be an external storage device of the computer device 4, such as a plug-in hard disk equipped on the computer device 4, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (FlashCard), etc. Of course, the memory 41 can also include both the internal storage unit of the computer device 4 and its external storage device. In this embodiment, the memory 41 is generally used to store the operating system and various application software installed on the computer device 4, such as computer-readable instructions of the behavior risk assessment method, etc. In addition, the memory 41 can also be used to temporarily store various types of data that have been output or are to be output.

[0169] The processor 42 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chip in some embodiments. The processor 42 is generally used to control the overall operation of the computer device 4. In this embodiment, the processor 42 is used to run computer-readable instructions or process data stored in the memory 41, such as computer-readable instructions for running the behavior risk assessment method.

[0170] The network interface 43 may include a wireless network interface or a wired network interface. The network interface 43 is generally used to establish a communication connection between the computer device 4 and other electronic devices.

[0171] This embodiment provides a computer device, which obtains user behavior data and user browsing data, makes content violation judgments on the user browsing data, obtains illegal browsing areas, overlays and maps the user behavior data to the user browsing data, obtains a user attention distribution map, performs correlation mining analysis on the user attention distribution map and the illegal browsing areas, and obtains a violation risk report. Through comprehensive user behavior data and browsing content analysis, illegal content areas are automatically identified and marked, and detailed violation risk reports are generated by associating areas that users actually pay attention to, thereby improving the efficiency and accuracy of monitoring illegal propaganda and accurately evaluating the influence of illegal information on users, so that financial institutions can implement more targeted regulatory measures, effectively protect consumer rights and interests, and prevent the spread of improper marketing behaviors.

[0172] The present application also provides another embodiment, namely, providing a computer-readable storage medium, wherein the computer-readable storage medium stores computer-readable instructions, and the computer-readable instructions can be executed by at least one processor to enable the at least one processor to perform the steps of the behavioral risk assessment method as described above.

[0173] This embodiment provides a computer-readable storage medium, which obtains user behavior data and user browsing data, makes content violation judgments on the user browsing data, obtains illegal browsing areas, overlays and maps the user behavior data to the user browsing data, obtains a user attention distribution map, performs correlation mining analysis on the user attention distribution map and the illegal browsing areas, and obtains a violation risk report. Through comprehensive user behavior data and browsing content analysis, illegal content areas are automatically identified and marked, and a detailed violation risk report is generated by associating the areas that users actually pay attention to, thereby improving the efficiency and accuracy of monitoring illegal propaganda and accurately evaluating the influence of illegal information on users, so that financial institutions can implement more targeted regulatory measures, effectively protect consumer rights and interests, and prevent the spread of improper marketing practices.

[0174] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0175] Obviously, the embodiments described above are only some embodiments of the present application, rather than all embodiments. The preferred embodiments of the present application are given in the accompanying drawings, but they do not limit the patent scope of the present application. The present application can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present application more thorough and comprehensive. Although the present application is described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions recorded in the aforementioned specific implementation methods, or to perform equivalent replacement of some of the technical features therein. Any equivalent structure made using the contents of the specification and drawings of this application, directly or indirectly used in other related technical fields, is similarly within the scope of patent protection of this application.

Claims

1. A behavior risk assessment method, characterized in that: The steps include: Obtain user behavior data and user browsing data; Conducting content violation judgment on the user's browsing data to obtain a violation browsing area; Overlaying and mapping the user behavior data onto the user browsing data to obtain a user attention distribution map; The user attention distribution map and the illegal browsing area are subjected to correlation mining analysis to obtain a violation risk report.

2. The behavior risk assessment method according to claim 1, characterized in that: The step of determining the violation of the content of the user's browsing data to obtain the violation browsing area specifically includes: Divide the user browsing data into regions by using a page structure analysis algorithm to obtain a target browsing region; The target browsing area is identified and processed by optical character recognition technology to obtain the illegal browsing area.

3. The behavior risk assessment method according to claim 2, characterized in that: The step of identifying the target browsing area by optical character recognition technology to obtain the illegal browsing area specifically includes: Performing text recognition processing on the target browsing area by optical character recognition technology to obtain target text data; Perform keyword matching processing on the target text data to obtain a target illegal text; Perform font detection processing on the target illegal text to determine the illegal browsing area.

4. The behavior risk assessment method according to claim 1, characterized in that: The user behavior data includes user eye movement data and mouse movement data, and the step of superimposing and mapping the user behavior data onto the user browsing data to obtain a user attention distribution map specifically includes: The user's eye movement data and the mouse movement data are superimposed on the user's browsing data by a heat map generation algorithm to obtain a target area distribution map; The target area distribution map is associated with the illegal browsing area through a spatial matching algorithm to obtain the user attention distribution map.

5. The behavior risk assessment method according to claim 1, characterized in that: The user attention distribution map includes a high attention area and a low attention area. The step of performing association mining analysis on the user attention distribution map and the illegal browsing area to obtain a violation risk report specifically includes: Analyzing the overlap between the high-attention area and the illegal browsing area through an association rule mining algorithm to obtain a first visual attraction index; Analyzing the overlap between the low-attention area and the illegal browsing area through an association rule mining algorithm to obtain a second visual attraction index; Performing weighted processing on the first visual attraction index and the second visual attraction index to obtain a target visual attraction index; The illegal browsing area is matched with the target visual attraction index to obtain the illegal risk report.

6. The behavior risk assessment method according to claim 5, characterized in that: After the step of matching the illegal browsing area by the target visual attraction index to obtain the illegal risk report, the method further includes: Performing path mining on the illegal browsing area to obtain a risk propagation path; Integrate the risk propagation path into the violation risk report, and update the violation risk report.

7. The behavior risk assessment method according to claim 6, characterized in that: The step of performing path mining on the illegal browsing area to obtain the risk propagation path specifically includes: Constructing a relationship network for the user behavior data and the user browsing data to obtain a user behavior relationship network; The risk propagation path is obtained by performing path mining processing on the user behavior relationship network through a graph mining algorithm.

8. A behavior risk assessment device, characterized in that: include: The acquisition module is used to obtain user behavior data and user browsing data; A judgment module, used to judge the violation of content on the user's browsing data and obtain a violation browsing area; A mapping module, used to map the user behavior data onto the user browsing data to obtain a user attention distribution map; The analysis module is used to perform correlation mining analysis on the user attention distribution map and the illegal browsing area to obtain a violation risk report.

9. A computer device, characterized in that: The method comprises a memory and a processor, wherein the memory stores computer-readable instructions, and the processor implements the steps of the behavior risk assessment method according to any one of claims 1 to 7 when executing the computer-readable instructions.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-readable instructions, and when the computer-readable instructions are executed by a processor, the steps of the behavior risk assessment method according to any one of claims 1 to 7 are implemented.

Images