A financial transaction system user interface development and design system and method

Abstract

The present invention discloses a user interface development and design system and method for a financial transaction system, relates to the field of financial transactions, and solves the problem of insufficient security in existing user interface development and design systems for financial transaction systems. The system comprises a behavior collection module: used for obtaining multiple interface design sub-periods, performing occurrence probability analysis on click events in each interface design sub-period based on historical login records, and obtaining user historical operation collection data based on the analysis results; an event prediction module: used for performing real-time operation prediction on a target user, and obtaining predicted click events based on the prediction results; a page layout module: obtaining multiple interface sub-areas, performing position distance analysis between each interface sub-area and the user's visual focus point, and marking the predicted click area based on the analysis results; and an operation control module: used for performing operation control on the user. The present invention can improve the security and intelligence of the user interface of the financial transaction system.

Description

Technical Field

[0001] The present invention belongs to the field of financial transactions and relates to system development technology, in particular to a system and method for developing and designing a user interface of a financial transaction system. Background Art

[0002] The existing user interface development and design system for financial transaction systems has the following specific defects when developing user interfaces:

[0003] 1. The user interface developed by the existing financial transaction system user interface development and design system cannot predict users' real-time operation behavior based on their historical login data. There is no separate operation window developed for the prediction results, which cannot effectively improve user operation efficiency.

[0004] 2. The user interface developed by the existing financial transaction system user interface development and design system is unable to conduct commonality analysis between the user's operation data in the current login state and historical operation data, and is unable to evaluate the user's operation security based on the analysis results, resulting in certain security risks in financial transactions.

[0005] To this end, we propose a system and method for developing and designing the user interface of a financial transaction system. Summary of the Invention

[0006] In view of the shortcomings of the existing technology, the purpose of the present invention is to provide a financial transaction system user interface development and design system and method, and the present invention aims to improve the security and operational efficiency of the financial transaction system.

[0007] To achieve the above objectives, the present invention adopts the following technical solutions: A financial transaction system user interface development and design system, comprising:

[0008] Behavior Collection Module: This module collects multiple historical login records of the target user on the user interface, divides the time period covered by the historical login records into multiple interface design sub-periods, captures each click event involved in the historical login records, analyzes the occurrence probability of click events in each interface design sub-period, and obtains multiple event transition probability matrices based on the analysis results to obtain the user's historical operation data.

[0009] Event prediction module: used to predict the real-time operation of the target user based on the user's historical operation data, and obtain the predicted click event based on the prediction results;

[0010] Page layout module: used to divide the user interface into several interface sub-areas, obtain the user's visual focus, analyze the relative position and relative distance between each interface sub-area and the user's visual focus, and mark the predicted click area based on the analysis results;

[0011] Operation control module: used to analyze the periodic number of clicks on the predicted click area and perform operation control on the user based on the analysis results.

[0012] Furthermore, the user's historical operation data is collected and acquired, as follows:

[0013] Randomly select one user from multiple logged-in users served by the user interface of the financial transaction system as a target user;

[0014] Acquire multiple historical login records of the target user in the user interface of the financial transaction system, obtain user click events in each historical login record, obtain multiple user click events, and mark the obtained multiple user click events as click event 1 to click event p respectively;

[0015] Divide the period during which the user interface of the financial transaction system is open into a plurality of interface design sub-periods, and arbitrarily select a sample design sub-period from the obtained plurality of interface design sub-periods;

[0016] Perform commonality analysis on click events in the sample design sub-period based on historical login records to obtain the event transition probability matrix corresponding to the sample design sub-period;

[0017] The event transition probability matrix corresponding to each interface design sub-period is obtained respectively to obtain the user's historical operation collection data.

[0018] Furthermore, the event transition probability matrix corresponding to the sample design sub-period is obtained as follows:

[0019] Acquire multiple historical login records in the sample design sub-period, and acquire click event sequences from click event 1 to click event p in each historical login record according to the historical login records to obtain multiple click event sequences;

[0020] Traverse each click event sequence, count the number of times each click event i directly transfers to click event j, and obtain the click calculation matrix C;

[0021] In multiple click event sequences, if there is a direct transfer from click event i to click event j, the event transfer probability matrix P corresponding to the sample design sub-period is calculated by the click calculation matrix C;

[0022] Calculate the event transition probability matrix P. The specific formula is as follows:

[0023] ;

[0024] Where P is the click calculation matrix, P(i, j) represents the probability value of click event i directly transferring to click event j, C is the click calculation matrix, C(i, j) represents the number of times click event i directly transfers to click event j, and C(i, k) represents the number of times click event i directly transfers to click event k;

[0025] In multiple click event sequences, if there is no direct transition from click event i to click event j, the event transition probability matrix P corresponding to the sample design sub-period is calculated by the click calculation matrix C;

[0026] Calculate the event transition probability matrix P. The specific formula is as follows:

[0027] ;

[0028] Where P is the click calculation matrix, P(i,j) represents the probability value of click event i directly transferring to click event j, C is the click calculation matrix, C(i,j) represents the number of times click event i directly transfers to click event j, C(i,k) represents the number of times click event i directly transfers to click event k, and p is the number value corresponding to the user click event.

[0029] Furthermore, the predicted click event is obtained as follows:

[0030] Obtain user historical operation collection data, and obtain the event transition probability matrix corresponding to each interface design sub-period based on the user historical operation collection data;

[0031] Mark the user click event closest to the current moment as a real-time click event, obtain the event transition probability matrix corresponding to the current moment, and obtain the real-time event probability matrix;

[0032] Substitute the real-time click event and click event 1 into the real-time event probability matrix to obtain the real-time transfer probability corresponding to click event 1. Substitute the real-time click event and click event 2 into the real-time event probability matrix to obtain the real-time transfer probability corresponding to click event 2. Similarly, substitute the real-time click event and click event p into the real-time event probability matrix to obtain the real-time transfer probability corresponding to click event p.

[0033] By analyzing the real-time transfer probability corresponding to each click event, the predicted click event is obtained.

[0034] Furthermore, the real-time transfer probability corresponding to each click event is obtained as follows:

[0035] Obtain the real-time transition occurrence probabilities corresponding to click event 1 to click event p, compare the values ​​of the obtained multiple real-time transition occurrence probabilities, and mark the click event corresponding to the real-time transition occurrence probability with the largest value as the maximum probability click event;

[0036] If there are multiple maximum probability click events, obtain the time value corresponding to the previous operation of each maximum probability click event, obtain multiple event operation time values, calculate the difference between the event operation time value and the current time value, and mark the maximum probability click event with the smallest time difference as the predicted click event;

[0037] If there is only one maximum probability click event, the maximum probability click event is marked as the predicted click event.

[0038] Furthermore, the predicted click area is obtained as follows:

[0039] Dividing the user interface into a plurality of interface sub-regions, and arbitrarily selecting a sample interface sub-region from the divided plurality of interface sub-regions;

[0040] Use a 3D structured light device to perform 3D scanning of the user interface and the spatial area where the user is located, and create a 3D model of the user interface based on the scanning results;

[0041] Creating a three-dimensional coordinate system for the user interface, performing a position analysis on the target user based on the three-dimensional coordinate system for the user interface, and obtaining a first sight feature point and a second sight feature point based on the analysis result;

[0042] Use the ELG algorithm to obtain the gaze ray starting from the first sight feature point to obtain the first sight line, and use the ELG algorithm to obtain the gaze ray starting from the second sight feature point to obtain the second sight line;

[0043] Obtain the intersection of the first target sight line and the first coordinate plane to obtain the first sight line intersection point, obtain the intersection of the second target sight line and the first coordinate plane to obtain the second sight line intersection point, connect the first sight line intersection point and the second sight line intersection point in the first coordinate plane, and mark the midpoint of the connection line as the user sight line focus point;

[0044] If the user's gaze focus is within the user interface, the sub-area of ​​the interface where the user's gaze focus is located is marked as a gaze concentration area;

[0045] If the user's gaze focus is not within the user interface, analyze the boundary distance between each interface sub-area and the user's gaze focus, and mark the gaze concentration area based on the analysis results;

[0046] Get the predicted click event, automatically set the eye-gathering area as the operation area of ​​the predicted click event, and name it the predicted click area.

[0047] Furthermore, the first sight line feature point and the second sight line feature point are obtained as follows:

[0048] In the three-dimensional model of the user interface, the geometric center of the user interface is obtained and marked as the coordinate origin. The coordinate plane where the user interface is located is marked as the first coordinate plane. A plane perpendicular to the first coordinate plane is drawn through the coordinate origin to obtain the second coordinate plane. The straight line at the intersection of the first coordinate plane and the second coordinate plane is marked as the coordinate x-axis. In the second coordinate plane, a straight line perpendicular to the coordinate x-axis is drawn through the coordinate origin to obtain the coordinate y-axis. A straight line perpendicular to the second coordinate plane is drawn through the coordinate origin to obtain the coordinate z-axis. The plane rectangular coordinate system composed of the coordinate origin, the coordinate x-axis, and the coordinate y-axis is marked as the three-dimensional coordinate system of the user interface.

[0049] In the three-dimensional coordinate system of the user interface, the center point of the pupil of the left eye of the target user is marked as the first sight line feature point, and the center point of the pupil of the right eye of the target user is marked as the second sight line feature point.

[0050] Furthermore, the sight convergence area is obtained as follows:

[0051] Acquire multiple boundary pixel points corresponding to the sample interface sub-region, and name the acquired multiple pixel points as a first boundary pixel point to a jth boundary pixel point respectively;

[0052] Acquire the three-dimensional coordinates of the first boundary pixel point in the user interface three-dimensional coordinate system to obtain a first three-dimensional coordinate, and acquire the three-dimensional coordinates of the user's visual focus point in the user three-dimensional coordinate system to obtain a second three-dimensional coordinate;

[0053] The first three-dimensional coordinate and the second three-dimensional coordinate are calculated to obtain a distance value between the user's sight focus point and the first boundary pixel point, and the distance value is named the first sight movement distance value;

[0054] Calculate the first sight moving distance value. The specific formula is as follows:

[0055] ;

[0056] Where Yj1 is the first sight moving distance value, (x1, y1, z1) is the first three-dimensional coordinate, and (x2, y2, z2) is the second three-dimensional coordinate;

[0057] Obtain the distance values ​​from the user's sight focus point and the second boundary pixel point to the j-th boundary pixel point respectively, and obtain the second sight movement distance value to the j-th sight movement distance value;

[0058] Calculate the average of the first sight movement distance value to the j-th sight movement distance value to obtain the average sight movement distance corresponding to the sample interface sub-area;

[0059] The average sight movement distance corresponding to each interface sub-area is obtained respectively to obtain multiple average sight movement distances;

[0060] The obtained multiple average sight movement distances are compared in numerical value, and the interface sub-area corresponding to the minimum average sight movement distance is marked as the sight gathering area.

[0061] Furthermore, the user is controlled by the following operations:

[0062] Obtain the predicted click area and mark an operation control monitoring period when the target user performs financial transaction operations through the user interface;

[0063] During the operation control monitoring cycle, the time value of each click of the target user on the predicted click area is obtained to obtain multiple click time values, and the obtained multiple click time events are marked in chronological order as D1 click time value to Dz click time value;

[0064] The D1 click time value to the Dz click time value are calculated to obtain the periodic predicted click interval duration;

[0065] The specific formula for calculating the periodic predicted click interval duration is as follows:

[0066] ;

[0067] Among them, Zyj is the periodic predicted click interval length, Dj i D i Click the time value, Dj i-1 D i-1 Click time value, z is the quantity value corresponding to the click time value;

[0068] Get the preset click interval duration range. If the periodic predicted click interval duration is within the preset click interval duration range, it is judged that the user's financial transaction is in a normal state. If the periodic predicted click interval duration is not within the preset click interval duration range, it is judged that the user's financial transaction is in an abnormal state. The user interface then returns to the login interface and re-authenticates the target user.

[0069] A method for developing and designing a user interface for a financial transaction system includes the following specific steps:

[0070] Step S1: Collect multiple historical login records of the target user on the user interface, divide the time period covered by the historical login records into multiple interface design sub-periods, obtain each click event involved in the historical login records, perform an occurrence probability analysis on the click event in each interface design sub-period, and obtain multiple event transition probability matrices based on the analysis results to obtain the user's historical operation collection data;

[0071] Step S2: Perform real-time operation prediction on the target user based on the user's historical operation data, and obtain predicted click events based on the prediction results;

[0072] Step S3: Divide the user interface into several interface sub-areas, obtain the user's visual focus point, perform relative position analysis and relative distance analysis between each interface sub-area and the user's visual focus point, and mark the predicted click area based on the analysis results;

[0073] Step S4: Analyze the periodic number of clicks on the predicted click area, and control the user's operation based on the analysis results.

[0074] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0075] 1. The present invention predicts the user's real-time operation behavior based on the user's historical login data, and develops a separate operation window for the prediction results, so that the user can perform one-click operation on the predicted event, thereby effectively improving the user's operation efficiency.

[0076] 2. The present invention performs commonality analysis on the user's current login operation data and historical operation data, conducts operation security assessment on the user based on the analysis results, and controls abnormal operations, which can effectively reduce the security risks existing in financial exchanges. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] To facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0078] Figure 1 is a block diagram of the overall system of the present invention;

[0079] Figure 2 It is a diagram of the implementation steps of the present invention;

[0080] Figure 3 A schematic diagram of the three-dimensional coordinate system of the user interface in the present invention;

[0081] Figure 4 Schematic diagram of the sample interface sub-region in the present invention. DETAILED DESCRIPTION

[0082] The technical solutions of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention. Example 1

[0083] See also Figure 1 The present invention provides a technical solution: a user interface development and design system for a financial transaction system, comprising a behavior collection module, an event prediction module, a page layout module, an operation control module, and a server, wherein the behavior collection module, event prediction module, page layout module, and operation control module are respectively connected to the server, and the server controls the behavior collection module, event prediction module, page layout module, and operation control module respectively;

[0084] The behavior collection module collects multiple historical login records of the target user on the user interface, divides the time period covered by the historical login records into multiple interface design sub-periods, obtains each click event involved in the historical login records, analyzes the occurrence probability of the click event in each interface design sub-period, and obtains multiple event transition probability matrices based on the analysis results to obtain the user's historical operation collection data;

[0085] The details are as follows:

[0086] Randomly select one user from multiple logged-in users served by the user interface of the financial transaction system as a target user;

[0087] Acquire multiple historical login records of the target user in the user interface of the financial transaction system, obtain user click events in each historical login record, obtain multiple user click events, and mark the obtained multiple user click events as click event 1 to click event p respectively;

[0088] It should be noted here that:

[0089] In this application, any user interface operation in the financial transaction system is a sequential user click event, and the interface operations involved here include but are not limited to login, deposit, transfer and loan;

[0090] In this application, p refers to the quantity value corresponding to the user click event;

[0091] The user click events involved here all correspond to a functional partition of the financial transaction system user interface.

[0092] Divide the period during which the user interface of the financial transaction system is open into a plurality of interface design sub-periods, and arbitrarily select a sample design sub-period from the obtained plurality of interface design sub-periods;

[0093] Perform commonality analysis on click events in the sample design sub-period based on historical login records to obtain the event transition probability matrix corresponding to the sample design sub-period;

[0094] The details are as follows:

[0095] Acquire multiple historical login records in the sample design sub-period, and acquire click event sequences from click event 1 to click event p in each historical login record according to the historical login records to obtain multiple click event sequences;

[0096] It should be noted here that:

[0097] In this application, each historical registration record corresponds to a click event sequence;

[0098] In a specific implementation, there is a historical login record, in which there are five click events. The corresponding click event sequence is as follows:

[0099] Click event 1 → click event 2 → click event 5 → click event 7 → click event 3;

[0100] “→” means direct transfer, that is, after a click event is completed, it directly transfers to the next click event. For example: “Click event 1 → Click event 2” means that the next click event completed by the user after completing click event 1 is click event 2;

[0101] Traverse each click event sequence, count the number of times each click event i directly transfers to click event j, and obtain the click calculation matrix C;

[0102] It should be noted here that:

[0103] In this application, since there are at most p click events in the click event sequence, the size of the click calculation matrix is ​​p×p;

[0104] In this application, the click event i involved here can be any click event from click event 1 to click event p, and the click event j involved here can be any click event from click event 1 to click event p, and both i and j are integers greater than 1;

[0105] In a specific implementation, in a plurality of click event sequences, if the number of occurrences of click event 3 immediately following click event 1 is 50, then C(1,3)=50.

[0106] In multiple click event sequences, if there is a direct transfer from click event i to click event j, the event transfer probability matrix P corresponding to the sample design sub-period is calculated by the click calculation matrix C;

[0107] Calculate the event transition probability matrix P. The specific formula is as follows:

[0108] ;

[0109] Where P is the click calculation matrix, P(i, j) represents the probability value of click event i directly transferring to click event j, C is the click calculation matrix, C(i, j) represents the number of times click event i directly transfers to click event j, and C(i, k) represents the number of times click event i directly transfers to click event k;

[0110] It should be noted here that:

[0111] In this application, the k involved here can be any click event from click event 1 to click event p.

[0112] In the specific implementation, there are the following experimental data:

[0113] If the click event sequence is as follows:

[0114] Click event 1 → click event 2 → click event 5 → click event 7 → click event 3 → click event 2 → click event 3 → click event 2 → click event 3 → click event 5 → click event 7 → click event 3 → click event 2;

[0115] The above click event sequence has 13 click events, the user clicks 12 times, and C(3,2)=3, so P(i,j)=0.25.

[0116] In multiple click event sequences, if there is no direct transition from click event i to click event j, the event transition probability matrix P corresponding to the sample design sub-period is calculated by the click calculation matrix C;

[0117] Calculate the event transition probability matrix P. The specific formula is as follows:

[0118] ;

[0119] Where P is the click calculation matrix, P(i,j) represents the probability value of click event i directly transferring to click event j, C is the click calculation matrix, C(i,j) represents the number of times click event i directly transfers to click event j, C(i,k) represents the number of times click event i directly transfers to click event k, and p is the number value corresponding to the user click event.

[0120] Repeat the process of obtaining the event transition probability matrix corresponding to the sample design sub-period, and obtain the event transition probability matrix corresponding to each interface design sub-period respectively to obtain the user's historical operation collection data;

[0121] The event prediction module collects data about the user's historical operations and makes real-time operation predictions for the target user, and obtains predicted click events based on the prediction results;

[0122] The details are as follows:

[0123] Obtain user historical operation collection data, and obtain the event transition probability matrix corresponding to each interface design sub-period based on the user historical operation collection data;

[0124] Mark the user click event closest to the current moment as a real-time click event, obtain the event transition probability matrix corresponding to the current moment, and obtain the real-time event probability matrix;

[0125] Substitute the real-time click event and click event 1 into the real-time event probability matrix to obtain the real-time transfer probability corresponding to click event 1. Substitute the real-time click event and click event 2 into the real-time event probability matrix to obtain the real-time transfer probability corresponding to click event 2. Similarly, substitute the real-time click event and click event p into the real-time event probability matrix to obtain the real-time transfer probability corresponding to click event p.

[0126] By analyzing the real-time transfer probability corresponding to each click event, the predicted click event is obtained;

[0127] The details are as follows:

[0128] Obtain the real-time transition occurrence probabilities corresponding to click event 1 to click event p, compare the values ​​of the obtained multiple real-time transition occurrence probabilities, and mark the click event corresponding to the real-time transition occurrence probability with the largest value as the maximum probability click event;

[0129] If there are multiple maximum probability click events, obtain the time value corresponding to the previous operation of each maximum probability click event, obtain multiple event operation time values, calculate the difference between the event operation time value and the current time value, and mark the maximum probability click event with the smallest time difference as the predicted click event;

[0130] If there is only one maximum probability click event, the maximum probability click event is marked as the predicted click event;

[0131] The page layout module divides the user interface into several sub-areas, obtains the user's visual focus, analyzes the relative position and distance between each sub-area and the user's visual focus, and marks the predicted click area based on the analysis results;

[0132] The details are as follows:

[0133] Dividing the user interface into a plurality of interface sub-regions, and arbitrarily selecting a sample interface sub-region from the divided plurality of interface sub-regions;

[0134] Use a 3D structured light device to perform 3D scanning of the user interface and the spatial area where the user is located, and create a 3D model of the user interface based on the scanning results;

[0135] See also Figure 3 In the three-dimensional model of the user interface, the geometric center of the user interface is obtained and marked as the coordinate origin. The coordinate plane where the user interface is located is marked as the first coordinate plane. A plane perpendicular to the first coordinate plane is drawn through the coordinate origin to obtain the second coordinate plane. The straight line at the intersection of the first coordinate plane and the second coordinate plane is marked as the coordinate x-axis. In the second coordinate plane, a straight line perpendicular to the coordinate x-axis is drawn through the coordinate origin to obtain the coordinate y-axis. A straight line perpendicular to the second coordinate plane is drawn through the coordinate origin to obtain the coordinate z-axis. The plane rectangular coordinate system composed of the coordinate origin, the coordinate x-axis, and the coordinate y-axis is marked as the three-dimensional coordinate system of the user interface.

[0136] It should be noted here that:

[0137] In the present application, the first coordinate plane involved here is an extended plane of the user interface, that is, the user interface only occupies a partial area of ​​the first coordinate plane.

[0138] In the three-dimensional coordinate system of the user interface, the center point of the pupil of the target user's left eye is marked as the first sight feature point, and the center point of the pupil of the target user's right eye is marked as the second sight feature point;

[0139] Use the ELG algorithm to obtain the gaze ray starting from the first sight feature point to obtain the first sight line, and use the ELG algorithm to obtain the gaze ray starting from the second sight feature point to obtain the second sight line;

[0140] It should be noted here that:

[0141] In this application, the ELG algorithm: a line of sight estimation method based on eye area landmarks can obtain the line of sight of the target user;

[0142] In the present application, the first sight line and the second sight line involved here both require that the target user faces the user interface, and it is necessary to ensure that the first sight line and the second sight line have an intersection with the first coordinate plane.

[0143] Obtain the intersection of the first target sight line and the first coordinate plane to obtain the first sight line intersection point, obtain the intersection of the second target sight line and the first coordinate plane to obtain the second sight line intersection point, connect the first sight line intersection point and the second sight line intersection point in the first coordinate plane, and mark the midpoint of the connection line as the user sight line focus point;

[0144] If the user's gaze focus is within the user interface, the sub-area of ​​the interface where the user's gaze focus is located is marked as a gaze concentration area;

[0145] If the user's gaze focus is not within the user interface, analyze the boundary distance between each interface sub-area and the user's gaze focus, and mark the gaze concentration area based on the analysis results;

[0146] The details are as follows:

[0147] See also Figure 4 , acquiring a plurality of boundary pixel points corresponding to the sample interface sub-region, and naming the acquired plurality of pixel points as a first boundary pixel point to a jth boundary pixel point respectively;

[0148] It should be noted here that:

[0149] In this application, j mentioned here is the number value corresponding to the boundary pixel point, and j is an integer greater than 0.

[0150] Acquire the three-dimensional coordinates of the first boundary pixel point in the user interface three-dimensional coordinate system to obtain a first three-dimensional coordinate, and acquire the three-dimensional coordinates of the user's visual focus point in the user three-dimensional coordinate system to obtain a second three-dimensional coordinate;

[0151] The first three-dimensional coordinate and the second three-dimensional coordinate are calculated to obtain a distance value between the user's sight focus point and the first boundary pixel point, and the distance value is named the first sight movement distance value;

[0152] Calculate the first sight moving distance value. The specific formula is as follows:

[0153] ;

[0154] Where Yj1 is the first sight moving distance value, (x1, y1, z1) is the first three-dimensional coordinate, and (x2, y2, z2) is the second three-dimensional coordinate;

[0155] It should be noted here that:

[0156] In a specific implementation, if the first three-dimensional coordinates are (3, 7, 9) and the second three-dimensional coordinates are (3, 10, 15), then the first sight movement distance value can be calculated to be 6.71.

[0157] Repeat the process of obtaining the first sight movement distance value, respectively obtain the distance values ​​from the user's sight focus point and the second boundary pixel point to the j-th boundary pixel point, and obtain the second sight movement distance value to the j-th sight movement distance value;

[0158] Calculate the average of the first sight movement distance value to the j-th sight movement distance value to obtain the average sight movement distance corresponding to the sample interface sub-area;

[0159] Repeat the average sight movement distance corresponding to the sample interface sub-area, respectively obtain the average sight movement distance corresponding to each interface sub-area, and obtain multiple average sight movement distances;

[0160] Compare the values ​​of the multiple average sight movement distances obtained, and mark the interface sub-area corresponding to the minimum average sight movement distance as the sight concentration area;

[0161] It should be noted here that:

[0162] In the present application, if there is a situation where the minimum average sight movement distances are parallel, the minimum sight movement distance values ​​of the parallel interface sub-areas are compared, that is, the minimum value from the first sight movement distance value to the jth sight movement distance value.

[0163] Get the predicted click event, and automatically set the eye-gathering area as the operation area of ​​the predicted click event, and name it the predicted click area;

[0164] It should be noted here that:

[0165] In order not to affect the user's operation of non-predicted click events, the predicted click area involved here appears in the user interface for 5 seconds, and the starting time of appearance is the time when the user's visual focus is on the user interface.

[0166] The page layout module obtains the predicted click area and transmits it to the operation control module;

[0167] The operation control module analyzes the periodic number of clicks on the predicted click area and controls the user's operation based on the analysis results;

[0168] The details are as follows:

[0169] Obtain the predicted click area and mark an operation control monitoring period when the target user performs financial transaction operations through the user interface;

[0170] During the operation control monitoring cycle, the time value of each click of the target user on the predicted click area is obtained to obtain multiple click time values, and the obtained multiple click time events are marked in chronological order as D1 click time value to Dz click time value;

[0171] It should be noted here that:

[0172] In this application, D is an identifier corresponding to the click time value, and z is a quantity corresponding to the click time value, and z is an integer greater than 0.

[0173] The D1 click time value to the Dz click time value are calculated to obtain the periodic predicted click interval duration;

[0174] The specific formula for calculating the periodic predicted click interval duration is as follows:

[0175] ;

[0176] Among them, Zyj is the periodic predicted click interval length, Dj i D i Click the time value, Dj i-1 D i-1 Click time value, z is the quantity value corresponding to the click time value;

[0177] It should be noted here that:

[0178] In the specific implementation, there are the following experimental data:

[0179] It is known that the click time value of D1 is 7 hours 54 minutes 31 seconds, the click time value of D2 is 7 hours 54 minutes 38 seconds, the click time value of D2 is 7 hours 54 minutes 44 seconds, the click time value of D3 is 7 hours 54 minutes 59 seconds, the click time value of D4 is 7 hours 55 minutes 08 seconds, and z is 5. It can be calculated that the periodic predicted click interval duration is 9.25 seconds.

[0180] Get the preset click interval duration range. If the periodic predicted click interval duration is within the preset click interval duration range, it is judged that the user's financial transaction is in a normal state. If the periodic predicted click interval duration is not within the preset click interval duration range, it is judged that the user's financial transaction is in an abnormal state. The user interface then returns to the login interface and re-authenticates the target user.

[0181] It should be noted here that:

[0182] The normal state involved here includes the situation where the periodically predicted click interval duration is within the boundary of a preset click interval duration interval.

[0183] If the periodically predicted click interval duration is not within the preset click interval duration range, the user's multiple click operations lack similarity with historical operations, and the login interface here determines that the user operation is abnormal;

[0184] The process of obtaining the preset click interval duration involved here is as follows:

[0185] Collect the periodic predicted click interval durations corresponding to multiple platform users who are known to be in normal status, calculate the average value and standard deviation of the multiple periodic predicted click interval durations, calculate the sum of the average value and the obtained standard to obtain the upper limit of the preset click interval duration range, calculate the difference between the average value and the obtained standard to obtain the lower limit of the preset click interval duration range, and mark the value between the lower limit of the preset click interval duration range and the upper limit of the preset click interval duration range as the preset click interval duration range.

[0186] In this application, if a corresponding calculation formula appears, the above calculation formula is dimensionless and its numerical calculation is performed. The weight coefficient, proportional coefficient and other coefficients in the formula are set to a result value obtained by quantifying each parameter. Regarding the size of the weight coefficient and the proportional coefficient, as long as it does not affect the proportional relationship between the parameter and the result value, it is acceptable.

[0187] Example 2

[0188] See also Figure 2 Based on another concept of the same invention, a method for developing and designing a user interface for a financial transaction system is proposed, comprising the following steps:

[0189] Step S1: Collect multiple historical login records of the target user on the user interface, divide the time period covered by the historical login records into multiple interface design sub-periods, obtain each click event involved in the historical login records, perform an occurrence probability analysis on the click event in each interface design sub-period, and obtain multiple event transition probability matrices based on the analysis results to obtain the user's historical operation collection data;

[0190] The step S1 further includes the following specific steps:

[0191] Step S11: randomly selecting one user from among multiple login users served by the user interface of the financial transaction system as a target user;

[0192] Step S12: Acquire multiple historical login records of the target user in the user interface of the financial transaction system, respectively acquire user click events in each historical login record to obtain multiple user click events, and mark the acquired multiple user click events as click event 1 to click event p, respectively;

[0193] Step S13: Divide the period during which the user interface of the financial transaction system is open into a plurality of interface design sub-periods, and arbitrarily select a sample design sub-period from the obtained plurality of interface design sub-periods;

[0194] Step S14: performing an operation commonality analysis on the click events in the sample design sub-period based on the historical login records to obtain an event transition probability matrix corresponding to the sample design sub-period;

[0195] The step S14 further includes the following specific steps:

[0196] Acquire multiple historical login records in the sample design sub-period, and acquire click event sequences from click event 1 to click event p in each historical login record according to the historical login records to obtain multiple click event sequences;

[0197] Traverse each click event sequence, count the number of times each click event i directly transfers to click event j, and obtain the click calculation matrix C;

[0198] In multiple click event sequences, if there is a direct transfer from click event i to click event j, the event transfer probability matrix P corresponding to the sample design sub-period is calculated by the click calculation matrix C;

[0199] Calculate the event transition probability matrix P. The specific formula is as follows:

[0200] ;

[0201] Where P is the click calculation matrix, P(i, j) represents the probability value of click event i directly transferring to click event j, C is the click calculation matrix, C(i, j) represents the number of times click event i directly transfers to click event j, and C(i, k) represents the number of times click event i directly transfers to click event k;

[0202] In multiple click event sequences, if there is no direct transition from click event i to click event j, the event transition probability matrix P corresponding to the sample design sub-period is calculated by the click calculation matrix C;

[0203] Calculate the event transition probability matrix P. The specific formula is as follows:

[0204] ;

[0205] Where P is the click calculation matrix, P(i,j) represents the probability value of click event i directly transferring to click event j, C is the click calculation matrix, C(i,j) represents the number of times click event i directly transfers to click event j, C(i,k) represents the number of times click event i directly transfers to click event k, and p is the number value corresponding to the user click event.

[0206] Step S15: Obtain the event transition probability matrix corresponding to each interface design sub-period to obtain user historical operation collection data;

[0207] Step S2: Perform real-time operation prediction on the target user based on the user's historical operation data, and obtain predicted click events based on the prediction results;

[0208] The step S2 further includes the following specific steps:

[0209] Step S21: Obtain user historical operation collection data, and obtain the event transition probability matrix corresponding to each interface design sub-period based on the user historical operation collection data;

[0210] Step S22: Mark the user click event closest to the current moment as a real-time click event, obtain the event transition probability matrix corresponding to the current moment, and obtain a real-time event probability matrix;

[0211] Step S23: Substitute the real-time click event and click event 1 into the real-time event probability matrix to obtain the real-time transfer probability corresponding to click event 1. Substitute the real-time click event and click event 2 into the real-time event probability matrix to obtain the real-time transfer probability corresponding to click event 2. Similarly, substitute the real-time click event and click event p into the real-time event probability matrix to obtain the real-time transfer probability corresponding to click event p.

[0212] Step S24: Analyzing the real-time transfer occurrence probability corresponding to each click event to obtain a predicted click event;

[0213] The step S24 further includes the following specific steps:

[0214] Obtain the real-time transition occurrence probabilities corresponding to click event 1 to click event p, compare the values ​​of the obtained multiple real-time transition occurrence probabilities, and mark the click event corresponding to the real-time transition occurrence probability with the largest value as the maximum probability click event;

[0215] If there are multiple maximum probability click events, obtain the time value corresponding to the previous operation of each maximum probability click event, obtain multiple event operation time values, calculate the difference between the event operation time value and the current time value, and mark the maximum probability click event with the smallest time difference as the predicted click event;

[0216] If there is only one maximum probability click event, the maximum probability click event is marked as the predicted click event;

[0217] Step S3: Divide the user interface into several interface sub-areas, obtain the user's visual focus point, perform relative position analysis and relative distance analysis between each interface sub-area and the user's visual focus point, and mark the predicted click area based on the analysis results;

[0218] The step S3 further includes the following specific steps:

[0219] Step S31: Divide the user interface into a plurality of interface sub-regions, and arbitrarily select a sample interface sub-region from the divided plurality of interface sub-regions;

[0220] Step S32: Using a 3D structured light device to perform a 3D scan of the user interface and the spatial area where the user is located, and creating a three-dimensional model of the user interface based on the scan results;

[0221] Step S33: creating a three-dimensional coordinate system of the user interface, performing position analysis on the target user based on the three-dimensional coordinate system of the user interface, and obtaining a first sight feature point and a second sight feature point based on the analysis results;

[0222] The step S33 further includes the following specific steps:

[0223] In the three-dimensional model of the user interface, the geometric center of the user interface is obtained and marked as the coordinate origin. The coordinate plane where the user interface is located is marked as the first coordinate plane. A plane perpendicular to the first coordinate plane is drawn through the coordinate origin to obtain the second coordinate plane. The straight line at the intersection of the first coordinate plane and the second coordinate plane is marked as the coordinate x-axis. In the second coordinate plane, a straight line perpendicular to the coordinate x-axis is drawn through the coordinate origin to obtain the coordinate y-axis. A straight line perpendicular to the second coordinate plane is drawn through the coordinate origin to obtain the coordinate z-axis. The plane rectangular coordinate system composed of the coordinate origin, the coordinate x-axis, and the coordinate y-axis is marked as the three-dimensional coordinate system of the user interface.

[0224] In the three-dimensional coordinate system of the user interface, the center point of the pupil of the target user's left eye is marked as the first sight feature point, and the center point of the pupil of the target user's right eye is marked as the second sight feature point;

[0225] Step S34: using the ELG algorithm to obtain the gaze ray starting from the first sight feature point to obtain the first sight line, and using the ELG algorithm to obtain the gaze ray starting from the second sight feature point to obtain the second sight line;

[0226] Step S35: Obtain the intersection of the first target sight line and the first coordinate plane to obtain the first sight line intersection point; obtain the intersection of the second target sight line and the first coordinate plane to obtain the second sight line intersection point; connect the first sight line intersection point and the second sight line intersection point in the first coordinate plane, and mark the midpoint of the connection line as the user sight line focus point;

[0227] Step S36: If the user's sight focus is within the user interface, the sub-region of the interface where the user's sight focus is located is marked as a sight concentration region;

[0228] Step S37: If the user's gaze focus is not within the user interface, analyze the boundary distance between each interface sub-area and the user's gaze focus, and mark the gaze concentration area based on the analysis results;

[0229] The step S37 further includes the following specific steps:

[0230] Acquire multiple boundary pixel points corresponding to the sample interface sub-region, and name the acquired multiple pixel points as a first boundary pixel point to a jth boundary pixel point respectively;

[0231] Acquire the three-dimensional coordinates of the first boundary pixel point in the user interface three-dimensional coordinate system to obtain a first three-dimensional coordinate, and acquire the three-dimensional coordinates of the user's visual focus point in the user three-dimensional coordinate system to obtain a second three-dimensional coordinate;

[0232] The first three-dimensional coordinate and the second three-dimensional coordinate are calculated to obtain a distance value between the user's sight focus point and the first boundary pixel point, and the distance value is named the first sight movement distance value;

[0233] Calculate the first sight moving distance value. The specific formula is as follows:

[0234] ;

[0235] Where Yj1 is the first sight moving distance value, (x1, y1, z1) is the first three-dimensional coordinate, and (x2, y2, z2) is the second three-dimensional coordinate;

[0236] Repeat the process of obtaining the first sight movement distance value, respectively obtain the distance values ​​from the user's sight focus point and the second boundary pixel point to the j-th boundary pixel point, and obtain the second sight movement distance value to the j-th sight movement distance value;

[0237] Calculate the average of the first sight movement distance value to the j-th sight movement distance value to obtain the average sight movement distance corresponding to the sample interface sub-area;

[0238] Repeat the average sight movement distance corresponding to the sample interface sub-area, respectively obtain the average sight movement distance corresponding to each interface sub-area, and obtain multiple average sight movement distances;

[0239] Compare the values ​​of the multiple average sight movement distances obtained, and mark the interface sub-area corresponding to the minimum average sight movement distance as the sight concentration area;

[0240] Step S38: Obtain the predicted click event, and automatically set the sight focus area as the operation area of ​​the predicted click event, and name it the predicted click area;

[0241] Step S4: Analyze the periodic number of clicks on the predicted click area, and control the user's operation based on the analysis results;

[0242] The step S4 further includes the following specific steps:

[0243] Step S41: obtaining a predicted click area, and marking an operation control monitoring period during the process of the target user performing a financial transaction operation through the user interface;

[0244] Step S42: During the operation control monitoring period, the time value of each click of the target user on the predicted click area is obtained to obtain multiple click time values, and the obtained multiple click time events are marked in chronological order as D1 click time value to Dz click time value;

[0245] Step S43: Calculate the periodic predicted click interval duration by combining the D1 click time value to the Dz click time value;

[0246] The specific formula for calculating the periodic predicted click interval duration is as follows:

[0247] ;

[0248] Among them, Zyj is the periodic predicted click interval length, Dj i D i Click the time value, Dj i-1 D i-1 Click time value, z is the quantity value corresponding to the click time value;

[0249] Step S44: Obtain the preset click interval duration interval. If the periodically predicted click interval duration is within the preset click interval duration interval, it is determined that the user's financial transaction is in a normal state. If the periodically predicted click interval duration is not within the preset click interval duration interval, it is determined that the user's financial transaction is in an abnormal state. The user interface then returns to the login interface and re-authenticates the target user.

[0250] The preferred embodiments of the present invention disclosed above are intended only to help illustrate the present invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the present invention to specific embodiments. Obviously, many modifications and variations are possible based on the contents of this specification. These embodiments are selected and described in detail in this specification to better explain the principles and practical applications of the present invention, thereby enabling those skilled in the art to better understand and utilize the present invention. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A financial transaction system user interface development and design system, characterized by: include: Behavior collection module: used to collect multiple historical login records, divide the time period covered by the historical login records into multiple interface design sub-periods, obtain each click event involved in the historical login records, perform occurrence probability analysis on the click events in each interface design sub-period, and obtain multiple event transition probability matrices based on the analysis results to obtain user historical operation collection data; Event prediction module: used to predict the real-time operation of the target user based on the user's historical operation data, and obtain the predicted click event based on the prediction results; Page layout module: used to divide the user interface into several interface sub-areas, obtain the user's visual focus, analyze the relative position and relative distance between each interface sub-area and the user's visual focus, and mark the predicted click area based on the analysis results; Operation control module: used to periodically monitor the number of clicks on the predicted click area and perform user operation control based on the monitoring results; Get the predicted click area as follows: Dividing the user interface into a plurality of interface sub-regions, and arbitrarily selecting a sample interface sub-region from the divided plurality of interface sub-regions; Perform 3D scanning of the user interface and the spatial area where the user is located, and create a 3D model of the user interface based on the scanning results; Creating a three-dimensional coordinate system for the user interface, performing a position analysis on the target user based on the three-dimensional coordinate system for the user interface, and obtaining a first sight feature point and a second sight feature point based on the analysis result; The gaze ray starting from the first sight feature point is marked as the first sight line, and the gaze ray starting from the second sight feature point is marked as the second sight line; Obtain the intersection of the first target sight line and the first coordinate plane to obtain the first sight line intersection point, obtain the intersection of the second target sight line and the first coordinate plane to obtain the second sight line intersection point, connect the first sight line intersection point and the second sight line intersection point in the first coordinate plane, and mark the midpoint of the connection line as the user sight line focus point; If the user's gaze focus is within the user interface, the sub-area of ​​the interface where the user's gaze focus is located is marked as a gaze concentration area; If the user's gaze focus is not within the user interface, analyze the boundary distance between each interface sub-area and the user's gaze focus, and mark the gaze concentration area based on the analysis results; Get the predicted click event, automatically set the eye-gathering area as the operation area of ​​the predicted click event, and name it the predicted click area.

2. A financial transaction system user interface development and design system according to claim 1, characterized in that: Acquire the user's historical operation data, as follows: Randomly select one user from multiple logged-in users served by the user interface of the financial transaction system as a target user; Acquire multiple historical login records of the target user in the user interface of the financial transaction system, obtain user click events in each historical login record, obtain multiple user click events, and mark the obtained multiple user click events as click event 1 to click event p respectively; Divide the period during which the user interface of the financial transaction system is open into a plurality of interface design sub-periods, and arbitrarily select a sample design sub-period from the obtained plurality of interface design sub-periods; According to the historical login records, the commonality analysis of the click events in the interface design sub-period is performed, and the event transfer probability matrix corresponding to each interface design sub-period is obtained according to the analysis results to obtain the user's historical operation collection data.

3. A financial transaction system user interface development and design system according to claim 2, characterized in that: The event transition probability matrix is ​​obtained as follows: Acquire multiple historical login records in the sample design sub-period, and acquire click event sequences from click event 1 to click event p in each historical login record according to the historical login records to obtain multiple click event sequences; Traverse each click event sequence, count the number of times click event i transfers to click event j, and obtain the click calculation matrix C; If C(i,j) is not 0, click on the calculation matrix C to calculate the event transition probability matrix P corresponding to the sample design sub-period; Calculate the event transition probability matrix P. The specific formula is as follows: ； Where P is the click calculation matrix, P(i, j) represents the probability value of click event i directly transferring to click event j, C is the click calculation matrix, C(i, j) represents the number of times click event i directly transfers to click event j, and C(i, k) represents the number of times click event i directly transfers to click event k; If C(i,j) is 0, click on the calculation matrix C to calculate the event transition probability matrix P corresponding to the sample design sub-period; Calculate the event transition probability matrix P. The specific formula is as follows: ； Among them, p is the quantity value corresponding to the user click event.

4. A financial transaction system user interface development and design system according to claim 1, characterized in that: Get the predicted click event as follows: Obtain user historical operation collection data, and obtain the event transition probability matrix corresponding to each interface design sub-period based on the user historical operation collection data; Mark the user click event closest to the current moment as a real-time click event, obtain the event transition probability matrix corresponding to the current moment, and obtain the real-time event probability matrix; Substitute the real-time click event and click event 1 into the real-time event probability matrix to obtain the real-time transfer probability corresponding to click event 1. Similarly, obtain the real-time transfer probability corresponding to click event p. By analyzing the real-time transfer probability corresponding to each click event, the predicted click event is obtained.

5. A financial transaction system user interface development and design system according to claim 4, characterized in that: The real-time transfer probability corresponding to each click event is obtained as follows: Obtain the real-time transition occurrence probabilities corresponding to click event 1 to click event p, compare the values ​​of the obtained multiple real-time transition occurrence probabilities, and mark the click event corresponding to the real-time transition occurrence probability with the largest value as the maximum probability click event; If there are multiple maximum probability click events, obtain the time value corresponding to the previous operation of each maximum probability click event, obtain multiple event operation time values, calculate the difference between the event operation time value and the current time value, and mark the maximum probability click event with the smallest time difference as the predicted click event; If there is only one maximum probability click event, the maximum probability click event is marked as the predicted click event.

6. A financial transaction system user interface development and design system according to claim 5, characterized in that: The first sight feature point and the second sight feature point are obtained as follows: In the three-dimensional model of the user interface, the geometric center of the user interface is obtained and marked as the coordinate origin. The coordinate plane where the user interface is located is marked as the first coordinate plane. A plane perpendicular to the first coordinate plane is drawn through the coordinate origin to obtain the second coordinate plane. The straight line at the intersection of the first coordinate plane and the second coordinate plane is marked as the coordinate x-axis. In the second coordinate plane, a straight line perpendicular to the coordinate x-axis is drawn through the coordinate origin to obtain the coordinate y-axis. A straight line perpendicular to the second coordinate plane is drawn through the coordinate origin to obtain the coordinate z-axis. The plane rectangular coordinate system composed of the coordinate origin, the coordinate x-axis, and the coordinate y-axis is marked as the three-dimensional coordinate system of the user interface. In the three-dimensional coordinate system of the user interface, the center point of the pupil of the left eye of the target user is marked as the first sight line feature point, and the center point of the pupil of the right eye of the target user is marked as the second sight line feature point.

7. A financial transaction system user interface development and design system according to claim 5, characterized in that: The sight gathering area is obtained as follows: Acquire multiple boundary pixel points corresponding to the sample interface sub-region, and name the acquired multiple pixel points as a first boundary pixel point to a jth boundary pixel point respectively; Acquire the three-dimensional coordinates of the first boundary pixel point in the user interface three-dimensional coordinate system to obtain a first three-dimensional coordinate, and acquire the three-dimensional coordinates of the user's visual focus point in the user three-dimensional coordinate system to obtain a second three-dimensional coordinate; The first three-dimensional coordinate (x1, y1, z1) and the second three-dimensional coordinate (x2, y2, z2) are calculated to obtain the first sight movement distance value Yj1; Obtain the distance values ​​from the user's sight focus point and the second boundary pixel point to the j-th boundary pixel point respectively, and obtain the second sight movement distance value to the j-th sight movement distance value; Calculate the average of the first sight movement distance value to the j-th sight movement distance value to obtain the average sight movement distance corresponding to the sample interface sub-area; The average sight movement distance corresponding to each interface sub-area is obtained respectively to obtain multiple average sight movement distances; The obtained multiple average sight movement distances are compared in numerical value, and the interface sub-area corresponding to the minimum average sight movement distance is marked as the sight gathering area.

8. A financial transaction system user interface development and design system according to claim 1, characterized in that: Perform operational control on users, as follows: Obtain the predicted click area and mark an operation control monitoring period when the target user performs financial transaction operations through the user interface; During the operation control monitoring cycle, the time value of each click of the target user on the predicted click area is obtained to obtain multiple click time values, and the obtained multiple click time events are marked in chronological order as D1 click time value to Dz click time value; The D1 click time value to the Dz click time value are calculated to obtain the periodic predicted click interval duration; The specific formula for calculating the periodic predicted click interval duration is as follows: ； Among them, Zyj is the periodic predicted click interval length, Dj i D i Click the time value, Dj i-1 D i-1 Click time value, z is the quantity value corresponding to the click time value; Obtain the preset click interval duration interval. If the periodically predicted click interval duration is within the preset click interval duration interval, it is determined that the user's financial transaction is in a normal state. If the periodically predicted click interval duration is not within the preset click interval duration interval, it is determined that the user's financial transaction is in an abnormal state, and the user interface re-authenticates the target user.

9. A method for developing and designing a user interface for a financial transaction system, characterized in that: A financial transaction system user interface development and design system applicable to any one of claims 1-8, comprising the following steps: Step S1: Collect multiple historical login records, divide the time period covered by the historical login records into multiple interface design sub-periods, obtain each click event involved in the historical login records, perform an occurrence probability analysis on the click event in each interface design sub-period, and obtain multiple event transition probability matrices based on the analysis results to obtain user historical operation collection data; Step S2: Perform real-time operation prediction on the target user based on the user's historical operation data, and obtain predicted click events based on the prediction results; Step S3: Divide the user interface into several interface sub-areas, obtain the user's visual focus point, perform relative position analysis and relative distance analysis between each interface sub-area and the user's visual focus point, and mark the predicted click area based on the analysis results; Step S4: Periodically monitor the number of clicks on the predicted click area, and perform user operation control based on the monitoring results.

Images