A software user experience evaluation method and system based on comprehensive data analysis

Abstract

The application discloses a kind of based on comprehensive data analysis's software user experience evaluation method and system, it is related to human-computer interaction technical field, this method is by embedding behavior perception engine in software operation interface, collect the use frequency of user in task path, jump operation frequency, process standard step number and access user total number, and further unified normalization processing is carried out in background server, constructs normalized behavior dataset, calculates path deviation joint index PDI.In based on path deviation joint index PDI joint consideration path selection information entropy and path jump rate mean two dimensions, whether the use of user path exists concentrated deviation or process jump behavior can be effectively identified, and then whether there is forced guidance, process hidden shortcut in system path design is judged.Compared with the evaluation mode of traditional only relying on click heat map or jump rate, the present method realizes higher dimensional path behavior modeling and deviation identification ability.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of human-computer interaction, in particular to a software user experience evaluation method and system based on comprehensive data analysis. BACKGROUND

[0002] The present application belongs to the technical field of human-computer interaction and intelligent software evaluation, and particularly relates to a software user experience quantitative evaluation method which is fused with user behavior collection, path selection modeling and multi-dimensional risk analysis. With the increasing complexity of software system functions, especially in multi-process and multi-path operation scenarios such as approval flow systems, online questionnaire systems and e-commerce ordering systems, when a user completes a task goal, there are often multiple feasible paths and different interaction modes. Precise and dynamic data modeling of the selection and use behavior of these interaction paths is the key to realizing software interaction optimization and intelligent intervention of user experience.

[0003] However, at the present stage, the existing software user experience evaluation methods mostly stay in post-investigation, such as questionnaires, ratings, coarse-grained indicators such as click volume, page dwell time, or heat map visualization stage, and lack of path-level deviation behavior modeling and logical structure identification of interaction abnormalities. Especially in software with multiple path completion mechanisms, users are often strongly guided by certain default recommended paths or system prompt processes, resulting in extremely concentrated use of paths, while other paths are almost unused, leading to unbalanced data feedback and hidden problems difficult to be discovered. At the same time, the behavior of users jumping steps, reworking input or misoperation in the path is also difficult to be accurately classified and responded, causing the system to only make delayed or rough adjustments in the later stage, affecting the interaction effect.

[0004] This non-uniformity in path design will essentially lead to a lack of freedom in user perception. Although users seem to have multiple options, they are actually forced to be guided or misled by system design, ultimately forming a main path that is excessively concentrated, while other paths are difficult to be effectively optimized or errors are difficult to be identified due to lack of use data; at the same time, since the system cannot timely discover behaviors such as users frequently skipping steps, repeatedly reworking or misoperating controls, it may cause negative experience indicators such as increased user cognitive load, increased task interruption rate and abnormally shortened page dwell time. This not only reduces user satisfaction, but also affects subsequent system recommendations, model optimization and version iteration decisions in the data accumulation process, leading the entire interaction system to fall into a vicious cycle of structural experience bias. SUMMARY

[0005] In view of the deficiencies of the prior art, the present application provides a software user experience evaluation method and system based on comprehensive data analysis, which solves the problems mentioned in the background art.

[0006] To achieve the above object, the present application is realized by the following technical solutions: comprising the following steps:

[0007] S1, by embedding behavior perception engine in the software operation interface, real-time collection of user experience behavior data, and transmitting the user experience behavior data to the software background server;

[0008] S2, pre-processing the user experience behavior data in the software background server, obtaining the normalized behavior data set, and constructing the behavior database, data storage of the normalized behavior data set;

[0009] S3, extracting the normalized behavior data set, calculating the output path deviation joint index PDI, and preliminarily comparing and evaluating the deviation interval threshold and the path deviation joint index PDI;

[0010] S4, triggering the risk identification mechanism based on the preliminary comparison and evaluation result, extracting the path risk data, and calculating the output behavior risk index PRLV based on the path risk data;

[0011] S5, based on the path deviation joint index PDI and the behavior risk index PRLV, comprehensive calculation, analysis of the path experience risk total index RRDV, and setting the path risk interval threshold and the path experience risk total index RRDV for secondary comparison and evaluation, and generating intervention measures.

[0012] Preferably, the S1 comprises S11 and S12;

[0013] S11, by embedding behavior perception engine in the software operation interface, the user logs in the software and authorizes the behavior perception, the behavior perception engine is automatically started, the user experience behavior data is collected in real time, and all the user experience behavior data collected by the behavior perception engine is collected into the log collector through the setting API application program interface;

[0014] The behavior perception engine comprises a path use frequency engine, a nonlinear jump operation engine, a path standard step number obtainer and a server engine;

[0015] The user experience behavior data comprises the use frequency U of the i-th path i , the nonlinear jump operation frequency J of the i-th path i , the total number of normal steps T of the i-th path i And the total number of users N;

[0016] The total number of users N is collected by the server engine of the software backend in real time;

[0017] S12, set the wireless communication interface of the software background server, and set the 5G wireless communication network, wirelessly connect the wireless communication interface with the API collection application interface, access the log collector of the software backend in real time, and extract the user experience behavior data obtained in real time.

[0018] Preferably, the S2 comprises S21 and S22.

[0019] S21, receiving user experience behavior data in real time in the software background server, and preprocessing the user experience behavior data in the software background server to obtain a normalized behavior data set;

[0020] The preprocessing comprises timestamp verification and data normalization.

[0021] The timestamp verification is performed by unifying the user experience behavior data into Unix timestamps.

[0022] The data normalization is performed by using a proportional normalization method to calculate the difference between the original value and the lower limit value of each parameter in the user experience behavior data after timestamp verification, and then calculating the difference between the upper limit value and the lower limit value, and calculating the ratio of the two differences, so as to normalize all parameters in the user experience behavior data to the interval [0, 1] and eliminate the dimensional influence of all parameters.

[0023] S22, configuring an original behavior event table and a processing data table in the behavior database, writing the normalized behavior data set obtained after preprocessing into the behavior database, storing it into the processing data table, and writing the user experience behavior data into the original behavior event table for data storage.

[0024] Preferably, the S3 comprises S31 and S32.

[0025] S31, extracting the normalized behavior data set in the behavior database to calculate and output a path deviation joint index PDI to measure the concentration and path deviation of the user path selection behavior.

[0026] The path deviation joint index PDI is calculated and output by the following algorithm formula.

[0027] ;

[0028] In the formula, n represents the total number of paths, and log represents a logarithmic function.

[0029] Preferably, S32 calculates the mean value by calculating the historical path deviation joint index PDI, and extracts the 85th percentile and the 20th percentile to set the deviation interval threshold, which includes the deviation upper threshold F1 and the deviation lower threshold F2. The deviation interval threshold is compared with the real-time acquired path deviation joint index PDI to evaluate the user's path deviation, and the specific evaluation content is as follows:

[0030] When the path deviation joint index PDI is greater than the deviation upper threshold F1, it indicates that the access path deviates abnormally, and the risk identification mechanism is triggered at this time;

[0031] When the deviation lower threshold F2 is less than the path deviation joint index PDI and the path deviation joint index PDI is less than or equal to the deviation upper threshold F1, it indicates that the current user access path deviates, and the acquisition frame rate of the behavior perception engine is increased by 50% at this time;

[0032] When the path deviation joint index PDI is less than or equal to the deviation lower threshold F2, it indicates that the current software user access experience is stable and the path distribution is balanced, and no intervention is performed at this time.

[0033] Preferably, the S4 includes S41 and S42;

[0034] S41, after triggering the risk identification mechanism in the preliminary comparison and evaluation, the API log application interface of the software system operation log is integrated and connected with the software background server, the path risk data is acquired in real time from the software system operation log, and the path risk data is normalized to eliminate the dimension influence of the path risk data;

[0035] The path risk data includes the total number of i-path interruption behaviors Zd i , the number of i-path rework behaviors Fg i , the number of i-path misoperation times W i , and the average operation time D of the i-path user i .

[0036] Preferably, S42, based on the extracted path risk data, the behavior risk index PRLV is calculated and output, which measures the risk state of the path when the current user deviates abnormally in accessing the software path;

[0037] The behavior risk index PRLV is calculated and output by the following algorithm formula;

[0038] ;

[0039] In the formula, PRLV i The risk index of the i-path.

[0040] Preferably, the S5 includes S51 and S52;

[0041] S51, based on the behavior risk index PRLV of all paths, combined with the path deviation joint index PDI, the path experience risk total index RRDV is calculated and output, which measures the behavior risk of the software in the case of path deviation;

[0042] The path experience risk total index RRDV is calculated and output by the following algorithm formula;

[0043] .

[0044] Preferably, S52, based on the calculation of the path experience risk total index RRDV in all historical periods, the mean and standard deviation are calculated to set the path risk interval threshold, which includes the risk upper limit threshold R1 and the risk lower limit threshold R2,

[0045] The risk upper limit threshold R1 is set by adding 75% of the standard deviation to the mean;

[0046] The risk lower limit threshold R2 is set by adding 35% of the standard deviation to the mean;

[0047] The path risk interval threshold is compared and evaluated again with the real-time acquired path experience risk total index RRDV, the risk state in the user interaction process is judged, and the level is divided, and the corresponding intervention measures are executed. The specific evaluation content is as follows:

[0048] When the path experience risk total index RRDV is greater than the risk upper limit threshold R1, the overall path of the current software is divided into a first risk at this time, and the first intervention measure is executed, which disables the current software path and prompts the software path reconstruction;

[0049] When the risk lower limit threshold R2 is less than the path experience risk total index RRDV, the overall path risk of the current software is divided into a second risk at this time, and the second intervention measure is executed;

[0050] The second intervention measure automatically identifies the mis-touch area by collecting the software operation interface click heat map and software system operation log, adjusts the spatial layout, and preferentially moves and presents the misoperation control;

[0051] When the path experience risk total index RRDV is less than or equal to the risk lower limit threshold R2, the overall path of the current software is divided into a third risk at this time, and no intervention measure needs to be executed.

[0052] A software user experience evaluation system based on comprehensive data analysis, comprising an experience behavior acquisition module, a data processing module, a path deviation analysis module, a behavior risk analysis module and a comprehensive experience risk evaluation module;

[0053] The experience behavior collection module collects user experience behavior data in real time by embedding a behavior perception engine in the software operation interface, and transmits the user experience behavior data to the software backend server.

[0054] The data processing module preprocesses user experience behavior data in the software backend server to obtain a normalized behavior dataset, constructs a behavior database, and stores the normalized behavior dataset.

[0055] The path deviation analysis module extracts the normalized behavior dataset, calculates and outputs the path deviation joint index (PDI), and performs a preliminary comparison and evaluation with the preset deviation interval threshold and the path deviation joint index (PDI).

[0056] The behavioral risk analysis module triggers a risk identification mechanism based on preliminary comparative assessment results, extracts path risk data, and calculates and outputs the behavioral risk index PRLV based on the path risk data.

[0057] The comprehensive experience risk assessment module performs a comprehensive calculation based on the Path Deviation Index (PDI) and the Behavioral Risk Index (PRLV) to analyze the overall path experience risk index (RRDV). It then sets a path risk interval threshold and performs a secondary comparative assessment with the overall path experience risk index (RRDV) to generate intervention measures.

[0058] This invention provides a method and system for evaluating software user experience based on comprehensive data analysis. It has the following beneficial effects:

[0059] (1) This method embeds a behavior-aware engine into the software interface to collect data on the number of times users use the task path, the number of skip operations, the number of standard steps in the process, and the total number of users accessing the task. This data is then uniformly normalized on the backend server to construct a normalized behavior dataset, and the Path Deviation Joint Index (PDI) is calculated. By considering both path selection information entropy and the mean path skip rate based on the PDI, this method can effectively identify whether there is concentrated deviation or process skipping behavior in user path usage, thereby determining whether there are problems such as forced guidance or hidden shortcuts in the system path design. Compared to traditional evaluation methods that rely solely on click heatmaps or bounce rates, this method achieves higher-dimensional path behavior modeling and deviation identification capabilities.

[0060] (2) The method further starts a risk identification mechanism after the path deviation index PDI exceeds a preset threshold, collects risk factors such as the number of interruption behaviors, the number of rework behaviors, the number of misoperations, and the average operation time consumption in the path, and realizes quantitative evaluation of the risk level of each path through the construction of the behavior risk index PRLV. The behavior risk index PRLV formula comprehensively considers the behavior intensity and the operation time compensation effect, realizes the expression of the behavior risk density per unit time, and has the adaptive detection capability for high-frequency rework and mis-touch paths. Further, the path experience risk total index RRDV calculated by combining the behavior risk index PRLV and the path deviation index PDI can not only form a unified risk stratification for all paths in the software, but also provide hierarchical decision support for subsequent interaction structure adjustment and control level intervention, and solve the problems of fuzzy risk identification and uncontrollable response strategy in the prior art.

[0061] (3) The method realizes three-level classification of the risk state in the user interaction process by setting the path risk interval threshold, dynamically comparing the path experience risk total index RRDV index result with the historical behavior data, and triggering the first intervention strategy for the path higher than the risk upper limit threshold R1, including path disabling and structure reconstruction suggestion; triggering the second intervention strategy for the path between the path risk interval threshold, such as spatial layout adjustment based on the heat map, control rearrangement and operation guidance optimization; and the path lower than the risk lower limit threshold R2 does not need intervention and maintains the current structure. The above intervention measures are generated based on real-time risk evaluation and have the characteristics of self-adaptation, closed loop and multi-level intervention, which can effectively improve the user experience performance of the software system in the complex multi-path scene, reduce the mis-touch rate, interruption rate and rework rate, and significantly enhance the system version optimization capability and user satisfaction. BRIEF DESCRIPTION OF DRAWINGS

[0062] Figure 1 It is a software user experience evaluation method based on comprehensive data analysis according to the present application;

[0063] Figure 2 It is a software user experience evaluation system flowchart based on comprehensive data analysis according to the present application;

[0064] Figure 3 It is a risk index PRLV risk value distribution diagram for each path. DETAILED DESCRIPTION

[0065] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0066] Embodiment 1: see Figure 1 and Figure 3 The application provides a software user experience evaluation method based on comprehensive data analysis, in order to achieve the above purposes, the application is realized through the following technical scheme: comprising the following steps:

[0067] S1, by embedding a behavior perception engine in the software operation interface, real-time collection of user experience behavior data, and transmitting the user experience behavior data to the software background server;

[0068] S2, pre-processing the user experience behavior data in the software background server, obtaining the normalized behavior data set, and constructing a behavior database, and storing the normalized behavior data set;

[0069] S3, extracting the normalized behavior data set, calculating the output path deviation joint index PDI, and preliminarily comparing and evaluating the deviation interval threshold value and the path deviation joint index PDI;

[0070] S4, triggering the risk identification mechanism based on the preliminary comparison and evaluation result, extracting the path risk data, and calculating the behavior risk index PRLV based on the path risk data;

[0071] S5, based on the path deviation joint index PDI and the behavior risk index PRLV, the path experience risk total index RRDV is calculated and analyzed, and the path risk interval threshold value and the path experience risk total index RRDV are compared and evaluated twice, and the intervention measures are generated.

[0072] In this embodiment, the method embeds a multi-type behavior perception engine into the software interface to collect key behavioral parameters in real time, including path usage frequency, non-linear jumps, and task completion status. The data is then uploaded to the server via a 5G communication interface, enabling real-time capture of user interaction behavior. The raw data is then synchronized and normalized to construct a unified-format behavior database, providing a high-quality data foundation for subsequent analysis. Furthermore, by constructing a Path Deviation Joint Index (PDI), the concentration of path selection distribution and jump rate changes are accurately characterized to identify whether there are guidance skew or path shortcut issues in the software process design. When the PDI exceeds the deviation threshold, a risk identification mechanism is triggered, collecting path risk data and calculating the Performance Risk Level (PRLV) to quantify the risk level of each path. Finally, the PDI and PRLV are comprehensively analyzed to output the Path Experience Risk Total Index (RRDV). Based on the PDI's value falling within different risk ranges, graded intervention strategies are automatically generated, including path reconstruction suggestions, control relocation, process decomposition, and visual feedback optimization. By implementing this method, potential user experience risks in multi-path processes can be dynamically identified, responded to in real time, and adaptively optimized without interrupting user task execution. Compared with traditional user experience evaluation methods that rely on static heatmaps or questionnaire ratings, this invention can detect process design defects earlier and more accurately, improve the software's ability to automatically identify and handle abnormal path behaviors, significantly enhance the software's path flexibility design capabilities, user behavior fault tolerance, and overall experience stability, ultimately achieving the technical effects of improving task completion rates, reducing rework and misoperations, and improving user satisfaction and system reliability.

[0073] Example 2: Please refer to Figure 1 Specifically: S1 includes S11 and S12;

[0074] S11. By embedding a behavior perception engine within the software's interface, the behavior perception engine is automatically activated after the user logs in and authorizes the behavior perception function. It collects user experience behavior data in real time and collects all user experience behavior data collected by the behavior perception engine into the log collector by setting up an API to collect application interface.

[0075] The behavior-aware engine includes a path usage count engine, a non-linear jump operation engine, a path standard step count acquirer, and a server-side engine.

[0076] User experience behavior data includes the number of times the i-th path is used (U). i The number of nonlinear jump operations J on the i-th path i The total number of normal steps T of the i-th path i And the total number of users N;

[0077] Usage times of the i-th path U i By embedding the path usage times engine on the task completion button and the submission event in the operation interface of the software, it is collected which path the user selects and whether the path task is successfully completed;

[0078] Non-linear jump operation times of the i-th path J i By embedding the non-linear jump operation engine on each step control in the operation interface of the software, the click order queue is recorded in real time;

[0079] Total number of normal steps of the i-th path T i By configuring the standard path process in the software backend, using the path standard step number obtainer in the backend, loading the user access path process, extracting the normal step process, and registering the defined standard process nodes for each path when entering;

[0080] The total number of users N is collected in real time by the server engine in the software backend;

[0081] S12, set the wireless communication interface of the software background server, set the 5G wireless communication network, wirelessly connect the wireless communication interface with the API collection application interface, access the log collector in the software backend in real time, and extract the user experience behavior data obtained in real time.

[0082] In this embodiment, the method constructs a front-end and back-end linkage mechanism that can collect, transmit and manage user experience behavior data in real time by embedding multiple types of behavior perception engines in the software operation interface. Among them, the path use frequency engine is used to record the task completion frequency of each path, the nonlinear jump operation engine is used to track the step skipping and path bypassing behavior in the user interaction process, the path standard step number acquirer provides the system preset process node sequence, and the server engine is used to dynamically count the total number of current access users, which constitutes the basic parameter set required for path behavior modeling. All engine collected data is transmitted to the log collector through the configured API collection interface, and further relies on the 5G wireless communication network to realize high-speed and low-delay data upload through the wireless communication interface of the software background, ensuring the integrity and timeliness of the collected behavior. Through the above implementation means, not only the high-precision, low-interference full-link collection of user path selection and behavior characteristics is realized, but also the real-time and uniformity of data collection is ensured, providing a complete, continuous and structured data basis for subsequent behavior modeling and path deviation analysis. Compared with the prior art, the method can effectively avoid the problems of traditional log lag, incomplete sampling and behavior omission, greatly improving the accuracy and usability of user behavior data. Finally, this way provides bottom support for building a software process evaluation mechanism with high reliability and multi-path dynamic optimization, significantly enhances the automatic perception ability and real-time response ability of the system in user experience feedback, and provides a key basis for realizing data-driven UI structure improvement and process redesign.

[0083] Embodiment 3: see Figure 1 , in particular: S2 includes S21 and S22;

[0084] S21, real-time receiving user experience behavior data in the software background server, and pre-processing the user experience behavior data in the software background server to obtain a normalized behavior data set;

[0085] The preprocessing includes timestamp verification and data normalization;

[0086] The timestamp verification is performed by unifying the user experience behavior data into Unix timestamp;

[0087] The data normalization is performed by using the proportional normalization method to calculate the difference between the original value and the lower limit value of each parameter in the user experience behavior data after timestamp verification, and then calculating the difference between the upper limit value and the lower limit value, and calculating the ratio of the two differences, so that all parameters in the user experience behavior data are normalized to the interval [0, 1], eliminating the dimensional influence of all parameters;

[0088] S22, configure the original behavior event table and the processing data table in the behavior database, write the normalized behavior data set obtained after preprocessing into the behavior database, store into the processing data table, write the user experience behavior data into the original behavior event table, and perform data storage.

[0089] In this embodiment, the method realizes the standardized preprocessing and structured storage management of the user experience behavior data by establishing a double-layer data processing mechanism in the software background server. Specifically, all the user experience behavior data is standardized to Unix timestamps by unifying the timestamp format, ensuring that the data acquisition results under different terminals and different operating system environments have consistent time sequence characteristics. Then, the proportion normalization method is used to perform dimensionless processing on each behavior parameter, such as path use frequency and jump frequency, and uniformly map the numerical range to the interval [0, 1], effectively eliminating the dimensional differences between parameters and improving the accuracy and comparability of subsequent model calculation. By configuring the original behavior event table and the processing data table in the behavior database, a hierarchical storage structure is established, so that the system can quickly extract and analyze the normalized data set while retaining the traceability of the original data, balancing data integrity and analysis efficiency. The introduction of this implementation manner significantly improves the processing capability of large-scale, multi-dimensional user behavior data, ensures the high quality and uniformity of algorithm input data, and provides a solid foundation for the accurate calculation of the path deviation index PDI and the risk level index PRLV. Compared with the existing data processing method, the present scheme effectively reduces data redundancy and information bias by normalization calibration and structured storage, improves the recognition sensitivity of the model to abnormal behavior, and finally enhances the data analysis accuracy and processing stability of the system in software user experience evaluation, promoting the seamless connection between user behavior perception and strategy intervention.

[0090] Embodiment 4: Please refer to Figure 1 , specifically: S3 includes S31 and S32;

[0091] S31, extract the normalized behavior data set in the behavior database, calculate and output the path deviation joint index PDI, and measure the concentration and path deviation of the user path selection behavior;

[0092] The path deviation joint index PDI is calculated and output by the following algorithm formula:

[0093] ;

[0094] In the formula, n represents the total number of paths, and log represents the logarithmic function.

[0095] Path selection entropy, which is a normalized form of information entropy, measures the distribution balance of user path selection. High entropy close to 1 indicates that the path usage is more balanced, and low entropy close to 0 indicates that most users are concentrated on a few paths.

[0096] Average jump rate, The jump rate of each path reflects whether the user performs tasks according to the system design process. Higher jump rate indicates more interaction jumps in the path, which may be caused by user confusion, complex interface, unclear guidance, etc. Finally, the average represents the user process compliance at the system level.

[0097] The product of the two reflects whether the user's path selection behavior is concentrated and whether the user's behavior deviates in the path. Essentially, it is a quantitative expression of the deviation of path structure design.

[0098] The path deviation joint index PDI is used to evaluate whether the user's selection of multiple paths in the task process is concentrated and whether it frequently deviates from the system's preset standard process. Essentially, it is used to determine whether the process structure is reasonable and the guidance is clear.

[0099] Path selection entropy is a normalized standard information entropy, which measures the concentration of selection. The unit is "bit", but since it is a normalized ratio, the result is [0, 1] interval without unit.

[0100] The average jump rate is a non-linear jump rate, with a dimensionless ratio, measuring behavior deviation.

[0101] The product of the two is a dimensionless number, avoiding the problem of dimension stacking.

[0102] S32, calculate the average of the historical path deviation joint index PDI, and extract the 85th percentile and 20th percentile as the deviation interval threshold. The deviation interval threshold includes the upper deviation threshold F1 and the lower deviation threshold F2. Compare the deviation interval threshold with the real-time path deviation joint index PDI to evaluate the user's path deviation. The specific evaluation content is as follows:

[0103] When the path deviation joint index PDI is greater than the upper deviation threshold F1, it indicates that the access path deviates abnormally, the path selection is too concentrated, and the operation jump is obvious. At this time, the risk identification mechanism is triggered.

[0104] When the lower deviation threshold F2 is less than the path deviation joint index PDI and the upper deviation threshold F1, it indicates that the current user access path deviates. At this time, the behavior perception engine's acquisition frame rate is increased by 50%.

[0105] When the path deviates from the joint index PDI ≤ deviation lower threshold F2, it indicates that the current software user access experience is stable, and the path distribution is balanced, and no intervention is performed.

[0106] In this embodiment, the method establishes a structural deviation analysis means for user path selection behavior by constructing a computer mechanism of path deviation joint index PDI. Specifically, by extracting normalized data in the behavior database, a joint calculation model is constructed by using information entropy and jump rate, on the basis of measuring the balance degree of user path distribution, the deviation factor of internal interaction behavior is further fused, so that the system can not only identify the concentration trend of path selection, but also quantify whether the internal process of the path is frequently jumped and bypassed, thereby realizing comprehensive evaluation of the rationality of path design and the consistency of user interaction. On this basis, a historical behavior statistical model is introduced to dynamically generate a deviation threshold interval, wherein the deviation interval threshold includes a deviation upper threshold F1 and a deviation lower threshold F2, and the real-time path deviation joint index PDI value is compared with the interval to judge the stability, deviation trend or abnormal concentration of the current user path behavior, and finally drive the collection density adjustment or risk identification mechanism. The introduction of this implementation manner significantly improves the recognition ability of the path deviation mode, effectively solves the technical blind spot of the traditional user behavior analysis method that "only looks at the number of clicks, but does not look at the path structure behavior characteristics". By converting the path behavior into a quantifiable structural deviation index and combining the dynamic threshold judgment strategy, the systematic early warning of potential process bottlenecks, hidden shortcuts or forced guide design is realized. Compared with the existing static index system, this method has higher real-time, sensitivity and structural analysis depth, and provides a pre-supporting basis for subsequent risk level identification and intervention strategy generation, and improves the user process design quality and interaction behavior transparency of the software system as a whole.

[0107] Embodiment 5: Please refer to Figure 1 , specifically: S4 includes S41 and S42;

[0108] S41, after triggering the risk identification mechanism by preliminary comparison and evaluation, the API log application program interface of the software system operation log is integrated and connected with the software background server, the path risk data is acquired in real time from the software system operation log, and the path risk data is subjected to data normalization processing to eliminate the dimension influence of the path risk data;

[0109] The path risk data includes the total number of i-path interruption behaviors Zd i , the number of i-path rework behaviors Fg i , the number of i-path misoperation W i , and the average user operation time D of the i-path i .

[0110] The total number of path interruption behaviors of the i-th path Zd i By identifying whether the user completes the target page, such as no end event and session end as interruption;

[0111] The number of times of rework behaviors of the i-th path Fg i By triggering an update event through the user clicking the previous step or modifying the input field, the difference before and after modification is recorded, and rework can be refined into repeated modification, submission failure rollback, etc.

[0112] The number of times of misoperation of the i-th path W i By collecting the DOM element attributes of the click event through the software system operation log to determine whether it is misoperation; such as clicking non-interactive elements, repeatedly clicking the submit button, clicking cancel immediately after clicking, etc., and it can also be collected through the frequency of triggering error message pop-up windows;

[0113] The average operation time consumption D of the i-th path user i By recording the time stamps of the start node and the completion node through the software system operation log when each user accesses the i-th path, the difference between the two is the time consumption of the user, and the average of the i-th path is obtained.

[0114] S42, based on the extracted path risk data, a ratio calculation is performed to output a behavior risk indicator PRLV, which measures the risk state of the path when the user deviates from the abnormality in accessing the software path;

[0115] The behavior risk indicator PRLV is calculated and output by the following algorithm formula;

[0116] ;

[0117] In the formula, PRLV i The risk indicator of the i-th path;

[0118] The numerator part represents the total negative interaction behavior of the user on the i-th path, that is, the weighted set of risk factors, and the higher the value, the more serious the potential risk of the path;

[0119] The denominator part introduces the operation time consumption factor, with two purposes:

[0120] Compensation effect: if the operation time consumption is very long, it means that the user is willing to invest time, which may partially buffer the negative effects of high-risk behaviors;

[0121] Time pressure punishment mechanism: if the operation time consumption is abnormally short, it may be used hastily, and there is a high risk of misoperation;

[0122] The risk indicator PRLV is a behavior risk density in a unit of time, which can also be understood as the number of abnormal interaction behaviors generated per second, measures the frequency and density of "interaction risk behaviors" in a specific path during user use, such as interruption, rework, misoperation, etc., and further combines operation time to form a risk density in a unit of time because;

[0123] The formula of the risk indicator PRLV is in a proportional form, and the parameters involved are all normalized data parameters. The output result of the risk indicator PRLV is dimensionless.

[0124] In this embodiment, the method further enhances the fine analysis ability after the path deviation behavior by establishing a risk identification and grade quantization mechanism based on the behavior log. Through the connection of the software system operation log API and the background integration, the core risk factor data such as path interruption, rework, misoperation and average time consumption are extracted in real time, and after unified normalization processing, a structured path risk data set is formed. This process realizes the systematic capture and standardized expression of the user's atypical operation behavior, and provides an accurate behavior quantization basis for subsequent risk calculation. Based on the above data, the behavior risk indicator PRLV is constructed, and the ratio calculation method of "total amount of negative behaviors divided by average time consumption of users" is adopted, the time density dimension is integrated, the frequency of user abnormal operation is identified, and the "time compensation effect" is introduced, which takes into account the dual dimensions of behavior intensity and operation rhythm. Through this implementation, the system can automatically enter the risk identification state after the user path deviation is detected, and quickly complete the quantization evaluation of the path risk grade, ensuring that the problem path can be accurately identified and treated differently. Compared with the traditional method of judging risk only by the number of operations, this method significantly improves the identification granularity and context perception ability. The risk indicator PRLV output by the method not only reflects the problem complexity of the path itself, but also reveals the potential cognitive or interaction obstacles behind the user behavior, providing a reliable basis for subsequent intervention decisions. Overall, the method effectively promotes the progressive transformation from "detecting deviation" to "judging risk", and makes the software system more targeted and intelligent in user experience management.

[0125] Embodiment 6: Please refer to Figure 1 and Figure 3 Specifically, S5 includes S51 and S52;

[0126] S51, based on the behavior risk indicator PRLV of all paths, combines the path deviation joint index PDI to perform comprehensive calculation and output the path experience risk total index RRDV, which measures the behavior risk of the software in the case of path deviation;

[0127] The path experience risk total index RRDV is calculated and output by the following algorithm formula;

[0128] ;

[0129] wherein, represents the average value of the behavior risk indicators of all paths, and the two key dimensions are fused in the form of product;

[0130] The path experience risk total index RRDV is the result of the overall risk assessment, which is obtained by combining the path deviation joint index PDI with the behavior risk indicators PRLV of all paths to assess the overall experience problem of the task flow;

[0131] The output results of the behavior risk indicators PRLV and the path deviation joint index PDI are both dimensionless, and the output result of the path experience risk total index RRDV is also dimensionless.

[0132] S52, based on the calculation of the path experience risk total index RRDV in all historical periods, the mean and standard deviation are calculated to set the path risk interval threshold, which includes the risk upper threshold R1 and the risk lower threshold R2,

[0133] The risk upper threshold R1 is set by adding 75% of the standard deviation to the mean;

[0134] The risk lower threshold R2 is set by adding 35% of the standard deviation to the mean;

[0135] After comparing the path risk interval threshold with the real-time path experience risk total index RRDV, the risk state in the user interaction process is judged, and the level is divided, and the corresponding intervention measures are executed. The specific evaluation content is as follows:

[0136] When the path experience risk total index RRDV is greater than the risk upper threshold R1, the overall path of the current software is divided into a first risk at this time, that is, the user experience of the software path design exists obvious deviation and a large number of risk behaviors are generated in the process of user experience, and the first intervention measure is executed. The first intervention measure disables the current software path and prompts the software path reconstruction;

[0137] When the risk lower threshold R2 is less than the path experience risk total index RRDV and the risk upper threshold R1, the overall path risk of the current software is divided into a second risk at this time, that is, the user behavior fluctuation is obvious but the path structure is not out of control, and the second intervention measure is executed;

[0138] The second intervention measure automatically identifies the mis-touch area by collecting the software operation interface click heat map and software system operation log, that is, the mis-touch frequency is greater than the mis-touch threshold, adjusts the spatial layout, and preferentially moves and presents the mis-operation control;

[0139] When the path experience risk total index RRDV≤risk lower threshold R2, the overall path of the current software is divided into three levels of risk at this time, that is, the current overall path experience of the software has a certain deviation, but the operation is risk-free, and no intervention measures need to be performed.

[0140] In this embodiment, the method calculates a composite path experience risk total index RRDV by fusing the two types of risk characteristics through a product form based on the path deviation joint index PDI obtained in the early stage and the behavior risk indicator PRLV of each path, so as to accurately reflect the superimposed risk of "structural imbalance" and "interaction anomaly" in user experience and comprehensively depict the controllability and stability of the current software path design in actual use. This index breaks away from the single-dimensional judgment mechanism and greatly enhances the overall visual expression ability of experience risk. Further based on the fluctuation of the path experience risk total index RRDV in the historical period, a dynamic path risk interval threshold is constructed in a "mean and multi-level standard deviation" manner, so as to form a three-level response intervention mechanism. For a first-level high-risk path, structural disabling and reconstruction prompts are automatically performed; for a second-level risk, a click heat map analysis and control space rearrangement optimization are intelligently triggered; and for a third-level low-risk state, no intervention is performed to avoid false positives and redundant adjustments. This risk level-based dynamic response mechanism significantly improves the precision, adaptability and practicality of the intervention strategy. Through the implementation of this step, not only can the path deviation and behavior risk be quantitatively fused, but also the threshold model can be combined to actively trigger operable intervention suggestions, realizing a closed-loop processing flow from "risk identification" to "experience optimization". The final effect is that the user operation fluency is improved, the false touch rate is reduced, the path structure is more flexible and reasonable, and the continuous improvement of software interaction experience and the scientific decision support for design iteration are effectively promoted.

[0141] Embodiment 7: Please refer to Figure 1 and Figure 2 A software user experience evaluation system based on comprehensive data analysis, comprising an experience behavior acquisition module, a collected data processing module, a path deviation analysis module, a behavior risk analysis module and a comprehensive experience risk evaluation module;

[0142] The experience behavior acquisition module acquires user experience behavior data in real time by embedding a behavior perception engine in the software operation interface, and transmits the user experience behavior data to the software background server.

[0143] The collected data processing module preprocesses the user experience behavior data in the software background server, obtains a normalized behavior data set, constructs a behavior database, and stores the normalized behavior data set.

[0144] The path deviation analysis module extracts the normalized behavior data set, calculates the output path deviation joint index PDI, and preliminarily compares and evaluates the deviation interval threshold value with the path deviation joint index PDI;

[0145] The behavior risk analysis module triggers the risk identification mechanism based on the preliminary comparison and evaluation result, extracts the path risk data, and calculates and outputs the behavior risk indicator PRLV based on the path risk data;

[0146] The comprehensive experience risk evaluation module comprehensively calculates based on the path deviation joint index PDI and the behavior risk indicator PRLV, analyzes the path experience risk total index RRDV, sets the path risk interval threshold value for the second comparison and evaluation with the path experience risk total index RRDV, and generates the intervention measures.

[0147] Although the embodiments of the present application have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the present application.

Claims

1. A software user experience evaluation method based on comprehensive data analysis, characterized in that: Includes the following steps: S1. By embedding a data acquisition listener in the software operation interface, user experience behavior data is collected in real time and transmitted to the software backend server; The user experience behavior data includes the number of times the i-th path is used (U). i The number of nonlinear jump operations J on the i-th path i The total number of normal steps T of the i-th path i And the total number of users N; S2. In the software backend server, preprocess the user experience behavior data to obtain a normalized behavior dataset, build a behavior database, and store the normalized behavior dataset. S3. Extract the normalized behavior dataset, calculate the output path deviation joint index PDI, and perform a preliminary comparison and evaluation with the preset deviation interval threshold and the path deviation joint index PDI. S3 includes S31 and S32; S31. Extract the normalized behavior dataset from the behavior database, calculate and output the Path Deviation Joint Index (PDI), and measure the concentration and path deviation of user path selection behavior. The Path Deviation Joint Index (PDI) is calculated and output using the following algorithm formula; ； In the formula, n represents the total number of paths, and log represents the logarithmic function; S32. The mean value is calculated by calculating the historical path deviation index (PDI), and 85% and 20% of the points are extracted as deviation interval thresholds. The deviation interval thresholds include an upper deviation threshold F1 and a lower deviation threshold F2. The deviation interval thresholds are then compared with the real-time path deviation index (PDI) for preliminary evaluation to determine the user's path deviation. S4. Based on the preliminary comparative assessment results, trigger the risk identification mechanism, extract path risk data, and calculate and output the behavioral risk indicator PRLV based on the path risk data; The path risk data includes the total number of interruption behaviors on the i-th path, Zd. i The number of rework actions on the i-th path, Fg i The number of erroneous operations W on the i-th path i The average user operation time D for the i-th path i S4 includes S42; S42. Calculate the ratio based on the extracted path risk data and output the behavioral risk index PRLV to measure the risk status of the path when the current user deviates from the path when accessing the software. The behavioral risk indicator PRLV is calculated and output using the following algorithm formula; ； In the formula, PRLV i Behavioral risk indicators for the i-th path; S5. Based on the Path Deviation Index (PDI) and the Behavioral Risk Index (PRLV), a comprehensive calculation is performed to analyze the Path Experience Risk Index (RRDV). A path risk interval threshold is set and the path experience risk index (RRDV) is compared and evaluated for the second time, and intervention measures are generated. S5 includes S51 and S52; S51. Based on the behavioral risk index PRLV of all paths, combined with the path deviation joint index PDI, a comprehensive calculation is performed to output the path experience risk index RRDV, which measures the behavioral risk of the software under path deviation. S52. Based on the calculation of the total path experience risk index RRDV over all historical periods, the mean and standard deviation are calculated to set the path risk interval threshold. The path risk interval threshold is then compared and evaluated with the real-time obtained total path experience risk index RRDV to determine the risk status during user interaction, classify the risk level, and implement corresponding intervention measures.

2. The software user experience evaluation method based on comprehensive data analysis according to claim 1, characterized in that: S1 includes S11 and S12; S11. By embedding a data collection listener in the software's operation interface, user experience behavior data is collected in real time. By setting up an API to collect application interface, all user experience behavior data collected by the data collection listeners is collected into a log collector. The data acquisition listener includes a path usage count listener, a non-linear jump operation listener, a path standard step count acquirer, and a server-side listener. The total number of users N is monitored in real time by the server-side listener on the software backend. S12. Configure the wireless communication interface of the software backend server and set up a 5G wireless communication network. Connect the wireless communication interface to the API collection application interface wirelessly to access the log collector of the software backend in real time and extract the real-time user experience behavior data.

3. The software user experience evaluation method based on comprehensive data analysis according to claim 2, characterized in that: S2 includes S21 and S22; S21. Receive user experience behavior data in real time in the software backend server, and preprocess the user experience behavior data in the software backend server to obtain a normalized behavior dataset. The preprocessing includes timestamp verification and data normalization; The timestamp verification unifies user experience behavior data into Unix timestamps; The data normalization is achieved by using a proportional normalization method on the user experience behavior data after timestamp verification. The difference between the original value and the lower limit value of each parameter in the user experience behavior data is calculated, and then the difference between the upper limit value and the lower limit value is calculated. The ratio of the two differences is then calculated to normalize all parameters in the user experience behavior data to the [0, 1] interval, eliminating the influence of the dimensions of all parameters. S22. Configure the original behavior event table and the processing data table in the behavior database. Write the normalized behavior dataset obtained after preprocessing into the behavior database and store it in the processing data table. Write the user experience behavior data into the original behavior event table for data storage.

4. The software user experience evaluation method based on comprehensive data analysis according to claim 1, characterized in that: When the path deviation index PDI is greater than the upper limit threshold F1, it indicates that the access path deviation is abnormal, and the risk identification mechanism is triggered at this time. When the deviation lower threshold F2 < path deviation joint index PDI ≤ deviation upper threshold F1, it indicates that the current user's access path has deviated, and the collection frame rate of the collection listener will be increased by 50%. When the path deviation index PDI is less than or equal to the lower deviation threshold F2, it indicates that the current user experience of the software is stable and the path distribution is balanced, and no intervention is required.

5. The software user experience evaluation method based on comprehensive data analysis according to claim 1, characterized in that: S4 includes S41; S41. After the initial comparative assessment triggers the risk identification mechanism, the API log application interface of the software system operation log is integrated with the software backend server to collect path risk data from the software system operation log in real time, and the path risk data is normalized to eliminate the influence of the dimension of the path risk data.

6. The software user experience evaluation method based on comprehensive data analysis according to claim 1, characterized in that: The overall risk index (RRDV) of the path experience is calculated and output using the following algorithm formula; 。 7. The software user experience evaluation method based on comprehensive data analysis according to claim 1, characterized in that: The path risk interval thresholds include an upper risk threshold R1 and a lower risk threshold R2. The risk ceiling threshold R1 is set based on the mean plus 75 percent of the standard deviation; The lower risk threshold R2 is set based on the mean plus 35 percent of the standard deviation; The specific assessment content is as follows; When the total path experience risk index RRDV is greater than the risk upper limit threshold R1, the overall path of the current software is classified as a level 1 risk, and the first intervention measure is executed. The first intervention measure is to disable the current software path and prompt the software path to be reconstructed. When the lower risk threshold R2 < the total path experience risk index RRDV ≤ the upper risk threshold R1, the overall path risk of the current software is classified as a secondary risk, and the second intervention measure is implemented. The second intervention measure automatically identifies areas of accidental touch by collecting click heatmaps of the software operation interface and software system operation logs, adjusts the spatial layout, and prioritizes moving and presenting erroneously operated controls forward and in stages. When the total path experience risk index RRDV is less than or equal to the lower risk threshold R2, the overall path of the current software is divided into three levels of risk, and no intervention measures are required.

8. A software user experience evaluation system based on comprehensive data analysis, applied to the software user experience evaluation method based on comprehensive data analysis as described in any one of claims 1-7, characterized in that: It includes an experience behavior monitoring module, a monitoring data processing module, a path deviation analysis module, a behavior risk analysis module, and a comprehensive experience risk assessment module; The user experience behavior monitoring module collects user experience behavior data in real time by embedding a data acquisition and monitoring device in the software operation interface, and transmits the user experience behavior data to the software backend server. The monitoring data processing module preprocesses user experience behavior data in the software backend server to obtain a normalized behavior dataset, constructs a behavior database, and stores the normalized behavior dataset. The path deviation analysis module extracts a normalized behavior dataset, calculates and outputs the path deviation joint index (PDI), and performs a preliminary comparison and evaluation with a preset deviation interval threshold and the path deviation joint index (PDI). The behavioral risk analysis module triggers a risk identification mechanism based on preliminary comparative assessment results, extracts path risk data, and calculates and outputs the behavioral risk index PRLV based on the path risk data. The comprehensive experience risk assessment module performs a comprehensive calculation based on the Path Deviation Index (PDI) and the Behavioral Risk Index (PRLV) to analyze the overall path experience risk index (RRDV). It then sets a path risk interval threshold and performs a secondary comparative assessment with the overall path experience risk index (RRDV) to generate intervention measures.

Images