User operation behavior prediction method and device, terminal device and server

Abstract

The application discloses a user operation behavior prediction method and device, terminal equipment and server. The method comprises the following steps: obtaining user operable function points of a software program to be tested, and constructing a function point directed graph according to the link relationship between the function points; constructing a three-dimensional model of the function point directed graph; calculating the initial concentration of pheromone of each function point; predicting the function points in a test level based on the operation habit of a test case and the initial concentration of pheromone of the function points; marking the connected function points and the path of the connected function points and the function points of the previous level in the three-dimensional model of the function point directed graph according to a marking strategy; updating the pheromone concentration of all the function points in the test level based on the prediction result after the prediction of one or more function points in the test level is completed; and displaying the three-dimensional model of the function point directed graph, the marking of the function points and the marking of the test case test path. The application can predict the operation behavior of the user, and the prediction result is displayed intuitively.

Description

Technical Field

[0001] This invention relates to the field of software program testing technology, specifically to a user operation behavior prediction method, apparatus, terminal device, server, computer-readable storage medium, and computer program product. Background Technology

[0002] With the development of computer technology, internet technology, and application technology, a wide variety of software programs with diverse functions, including those applied to web platforms and mobile terminals, are playing a significant role in various industries. Software development typically requires substantial financial and human resources. To ensure the developed software achieves the desired results, developers usually conduct user market research before starting development. After the software is launched, user interaction data is collected through tracking points in the code. Once a certain number of users have been accumulated, the tracking data is analyzed, and targeted optimization plans or other commercial decisions are made based on the analysis results. For example, adjusting functions, changing interface layouts, and setting up advertising placements, etc.

[0003] However, this method of analyzing user behavior has certain drawbacks. For example, user behavior data can only be obtained after the software program is launched, and the data tracking only becomes representative after a certain number of users have been accumulated, thus requiring a certain amount of time to accumulate user data. Therefore, this method of analyzing user behavior leads to delays in software program optimization and business strategy, not only hindering user growth but also missing business opportunities, ultimately resulting in the failure of software program development. Summary of the Invention

[0004] In view of this, embodiments of the present invention provide a method, apparatus, terminal device, server, computer-readable storage medium, and computer program product for predicting user operation behavior, in order to solve the problem of lag in obtaining user operation behavior data of software programs.

[0005] According to one aspect of the present invention, a method for predicting user operation behavior is provided, comprising the following steps:

[0006] Obtain the user-operable functional points of the software program to be tested, and construct a directed graph of functional points based on the link relationships between the functional points. The directed graph includes multiple levels of functional point sets with logical order. Each level of functional point set includes one or more functional points, and each upper-level functional point corresponds to a lower-level functional point set.

[0007] Construct a three-dimensional model of the directed graph of the functional points, wherein multiple functional points at the same level are located in the same spatial region, and functional points at different levels are located in different spatial regions.

[0008] Calculate the initial pheromone concentration for each function point based on the probability that the function point will be operated by the user.

[0009] In response to the start of testing at the test terminal, the set of functional points with the lowest logical order in the directed graph is taken as the test level; based on the operational habits of the test cases and the initial pheromone concentration of the functional points, one or more functional points in the test level are predicted;

[0010] In response to the prediction of a connected functional point, the set of next-level functional points of the connected functional point is taken as a new test level; based on the operation habits of the test cases and the pheromone concentration of the functional points, one or more functional points in the new test level are predicted until the termination condition is met, and the testing process of the software program ends.

[0011] Specifically, when a connected functional point is predicted, the connected functional point and its path to the previous level connected functional point are marked in the 3D model of the directed graph of functional points according to the marking strategy.

[0012] After predicting one or more function points in the test level, the pheromone concentration of all function points in the test level is updated based on the prediction results; and

[0013] During the testing process, a 3D model of the directed graph of the functional points, along with the markers of the functional points and the markers of the test cases and test paths, is displayed on the screen of the display terminal.

[0014] According to another aspect of the present invention, a user operation behavior prediction method is provided, applied to a server, wherein the method includes the following steps:

[0015] Obtain the user-operable functional points of the software program to be tested, and construct a directed graph of functional points based on the link relationships between the functional points. The directed graph includes multiple levels of functional point sets with logical order. Each level of functional point set includes one or more functional points, and each upper-level functional point corresponds to a lower-level functional point set.

[0016] Construct a three-dimensional model of the directed graph of the functional points, wherein multiple functional points at the same level are located in the same spatial region, and functional points at different levels are located in different spatial regions.

[0017] Calculate the initial pheromone concentration for each function point based on the probability that the function point will be operated by the user.

[0018] In response to the test start request from the test terminal, the directed graph data of the function points of the software program under test and the test case data are sent to the test terminal, and the 3D model of the directed graph data is displayed on the display terminal; wherein, the directed graph data includes multiple levels of function point sets, the mapping relationship between each function point and the lower-level function point sets, and the initial pheromone concentration of each function point; the test case data includes operation habit data; and

[0019] The prediction data sent by the response test terminal includes connected function point identifiers. In the directed graph 3D model of the function points in the display terminal, the corresponding connected function points and their paths to the previous level connected function points are marked according to the marking strategy.

[0020] According to another aspect of the present invention, a user operation behavior prediction method is provided, applied to a test terminal device, comprising the following steps:

[0021] Start the software program to be tested and obtain the directed graph data of the function points and the test case data of the software program to be tested; wherein, the directed graph data of the function points includes a multi-level set of function points with logical order, the correspondence between each function point and the lower-level set of function points, and the initial pheromone concentration of each function point; the test case data includes operation habit data.

[0022] The test level is determined starting from the lowest level, according to the logical order of the directed graph.

[0023] Predict one or more functional points in the test level based on the operational habits of test cases and the pheromone concentration of functional points;

[0024] In response to the prediction of a connected functional point, the set of lower-level functional points corresponding to the connected functional point is taken as a new test level. Based on the operational habits of the test cases and the pheromone concentration of the functional points, predictions are made for one or more functional points in the new test level. The testing process for the software program ends when a termination condition is met.

[0025] During the testing process, the pheromone concentration of all functional points in the test level is updated based on the prediction results.

[0026] According to another aspect of the present invention, a user operation behavior prediction device is provided, applied to a server, comprising a function point directed graph construction module, a model construction module, a pheromone concentration calculation module, a data interaction module, and a display marking module. The function point directed graph construction module is configured to acquire user-operable function points of a software program under test, and constructs a function point directed graph based on the link relationships between function points. The directed graph includes multiple levels of function point sets with logical order, each level of function point set including one or more function points, and each higher-level function point corresponding to a lower-level function point set. The model construction module is connected to the function point directed graph construction module and is configured to construct a three-dimensional model of the function point directed graph based on the function point directed graph of the software program under test, wherein multiple function points at the same level are located in the same spatial region, and function points at different levels are located in different spatial regions. The pheromone concentration calculation module calculates the pheromone concentration based on the function points operated by the user. The initial pheromone concentration for each functional point is calculated. The data interaction module is connected to the test terminal and the display terminal, and is configured to respond to the test start request from the test terminal. It sends the directed graph data of the functional points of the software program under test and the test case data to the test terminal, and sends the 3D model of the directed graph of functional points to the display terminal for display. The display marking module is connected to the data interaction module, and is configured to receive prediction data from the test terminal from the data interaction module. Based on the connected functional point identifier, it marks the corresponding connected functional points and their paths to the previous level connected functional points in the 3D model of the directed graph of functional points in the display terminal according to the marking strategy.

[0027] According to another aspect of the present invention, a user operation behavior prediction device is provided, applied to a terminal device as a test terminal, comprising a data acquisition module, a test process monitoring module, a function point prediction module, a simulated operation module, and a pheromone concentration update module. The data acquisition module is configured to acquire directed graph data of function points and test case data of the software program to be tested. The directed graph data includes a multi-level set of function points with logical order, the correspondence between each function point and its subordinate set of function points, and the initial pheromone concentration of each function point. The test case data includes operation habit data. The test process monitoring module is connected to the data acquisition module and is configured to start the software program to be tested, using the lowest-level set of function points in the directed graph as the starting point for testing. The test level terminates the testing process of the software program when the termination condition is met. The function point prediction module is connected to the test process monitoring module and is configured to predict one or more function points in the test level based on the operation habits of the test cases and the pheromone concentration of the function points. The simulation operation module is connected to both the test process monitoring module and the function point prediction module and is configured to respond to the function point prediction module's prediction of connected function points and perform operations corresponding to the operation type of the connected function points. Correspondingly, the test process monitoring module uses the set of lower-level function points corresponding to the connected function points as the new test level. The pheromone concentration update module is connected to the function point prediction module and is configured to update the pheromone concentration of all function points in the test level based on the prediction results during the test.

[0028] According to another aspect of the present invention, a server is provided, including a processor and a memory storing computer program instructions, wherein the processor, when executing the computer program instructions, implements a user operation behavior prediction method applied to the server.

[0029] According to another aspect of the present invention, a terminal device is provided, including a processor and a memory storing computer program instructions; the processor, when executing the computer program instructions, implements a user operation behavior prediction method applied to the terminal device.

[0030] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein computer program instructions are stored thereon, and when executed by a processor, the computer program instructions implement an application testing method applied to a server, or implement a user operation behavior prediction method applied to a test terminal device.

[0031] According to another aspect of the present invention, a computer program product is provided, comprising computer program instructions, which, when executed by a processor, implement the application testing method applied to a server, or implement the user operation behavior prediction method applied to a test terminal device.

[0032] As can be seen from the foregoing embodiments, the present invention obtains user operating habits based on the operational behavior data of users of other existing applications, and predicts the user's operational behavior of the software program under test based on the idea of ​​ant colony algorithm. This allows developers to determine whether the design of various functions of the software program under test meets expectations, whether adjustments are needed, and what pages and methods of advertising and other commercial behaviors are used in the application based on the predicted user operational behavior. Thus, it solves the problem of the lag in obtaining user operational behavior in existing applications and other negative effects caused by the lag. Furthermore, the results of the user operational behavior prediction of the present invention are displayed intuitively. Attached Figure Description

[0033] To more clearly illustrate the implementation of the embodiments of the present invention, the accompanying drawings of the embodiments of the present invention will be briefly described below.

[0034] Figure 1 This is a schematic diagram of an AR system architecture based on a server and terminal device according to an embodiment of the present invention;

[0035] Figure 2 This is a schematic diagram of a virtual-real image fusion method for AR navigation using a mobile app.

[0036] Figure 3 This is a flowchart of a user operation behavior prediction method according to an embodiment of the present invention;

[0037] Figure 4 This is a schematic diagram of a three-dimensional model of a directed graph of function points according to an embodiment of the present invention;

[0038] Figure 5 This is a flowchart of a method for obtaining test case data according to an embodiment of the present invention;

[0039] Figure 6 This is a flowchart of a user operation behavior prediction method applied to a test terminal according to an embodiment of the present invention;

[0040] Figure 7 This is a flowchart of a method for predicting one or more function points in a test level according to an embodiment of the present invention.

[0041] Figure 8 This is a flowchart of a user operation behavior prediction method applied to a server according to an embodiment of the present invention;

[0042] Figure 9 This is a flowchart of a method for interaction between a test terminal, a display terminal, and a server according to an embodiment of the present invention;

[0043] Figure 10 This is a schematic diagram of functional points in the current visible range of an AR terminal according to an embodiment of the present invention;

[0044] Figure 11 This is a schematic diagram of a user operation behavior prediction device applied to a server according to an embodiment of the present invention;

[0045] Figure 12 This is a schematic diagram of a user operation behavior prediction device applied to a test terminal according to an embodiment of the present invention;

[0046] Figure 13 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present invention; and

[0047] Figure 14 This is a schematic diagram of the software program structure of an exemplary terminal device according to an embodiment of the present invention. Detailed Implementation

[0048] The principles and spirit of the present invention will be described below with reference to several exemplary embodiments. It should be understood that these embodiments are provided to make the principles and spirit of the present invention clearer and more thorough, enabling those skilled in the art to better understand and implement the principles and spirit of the present invention. The exemplary embodiments provided herein are only a part of the embodiments of the present invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments described herein without inventive effort are within the scope of protection of the present invention.

[0049] Those skilled in the art will recognize that embodiments of the present invention can be implemented as a system, apparatus, device, method, computer-readable storage medium, or computer program product. Therefore, the present invention can be specifically implemented in at least one of the following forms: entirely hardware, entirely software, or a combination of hardware and software. According to specific embodiments of the present invention, the present invention claims protection for a user operation behavior prediction method, apparatus, terminal device, server, computer-readable storage medium, and computer program product.

[0050] In this document, terms such as first, second, and third are used only to distinguish one entity (or operation) from another, and are not intended to require or imply any order or relationship between these entities (or operations).

[0051] Embodiments of the present invention can be applied to servers and terminal devices. Please refer to... Figure 1This diagram schematically illustrates an AR system architecture for an AR application, including a server 10 and several terminal devices 20. In some examples, the terminal devices 20 are AR devices, which can be dedicated AR devices such as head-mounted displays (HMDs), smart gloves, clothing, and other smart wearable electronic devices. In other examples, the terminal devices 20 can be general-purpose AR devices, such as mobile phones, laptops, tablets, virtual reality (VR) devices, in-vehicle devices, navigation devices, gaming devices, etc.

[0052] Taking AR helmets or AR glasses as an example, a head-mounted display, machine vision system, and mobile computer can be integrated into a wearable device. This device has a display resembling glasses and is worn on the user's head. It transmits augmented reality information to the display or projects it onto the user's eyes, enhancing the user's visual immersion. In some examples, AR devices also include cameras, which can be wide-angle, telephoto, or structured light cameras (also known as point cloud depth cameras, 3D structured light cameras, or depth cameras). Structured light cameras, based on 3D vision technology, can acquire the planar and depth information of objects. A structured light camera projects light with specific structural features onto the object being photographed using a near-infrared laser. The reflected light is then collected by an infrared camera and processed by a processor chip. The calculation principle involves calculating the object's position and depth information based on changes in the light signal caused by the object, presenting a 3D image. Typical terminal devices, such as mobile phones, display two-dimensional images and cannot show the depth of different locations within the image. Structured light cameras can capture and acquire 3D image information, obtaining not only color and other information at different locations but also depth information, which can be used for AR ranging. Of course, ordinary terminal devices can also acquire 2D images using optical cameras and combine this with deep learning algorithms to obtain depth information, ultimately displaying 3D images as well.

[0053] In some examples, the terminal device 20 has an AR-enabled software program or application (APP) installed. The server 10 can be a management server or application server for this software program or APP. The server 10 can be a single server, a server cluster consisting of multiple servers, or a cloud server, etc. The terminal device 20 integrates a network-enabled module, such as a Wireless-Fidelity (Wi-Fi) module, a Bluetooth module, or a 2G / 3G / 4G / 5G communication module, to connect to the server 10 via a network.

[0054] For example, users can log in to their user accounts through an app installed on their mobile phones, or through software installed on AR glasses.

[0055] Taking an AR navigation app as an example, the app can possess capabilities such as high-precision map navigation, environmental understanding, and virtual-real fusion rendering. The app can report its current geographical location information to the server 10 through the terminal device 20, and the server 10 provides AR navigation services to the user based on the real-time geographical location information. For example, if the terminal device 20 is a mobile phone, in response to the user's operation of launching the app, the mobile phone can activate its camera to capture images of the real environment. Then, the system performs AR enhancement on the real environment images captured by the camera, integrating or overlaying rendered AR effects (such as navigation route signs, road names, merchant information, advertising displays, etc.) into the real environment images, and displaying the virtual-real fusion image on the mobile phone screen.

[0056] Figure 2 The illustration schematically shows a virtual-real fusion image for AR navigation using a mobile app, where the AR navigation pointer arrows are superimposed on the real road surface and space in the image, and the electronic promotional materials of merchants float in the space in the form of parachutes carrying gift boxes at designated locations.

[0057] Embodiments of the present invention relate to terminal devices and / or servers. The principles and spirit of the present invention will be explained in detail below through several exemplary embodiments or representative implementations.

[0058] Figure 3 This is a flowchart of a user operation behavior prediction method according to an embodiment of the present invention. In this embodiment, the user operation behavior prediction method includes the following steps:

[0059] Step S101: Construct a directed graph of function points. Specifically, first, obtain the user-operable function points of the software program to be tested, and then construct a directed graph of function points based on the link relationships between the function points. The directed graph includes multiple levels of function point sets with logical order. Each level of function point set includes one or more function points, and each higher-level function point corresponds to a lower-level function point set.

[0060] The user-operable functional points refer to the operation points, or functional atoms, within an application page that respond to user actions and perform a function based on those actions. Examples include page-turning buttons, like icons, navigation icons, etc., on an application page. User actions are not limited to clicking, swiping, or specific gestures. When the software program under test is a terminal application, the levels in the directed graph of these functional points correspond to application pages, and each level of functional points corresponds to a user-operable functional atom within that application page. When a functional point on one application page is accessed by the user, the user navigates to another page; therefore, there is a logical order between pages, meaning there is a specific logical order between different levels.

[0061] Step S102: Construct a 3D model of the directed graph of function points. In this model, multiple function points at the same level are located in the same spatial region, while function points at different levels are located in different spatial regions. For example... Figure 4 As shown, this is a schematic diagram of a directed graph of function points according to one embodiment. In this embodiment, the function point set of the first layer L1 consists of the function points in the homepage P1, which is also the first or lowest level function point set in the logical order. The homepage P1 includes four function points, therefore the function point set of the second layer L2 includes four function point sets that are subordinate to the four function points in the homepage P1, and so on down to the deepest and last page. Figure 4 As shown, page P13 is the set of subordinate function points of function point A3, page P132 is the set of subordinate function points of function point A31, and page P1322 is the set of subordinate function points of function point A312. Pages P13, P132, and P1322 have a logical order. To clearly display the different levels of function points and their logical order, this example sets each level of function point set on a vertical plane. It could also be a horizontal plane, any plane in three-dimensional space, or a three-dimensional spatial region. Furthermore, to distinguish the different levels of function points, different image textures can be used. Alternatively, the 3D model can be given a certain spatial shape, such as a tree or mountain, increasing the readability of the displayed content while clearly distinguishing the different levels of function point sets.

[0062] Step S103: Calculate the initial pheromone concentration for each function point. The pheromone concentration of a function point refers to the probability that the function point will be operated by the user. This probability changes over time depending on whether it is operated by the user. In one embodiment, the pheromone concentration D of each function point is calculated based on the following formula:

[0063] D = [F1, F2, ..., F] i ...F n ]

[0064] Among them, Fi F represents the operational impact factor of the i-th dimension of the current function point, where n is the total number of dimensions; i =d i ·f i f i Let d be the actual impact factor value of the i-th dimension. i The weights of the actual influencing factors are denoted as , where . k is the test case number, x k x is a constant, which is the value of x when the current function point is predicted to be a connected function point. k =1; when the current function point is not predicted as a connected function point, x k =0. The system stores the actual influencing factor values ​​f for each dimension. i The specific values, the dimensions include, for example, function, position on the page, icon size, color, etc., and the actual influence factor value of each dimension has an initial value, whose weight changes in real time according to the click situation of the test cases as the test progresses. When the test has not started, the weight of the actual influence factor value of each dimension is 1. Therefore, in step S103, the initial pheromone concentration of each function point can be obtained according to the above calculation formula.

[0065] Step S104: In response to the start of testing at the test terminal, determine the test level. Specifically, at the start of testing, the set of function points with the lowest logical order in the directed graph is used as the test level. Figure 4 Taking the 3D model as an example, the set of functional points at the very beginning or lowest level of the logical sequence, the first layer L1 (home page P1), is used as the test level. When a connected functional point is predicted, and the test termination condition is not met, the set of functional points at the next level corresponding to that connected functional point is used as the test level. For example, if functional point A3 is obtained on the home page P1, and the test termination condition is not met, P13 corresponding to functional point A3 is used as the test level, and so on, until the test ends.

[0066] Step S105: Based on the operational habits of the test cases and the initial pheromone concentration of the function points, predict one or more function points in the test level.

[0067] In this invention, the test case refers to a simulated user, who has operating habits based on those of real users, and these operating habits are expressed in the form of data. In one embodiment, such as... Figure 5 The diagram shown is a flowchart of the method for obtaining test case data.

[0068] Step S401: Obtain user operation behavior data from one or more software programs of the same type as the software program to be tested. The user operation behavior data includes multiple user operation records, and each user operation record includes user information, function point description information, and operation information.

[0069] Step S402: Based on user information, function description information, and operation information from multiple user operation records, obtain operation habit factor data h of various types of multiple users across multiple dimensions. ij , where i is the dimension index and j is the user type index.

[0070] Step S403: Combine all dimensions of each user's operational habit factor data to form the first array as test case data. in, Where j is the user type index, k is the kth user in the j-th user type, and n is the total number of dimensions.

[0071] Through the above process, various types of user operation habit data can be obtained, thus providing sufficient test case data for the testing of this invention.

[0072] In step S106, when a connected functional point is predicted, the connected functional point and its path to the previous level connected functional point are marked in the 3D model of the directed graph of the functional point according to the marking strategy. In step S107, the 3D model and the marking of the functional point and the marking of the test path of the test case are displayed on the display terminal.

[0073] After predicting one or more function points in the test level, in step S108, the pheromone concentration of all function points in the test level is updated based on the prediction results. In step S109, it is determined whether to end the test. If the test has ended, the current processing flow ends; otherwise, it returns to step S104 to redetermine the test level and perform the test again.

[0074] As seen in the foregoing embodiments, after a software program is developed, in order to understand user behavior, this invention obtains user operating habits based on existing user behavior data from other software programs. Based on the ant colony algorithm, pheromone concentrations are set for the functional points of the software program under test to simulate users as ants, thereby predicting user behavior. This allows developers to determine whether the design of various functions of the software program under test meets expectations, whether adjustments are needed, and where and how to place advertisements and other commercial activities within the software program. This solves the problem of the lag in obtaining user behavior data and the other negative effects of such lag. Because this invention displays the predicted user behavior results in an intuitive three-dimensional model, by adjusting the display angle, the overall situation of user operation throughout the entire process from start to finish can be observed from all angles. For example, it can be intuitively observed on which page a user operation stops. After extensive testing with numerous test cases, based on the display markers, the different user operation paths can be clearly observed. Combined with the data display settings during marking, the test results are displayed intuitively, improving the readability of the displayed content.

[0075] The foregoing Figure 3-5 The user operation behavior prediction method shown is performed by a server, a test terminal, and a display terminal. In one embodiment, the test terminal connects to and communicates with the server, obtains the data required for testing from the server, such as directed graph data of function points and test case data, and sends the connected function point data predicted for each test case during the testing process to the server. The server marks the 3D model based on the connected function point data. When the display terminal is a regular monitor, the server establishes a 3D space and displays the 3D model, connected function points, and paths formed by connecting these connected function points in the 3D space, and displays this data on the monitor. When the display terminal is a PC or mobile terminal with processing capabilities, the display terminal establishes a 3D space, receives the 3D model and marking data sent by the server, and marks the 3D model according to the marking strategy. The following descriptions illustrate different scenarios.

[0076] Figure 6 This is a flowchart of a user operation behavior prediction method in a test terminal according to an embodiment of the present invention. In this embodiment, the test terminal has a software program to be tested installed, and performs prediction of user operation behavior, specifically including the following steps:

[0077] Step S201: Start the software program to be tested and obtain the directed graph data of the function points and the test case data of the software program to be tested; wherein, the directed graph data of the function points includes a multi-level set of function points with logical order, the correspondence between each function point and the lower-level set of function points, and the initial pheromone concentration of each function point, and the test case data includes operation habit data.

[0078] Step S202: Determine the test level. At the beginning, the set of function points with the lowest logical order in the directed graph is used as the test level to start the test. During the test, if a connected function point is predicted and the test is not yet finished, the lower level of that connected function point is used as the test level.

[0079] Step S203: Based on the operational habits of the test cases and the initial pheromone concentration of the function points, predict one or more function points in the test level.

[0080] The steps for predicting one or more functional points in a test level based on the operational habits and initial pheromone concentration of a test case are as follows: Figure 7 As shown:

[0081] Step S501: Obtain the operation habit data of the target test case. When a test case is selected for testing, it is used as the target test case, and its operation habit data is read. The operation habit data is a first array composed of operation habit factors of multiple dimensions; for example, the first array obtained above. Where j is the user type index, k is the kth user in the j-th user type, and n is the total number of dimensions.

[0082] Step S502: Obtain a second array of pheromone concentrations for each functional point in the test level; the pheromone concentration D is, for example, the aforementioned data D = [F1, F2, ..., F...]. i ...F n ].

[0083] Step S503: Calculate the matching degree between the first array and the second array of pheromone concentrations for each function point in the test level. For example, when calculating the matching degree between the first array and a second array, first calculate the similarity value for each dimension, then normalize the similarity values, and finally obtain the sum of the normalized similarity values ​​for all dimensions. This sum is used as the matching degree value between the test case and the function point. When there are multiple function points in the current test level, repeatedly execute step S503 to obtain the matching degree value between the test case and each function point.

[0084] Step S504: Compare the matching degree values ​​of the function points and determine the matching degree with the highest value.

[0085] Step S505: In one embodiment, the highest matching degree value is compared with a threshold. Step S506: Determine if the highest matching degree value is greater than or equal to the threshold. If so, in step S507, the function point with the highest matching degree value is taken as a connected function point. If the highest matching degree value is less than the threshold, it indicates that there is no function point matching the test case in the current test level, and in step S508, it is confirmed that no connected function point has been found. In another embodiment, the function point with the highest matching degree value can also be directly taken as a connected function point.

[0086] Step S204: Determine if a connected functional point has been predicted. Based on the aforementioned prediction process, if a connected functional point is predicted, proceed to step S205. If no connected functional point is predicted, in step S2041, determine if the number of test cases has reached a preset number. If it has, the test termination condition is met, and the test process ends. If the number of test cases has not reached the preset number, in step S2042, replace the test cases and return to step S202. In another embodiment, the test time can also be used as the standard for test termination. For example, timing can be started at the beginning of the test, and then in step S2041, it can be determined whether a preset test time period has been reached. If it has, the test ends; if not, in step S2042, replace the test cases.

[0087] Step S205: Send the connectivity function point identifier and test case identifier to the server.

[0088] Step S206: Determine if the current level is the last level in the directed functional graph. If yes, return to step S2041. If not, return to S202.

[0089] Figure 8 This is a flowchart of a user operation behavior prediction method applied to a server according to an embodiment of the present invention. In this embodiment, the server cooperates with... Figure 6 The test terminal shown displays a 3D model of the test process and the operation path of the test cases on a display-only screen. Specifically, it includes the following steps:

[0090] Step S110: Based on the test start request from the test terminal, send the directed graph data of the function points of the software program to be tested and the test case data to the test terminal.

[0091] Step S120: A three-dimensional space is established according to the display configuration, and the directed graph stereo model of the function points is displayed in the three-dimensional space. In one embodiment, when the display includes an image acquisition device such as a camera, an image of the environment where the display is located is acquired, and a three-dimensional space overlapping with the real environment is established based on the environmental image. When displaying the directed graph stereo model of the function points, a plane in the current space is identified, and the directed graph stereo model of the function points is merged into the plane. In another embodiment, when the display does not have an image acquisition device such as a camera, a three-dimensional space is constructed using a three-dimensional modeling software program or platform, such as constructing a three-dimensional space based on the Unity platform. The directed graph stereo model of the function points is then merged into the three-dimensional space and sent to the display for display, which the user can view directly. Of course, a virtual reality three-dimensional space can also be constructed, which the user can view through a VR device.

[0092] Step S130: Receive the prediction data sent by the test terminal during the test, and parse it to obtain the connectivity function point identifier and test case identifier.

[0093] Step S140: In the directed graph 3D model of function points, the corresponding connected function points and their paths to the next-level connected function points are marked according to a marking strategy, thereby forming the operation behavior path. The marking strategy includes, for example, using different colors and icons to mark connected function points and their paths to the next-level connected function points. In one embodiment, when testing based on the first test case, the marking is based on the connected function point identifiers. When the test case is changed, after marking the current connected function points, the next-level connected function point is found based on the identifier of the current test case, and the two are then connected to form the operation path of that test case, distinguishing it from the operation paths of other test cases. In an optional embodiment, the number of test cases is displayed next to the same test path.

[0094] Step S150: Count the number of test cases for the same path.

[0095] Step S160: Determine if the number of test cases for the same path reaches a preset order of magnitude, such as 10 or 100. If it does, in step S170, mark the path with appropriate identifiers, such as thickening the connecting lines or changing their color. If the number of test cases for the same path does not reach the preset order of magnitude, return to step S130 to continue receiving prediction data from the test terminal and proceed with the subsequent processes.

[0096] Step S180: Determine whether to end the test. For example, after the test is completed, the test terminal will send a notification to the server. Upon receiving the notification that the test is over, the server confirms the end of the test, terminates the current process, and records all test data, including the current marking status of the 3D model and its corresponding data. If the test is not over, return to step S130 to continue receiving prediction data from the test terminal and proceed with the subsequent processes.

[0097] Figure 9 This is a flowchart illustrating a method for interaction between a test terminal, a display terminal, and a server according to an embodiment of the present invention. In this embodiment, the display terminal is a terminal with processing capabilities, and the interaction process among the three is briefly described below:

[0098] In step S101a, the test terminal sends a test request to the server, which includes the identifier of the software program to be tested, such as its name and version number, as well as the identifier of the test terminal, such as its username, nickname, and device number.

[0099] Based on the test request, the server reads the functional directed graph data and test case data of the software program to be tested from the database, and sends them to the test terminal in step S201a. In one embodiment, the server sends the required number of test case data at once, or during the test, it sends one test case data to the test terminal each time it requests test cases.

[0100] In step S202a, based on the test request, the server sends a notification to the display terminal to construct a three-dimensional space and three-dimensional model data.

[0101] In step S301a, the display terminal constructs a three-dimensional space based on the notification and displays the received functional point stereoscopic model in the three-dimensional space. In this embodiment, the display terminal is a terminal with processing capabilities such as a processor and memory, like a PC, mobile smart terminal, AR / VR device, etc. Different three-dimensional spaces are created depending on the type of display terminal. For example, when the display terminal is a mobile smart terminal, AR device, etc., an AR digital space is created based on V-SLAM technology, which overlaps with the real scene. When displaying the functional point stereoscopic model, the user operating the display terminal can move around the model and view it from various angles. When the display terminal is a PC, a three-dimensional space can be constructed using 3D software programs or a 3D platform. The user can view the functional point stereoscopic model like watching a 3D game and change the display angle of the current scene based on certain operations, thus viewing the model from different angles. When the display terminal is a VR device, a virtual reality three-dimensional space is created. The user can change the display angle of the functional point stereoscopic model using gestures, etc., thus viewing it from different angles.

[0102] The test terminal performs the test in step S102a, and sends the prediction data to the server in step S103a when the prediction data is obtained. The test process is as described in the previous embodiment and will not be repeated here.

[0103] After receiving the prediction data sent by the test terminal, the server extracts the connectivity function point identifier and the test case identifier from it, and sends the connectivity function point identifier and the test case identifier to the display terminal in step S203a.

[0104] In S302a, the display terminal marks the connected functional points and their corresponding paths in the three-dimensional model.

[0105] The above process is repeated until the test is completed.

[0106] When the display terminal is a PC or mobile terminal with processing capabilities, it connects to the server, establishes a three-dimensional space according to the server's instructions, receives the stereo model and marking data sent by the server, and marks the stereo model according to the marking strategy.

[0107] Figure 9 The display terminal in the illustrated embodiment differs from the test terminal. In another embodiment, the display terminal can be a test terminal. During testing, the test terminal no longer sends test data to the server, but instead displays the 3D model and the operation path of the test cases in a separate display window on the screen. This window can be hidden and can automatically move its position on the screen.

[0108] Application Examples

[0109] In this embodiment, the test terminal is a smartphone, and the software program to be tested is an AR application. The user-operable objects such as the UI, digital content, and interactive icons in each application scenario of the AR application constitute a set of functional points at one level. Due to the limitations of the current field of view of a smartphone, not all application scenarios can be placed within the field of view. Therefore, in this embodiment, the current application scenario is dynamically divided into multiple scenarios. For example, each field of view is considered as one scenario, and the set of user-operable objects within each scenario is considered as a sub-set of scenario functional points. Figure 10 As shown, the functional points within the current visible range are treated as a subset of scene functional points. Figure 10 Each box in the diagram represents a user-operable functional atom.

[0110] In this embodiment, when the tester holds a smartphone and launches the software program under test, the smartphone constructs a dynamic AR digital 3D space based on V-SLAM technology and overlays it onto the current real-world scene. The software program under test integrates digital content into the current AR digital 3D space, such as the anime-style doll shown in the figure. According to the developers' design, the software program under test has multiple application scenarios, each with multiple user-operable functionalities, such as... Figure 10 The boxes in the diagram represent new application scenarios that arise in response to the operation of each function point, or new perspectives within the current application scenario. These scenarios are considered as the sub-scenarios of that function point, and the set of function points within them is considered as the sub-set of function points.

[0111] Testers can set the current test mode to single-scenario or multi-scenario mode via their smartphones. In single-scenario mode, after starting the test, the smartphone sends a data request to the server, receives directed graph data of functional points and a test case, and determines the test level. In this embodiment, one application scenario is defined as one test level. Since the application scenario in this embodiment is larger than the smartphone's field of view, before testing, the functional points within the currently visible scene are dynamically determined as a subset of scene functional points for the current application scenario. Then, based on the test case data, the functions in the current scene functional point subset are matched and calculated to predict whether the simulated user corresponding to the test case will operate on a functional point within the current scene functional point subset, and further determine which functional point will be operated on. The tester can then enter the corresponding application scenario with the selected functional point. If no functional point is predicted in the current scene functional point subset, the tester can continue to move the location or change the angle within the application scenario to obtain a new field of view, i.e., a new scene functional point subset. Prediction continues until the entire current application scenario has been predicted. If no functional point is predicted in the current application scenario, the smartphone sends a test case data request to the server to obtain new test case data and re-performs the test. If a functional point is predicted, the smartphone sends the functional point, the current subset of functional points, and the corresponding test case identifier to the server. The server dynamically divides the corresponding level from the 3D model based on the received subset of functional points and marks the connecting functional points in the 3D model on the display terminal. The smartphone's prediction process and the marking process in the 3D model are the same as in the previous embodiment and will not be described again here.

[0112] When the tester is set to multi-scenario mode, the tester needs to walk through all the locations involved in the current application scenario, even if the current application scenario has appeared within the smartphone's field of view. At this time, the smartphone makes predictions based on the test case data. The prediction process is the same as in single-scenario mode, so it will not be described in detail here.

[0113] In a further embodiment, when the aforementioned application scenario involves a large area, multiple rest points can be set up in the scene to facilitate the determination of the scene's functional point set. Each time a tester reaches a rest point, they gain a visible range, thus obtaining a scene functional point set. A smartphone (i.e., the terminal device) can prompt the tester to move towards a rest point, for example, through voice prompts or by overlaying directional signs within the current visible range to guide the tester to the rest point, thereby obtaining the scene functional point set for prediction.

[0114] The software program to be tested in the aforementioned application examples is an AR application, but it can also be other applications (APPs) applied to mobile terminals, web applications, or other types of software programs.

[0115] This invention also provides a user operation behavior prediction device, which is applied to a server, such as... Figure 11 The diagram shown is a principle block diagram of a user operation behavior prediction device applied to a server according to an embodiment of the present invention. It includes a function point directed graph construction module 11, a model construction module 12, a pheromone concentration calculation module 13, a data interaction module 14, and a display marking module 15. The function point directed graph construction module 11 is used to obtain the user-operable function points of the software program under test, and constructs a function point directed graph based on the link relationships between the function points. The directed graph includes multiple levels of function point sets with logical order, each level of function point set including one or more function points, and each higher-level function point corresponds to a lower-level function point set. The model construction module 12 is connected to the function point directed graph construction module 11, and constructs a three-dimensional model of the function point directed graph based on the function point directed graph of the software program under test. Multiple function points at the same level are located in the same spatial region, and function points at different levels are located in different spatial regions. One example of a three-dimensional model of a function point directed graph is shown below. Figure 4As shown. The pheromone concentration calculation module 13 calculates the initial pheromone concentration of each function point based on the probability of the function point being operated by the user. The function point directed graph construction module 11, model construction module 12, and pheromone concentration calculation module 13 store the obtained data, models, etc., in the server-side database 101. When needed, the corresponding data can be read from the database 101. The data interaction module 14 is connected to the test terminal and the display terminal through the server's communication module 102. In response to the test start request of the test terminal, it reads the function point directed graph data and test case data of the software program to be tested from the database 101 and sends them to the test terminal. It also sends the function point directed graph 3D model to the display terminal so that the function point directed graph 3D model can be displayed on the display terminal. The display marking module 15 is connected to the data interaction module 14 and is used to receive the prediction data from the test terminal from the data interaction module. According to the function point identifier, it marks the corresponding connected function points and their paths to the previous level connected function points in the function point directed graph 3D model in the display terminal according to the marking strategy.

[0116] This invention also provides a user operation behavior prediction device, which is applied to a test terminal, such as... Figure 12 As shown, this is a user operation behavior prediction device applied to a test terminal according to an embodiment of the present invention. It includes a data acquisition module 21, a test process monitoring module 22, a function point prediction module 23, a simulation operation module 24, and a pheromone concentration update module 25. The data acquisition module 21 acquires directed graph data of function points and test case data of the software program to be tested. The data acquisition module 21 interacts with a server through the communication module 201 of the test terminal, sending data requests and receiving data returned by the server. The directed graph data includes a multi-level set of function points with logical order, the correspondence between each function point and its subordinate set of function points, and the initial pheromone concentration of each function point. The test case data includes operation habit data. The test process monitoring module 22 is connected to the data acquisition module 21 and is used to start the software program to be tested, using the lowest-level set of function points in the directed graph as the initial test level. Upon obtaining a connected function point, the next-level set of function points of that connected function point is used as the test level. The test process ends when the termination condition is met. The function point prediction module 23 is connected to the test process monitoring module 22. Based on the operational habits of the test cases and the pheromone concentration of the function points, it predicts one or more function points in the test level. The prediction process is as follows: Figure 7As shown, further details are omitted here. The simulation operation module 24 is connected to the test process monitoring module 22 and the function point prediction module 23 respectively. In response to the function point prediction module 23 predicting connected function points, it performs operations corresponding to the operation type of the connected function points. Correspondingly, the test process monitoring module 22 uses the set of lower-level function points corresponding to the connected function points as the new test level. The pheromone concentration update module 25 is connected to the function point prediction module 23 and updates the pheromone concentration of all function points in the test level based on the prediction results during the test. The update includes recalculating the pheromone concentration of each function point.

[0117] According to the method and apparatus described above, test case data is used to predict user behavior on various functional points in an application. Based on the comparison between the predicted data and the expected data of the application, specific application optimization schemes can be determined, such as increasing the function point icons, changing the position of the function points in the interface, or setting function point labels, etc. Furthermore, by analyzing the predicted data, some specific suggestions for commercial applications can be obtained, such as determining the best placement for commercial advertisements. Therefore, this invention has guiding significance for both the optimization of the application product itself and commercial applications.

[0118] It should be noted that, for clarity, all embodiments of the present invention are described as a combination of a series of actions or processes. Those skilled in the art should understand that the implementation process is not limited by the order of the described actions or processes, and some steps in the embodiments of the present invention may be processed in other orders or simultaneously.

[0119] Those skilled in the art will understand that the embodiments described herein are preferred embodiments, and the actions, steps, modules, or units involved are not necessarily essential to the embodiments of the present invention. In the above embodiments, the descriptions of each embodiment have their own emphasis; for parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0120] Figure 13 This is a schematic diagram of the structure of an electronic device 60 provided in an embodiment of the present invention. The electronic device 60 includes a processor 61, a memory 62, and a communication bus for connecting the processor 61 and the memory 62. The memory 62 stores a computer program that can run on the processor 61. When the processor 61 runs the computer program, it can execute or implement the steps of the methods in the various embodiments of the present invention. The electronic device 60 also includes a communication interface for receiving and sending data. The electronic device 60 can be a server in the embodiments of the present invention, or it can be a cloud server. The electronic device 60 can also be a terminal device or an AR device in the embodiments of the present invention, or it can be a server in the present invention. Where appropriate, the electronic device can also be called a computing device.

[0121] In some embodiments, processor 61 may be a central processing unit (CPU), graphics processing unit (GPU), application processor (AP), modem processor, image signal processor (ISP), controller, video codec, digital signal processor (DSP), baseband processor, neural-network processing unit (NPU), etc. Processor 61 may also be other general-purpose processors, application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. General-purpose processors may be microprocessors or any conventional processor. The neural network processor (NPU), by drawing inspiration from biological neural network structures, can rapidly process input information and continuously learn itself. The NPU electronic device 60 can realize applications such as intelligent cognition, including image recognition, face recognition, semantic recognition, speech recognition, and text understanding.

[0122] In some embodiments, memory 62 may be an internal storage unit of electronic device 60, such as a hard disk or memory of electronic device 60; memory 62 may also be an external storage device of electronic device 60, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on electronic device 60. Memory 62 may include both internal storage units and external storage devices of electronic device 60. Memory 62 can be used to store operating system, application programs, bootloader, data, and other programs, such as program code of computer programs. Memory 62 includes, but is not limited to, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), or compact disc read-only memory (CD-ROM). Memory 62 is used to store program code executed by electronic device 60 and data transmitted. Memory 62 can also be used to temporarily store data that has been output or will be output.

[0123] Those skilled in the art will understand that Figure 13 This is merely an example of electronic device 60 and does not constitute a limitation on electronic device 60. Electronic device 60 may include more or fewer components than shown, or combine certain components, or include different components, such as input / output devices, network access devices, etc.

[0124] Figure 14 This is a schematic diagram of the software program structure of a terminal device according to an embodiment of the present invention. Taking the Android mobile operating system as an example, in some embodiments, the Android system is divided into four layers: the application layer, the application framework layer (FWK), the system layer, and the hardware abstraction layer. The layers communicate with each other through software program interfaces.

[0125] First, the application layer can include multiple application packages, which can be various application apps such as calling, camera, video, navigation, weather, instant messaging, education, etc., or they can be application apps based on AR technology.

[0126] Second, the Application Framework Layer (FWK) provides application programming interfaces (APIs) and programming frameworks for applications within the application layer. The application framework layer can include predefined functions, such as functions for receiving events sent by the application framework layer.

[0127] The application framework layer may include a window manager, a resource manager, and a notification manager, among others.

[0128] The window manager manages the windowed applications. It can determine the screen size, the presence of a status bar, screen lock, and screen capture capabilities. The content provider stores and retrieves data, making it accessible to applications. This data may include video, images, audio, made and received phone calls, browsing history and bookmarks, and phonebook entries.

[0129] The resource manager provides various resources for applications, such as localized strings, icons, images, layout files, video files, and so on.

[0130] The notification manager allows applications to display notifications in the status bar. These notifications can be used to deliver informational messages and can disappear automatically after a short pause, requiring no user interaction. For example, the notification manager can be used to notify users of download completion or message alerts. The notification manager can also display notifications as icons or scrolling text in the top status bar, such as notifications from background applications, or as dialog boxes on the screen. Examples include displaying text messages in the status bar, emitting sounds, vibrating electronic devices, and flashing indicator lights.

[0131] In addition, the application framework layer may include a view system, which includes visual controls, such as controls for displaying text and controls for displaying images. The view system can be used to build the application. The display interface can consist of one or more views; for example, the display interface of a text notification icon may include a view for displaying text and a view for displaying images.

[0132] Third, the system layer can include multiple functional modules, such as sensor service modules, physical state recognition modules, 3D graphics processing libraries (e.g., OpenGLES), and so on.

[0133] The sensor service module monitors sensor data uploaded by various sensors at the hardware layer to determine the physical state of the phone; the physical state recognition module analyzes and recognizes user gestures, faces, etc.; and the 3D graphics processing library is used to implement 3D graphics drawing, image rendering, compositing, and layer processing.

[0134] In addition, the system layer may include a surface manager and a media library. The surface manager manages the display subsystem and provides 2D and 3D layer blending for multiple applications. The media library supports playback and recording of various common audio and video formats, as well as still image files.

[0135] Finally, the hardware abstraction layer (HAL) is the layer between hardware and software programs. The HAL can include display drivers, camera drivers, sensor drivers, etc., used to drive the relevant hardware in the hardware layer, such as displays, cameras, and sensors.

[0136] This invention also provides a computer-readable storage medium storing a computer program or instructions that, when executed, implement the steps of the method designed in the above embodiments.

[0137] This invention also provides a computer program product, including a computer program or instructions, which, when executed, implement the steps of the method designed in the above embodiments. For example, the computer program product may be a software program installation package.

[0138] Those skilled in the art should understand that the functions of the methods, steps, or related modules / units described in the embodiments of the present invention can be implemented, in whole or in part, by software programs, hardware, firmware, or any combination thereof. When implemented using software programs, they can be implemented, in whole or in part, in the form of a computer program product, or by a processor executing computer program instructions. The computer program product includes at least one computer program instruction, which can be composed of corresponding software program modules. These software program modules can be stored in RAM, flash memory, ROM, EPROM, EEPROM, registers, hard disk, portable hard disk, read-only optical disc (CD-ROM), or any other form of storage medium known in the art. The computer program instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer program instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device such as a server or data center that integrates one or more available media. The available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media, or semiconductor media (e.g., SSDs).

[0139] Regarding the various devices / products described in the above embodiments, the modules / units included can be software program modules / units, hardware modules / units, or a combination of both. For example, for devices / products applied to or integrated into a chip, all of its modules / units can be implemented using hardware methods such as circuits, or at least some modules / units can be implemented using software programs and run on a processor integrated within the chip, while the remaining modules / units can be implemented using hardware methods such as circuits. Similarly, for devices / products applied to or integrated into a terminal, all of its modules / units can be implemented using hardware methods such as circuits, or at least some modules / units can be implemented using software programs and run on a processor integrated within the terminal, while the remaining modules / units can be implemented using hardware methods such as circuits.

[0140] The above description is merely a specific embodiment of the present invention. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, modules, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. It should be understood that the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the protection scope of the present invention.

Claims

1. A user operation behavior prediction method, characterized by, The method comprises: obtaining user-operable function points of a software program to be tested, constructing a function point directed graph according to the link relationship between the function points, the function point directed graph comprising a plurality of levels of function point sets with logical order, each level of function point set comprising one or more function points, and each upper-level function point corresponding to a lower-level function point set; constructing a three-dimensional model of the function point directed graph, wherein the plurality of function points at the same level are located in the same spatial region, and the function points at different levels are located in different spatial regions; calculating the initial pheromone concentration of each function point based on the possibility of the function point being operated by a user, wherein the pheromone concentration D of each function point is calculated based on the following formula: wherein, is the operation impact factor of the current function point in the ith dimension, and n is the total number of dimensions; , is the actual impact factor value of the ith dimension, is the weight of the actual impact factor, wherein, , k is the test case serial number, is a constant, when the current function point is predicted as a connected function point, ; when the current function point is not predicted as a connected function point, ; in response to the test terminal test starting, taking the function point set at the lowest level of logical order in the directed graph as a test level; and predicting one or more function points in the test level based on the operation habit of the test case and the initial pheromone concentration of the function points; in response to a connected function point being predicted, taking the next-level function point set of the connected function point as a new test level; predicting one or more function points in the new test level based on the operation habit of the test case and the pheromone concentration of the function points until the termination condition is met, and ending the test process of the software program; wherein, when a connected function point is predicted, the connected function point and the path of the connected function point from the upper-level connected function point are marked in the three-dimensional model of the function point directed graph according to a marking strategy; after the prediction of one or more function points in the test level is completed, updating the pheromone concentration of all function points in the test level based on the prediction result; and during the test process, displaying the three-dimensional model of the function point directed graph and the marking of the function points and the marking of the test case test path on the screen of the display terminal.

2. The method of claim 1, wherein, When the software program to be tested is a terminal application, the levels in the function point directed graph correspond to application pages, and the function points at each level correspond to function atoms that are user-operable in the application pages.

3. The method of claim 1, wherein, During the test process, the plurality of function points at the current test level are dynamically divided into a plurality of scene function point sub-sets based on application scenarios; correspondingly, when predicting the one or more function points in the test level, one or more function points in a single scene function point sub-set are predicted respectively, or when the function points of all scene function point sub-sets are obtained, a plurality of function points of the total function point set of all scenes are predicted.

4. The method of claim 1, wherein, The step of predicting the function points in the test level based on the operation habit of the test case and the initial pheromone concentration of the function points comprises: obtaining operation habit data of the test case, wherein the operation habit data is a first array composed of a plurality of dimension operation habit factors; obtaining a second array of the pheromone concentration of each function point in the test level; calculating the matching degree of the first array and the second array of the pheromone concentration of each function point in the test level respectively; and taking the function point with the highest matching degree as the connected function point.

5. A user operation behavior prediction method characterized by comprising: The method is applied to a server, and the method comprises: obtaining user-operable function points of a software program to be tested, constructing a function point directed graph according to a link relationship between the function points, the function point directed graph including a plurality of function point sets of levels with logical sequences, each function point set of a level including one or more function points, and each upper-level function point corresponding to a lower-level function point set; constructing a three-dimensional model of the function point directed graph, wherein a plurality of function points of the same level are located in a same spatial region, and function points of different levels are located in different spatial regions; calculating an initial pheromone concentration of each function point based on a possibility of the function point being operated by a user, wherein the pheromone concentration D of each function point is calculated based on the following formula: wherein, is the operation impact factor of the i-th dimension of the current function point, and n is the total number of dimensions; , is the actual impact factor value of the i-th dimension, is the weight of the actual impact factor, wherein, k is the test case serial number, is a constant, when the current function point is predicted to be a connected function point, ; when the current function point is not predicted to be a connected function point, ; in response to a test start request of a test terminal, sending function point directed graph data and test case data of the software program to be tested to the test terminal, and displaying the function point directed graph three-dimensional model on a display terminal, wherein the function point directed graph data includes a plurality of function point sets of levels, a mapping relationship between each function point and a lower-level function point set, and an initial pheromone concentration of each function point, and the test case data includes operation habit data; and in response to the test terminal sending a prediction data including a connected function point identifier, marking a corresponding connected function point and a path of the connected function point and an upper-level connected function point in the function point directed graph three-dimensional model in the display terminal according to a marking strategy.

6. The method of claim 5, wherein, further analyzing a test case identifier from the prediction data received from the test terminal, and determining the upper-level connected function point based on the test case identifier when marking the path of the current connected function point and the upper-level connected function point.

7. The method of claim 6, wherein, further comprising: counting a number of test cases of a same path, and marking the path according to an identifier corresponding to a number level in the marking strategy when the number of test cases reaches the number level.

8. The method of claim 5, wherein, the step of displaying the function point directed graph three-dimensional model on the display terminal further includes: sending a notification of constructing a three-dimensional space to the display terminal; receiving three-dimensional space data returned by the display terminal, obtaining three-dimensional coordinates and model data of each function point in the three-dimensional space based on a structure of the function point directed graph three-dimensional model, the model data including at least texture image data of function points of the same level; and sending the three-dimensional coordinates and the model data of the function points to the display terminal to display the function point directed graph three-dimensional model on the display terminal.

9. The method of claim 8, wherein, the three-dimensional space is an AR digital three-dimensional space constructed by the display terminal based on an environment image collected by the display terminal and coinciding with a current real environment, or the three-dimensional space is a VR digital three-dimensional space constructed by the display terminal.

10. The method of claim 5, wherein, the method further includes a step of obtaining test case data: obtaining user operation behavior data of one or more software programs of the same type as the software program to be tested, the user operation behavior data including a plurality of user operation records, each user operation record including user information, function point description information, and operation information; Obtaining operation habit factor data h of multiple types of users in multiple dimensions based on user information, function point description information and operation information in multiple user operation records ij wherein i is a dimension serial number, and j is a user type serial number. combining all the dimension operation habit factor data of each type of user together to form a first array as test case data ; wherein, wherein, j is the user type serial number, k is the kth user in the jth type of user, and n is the total number of dimensions.

11. A user operation behavior prediction method characterized by comprising: the method is applied to a test terminal device, and includes starting the software program to be tested, and obtaining functional point directed graph data and test case data of the software program to be tested; wherein the functional point directed graph data comprises a multi-level functional point set with logical order, a corresponding relationship between each functional point and a lower-level functional point set, and initial pheromone concentration of each functional point, and the test case data comprises operation habit data; determining a test level starting from the lowest level in the logical order in the directed graph; predicting one or more functional points in the test level based on operation habit of the test case and pheromone concentration of the functional point, wherein the pheromone concentration D of each functional point is calculated based on the following formula: wherein, is the operation impact factor of the current function point in the i-th dimension, and n is the total number of dimensions; , is the actual impact factor value of the i-th dimension, is the weight of the actual impact factor, wherein, , k is the test case serial number, is a constant, when the current function point is predicted as a connected function point, ; when the current function point is not predicted as a connected function point, ; in response to predicting a connected functional point, taking a lower-level functional point set having a corresponding relationship with the connected functional point as a new test level, and predicting one or more functional points in the new test level based on operation habit of the test case and pheromone concentration of the functional point; until a termination condition is met, the testing process of the software program is ended; and updating the pheromone concentration of all functional points in the test level based on the prediction result during the testing process.

12. The method of claim 11, wherein, The step of predicting the functional point in the test level based on operation habit of the test case and pheromone concentration of the functional point comprises: Obtaining operation habit data of a test case, wherein the operation habit data is a first array composed of operation habit factors of multiple dimensions ; wherein, wherein, j is the user type serial number, k is the kth user in the jth user type, and n is the total number of dimensions. obtaining a second array D of the pheromone concentration of each functional point in the test level; respectively calculating the matching degree of the first array and the second array of the pheromone concentration of each functional point in the test level; and taking the functional point with the highest matching degree as the connected functional point.

13. The method of claim 12, wherein, Further comprising: when calculating the matching degree of the first array of a test case and the second array of the pheromone concentration of each functional point in the test level, when the matching degree of the first array and all functional points is less than a threshold value, replacing the test case; or when the connected functional point is one of the functional point set in the highest level in the logical order in the functional point directed graph, replacing the test case; or when the number of test cases reaches a threshold value or the testing time reaches a threshold value, the termination condition is met.

14. The method of claim 11, wherein, Further comprising: when a connected functional point is predicted, sending the connected functional point data and its test case identification as prediction data to a server.

15. The method of claim 11, wherein, Further comprising after the test level is determined: dynamically dividing the multiple functional points in the test level into multiple scene functional point sub-sets corresponding to scenes according to scenes; correspondingly, when predicting one or more functional points in the test level, predicting one or more functional points in a single scene functional point sub-set, or predicting multiple functional points in the total functional point set of all scenes when the functional points of all scene functional point sub-sets are obtained.

16. The method of claim 11, wherein, Further comprising: constructing a three-dimensional space, and displaying a functional point directed graph solid model of the software program to be tested in the three-dimensional space; and when a connected functional point is predicted, marking the corresponding connected functional point and its path with the upper connected functional point in the functional point directed graph solid model according to a marking strategy.

17. The method of claim 16, wherein, Further comprising: counting the number of test cases in the same path, and when the number of test cases reaches a number level in the marking strategy, marking the path according to an identification corresponding to the number level.

18. The method of claim 16, wherein, The three-dimensional space is an AR digital three-dimensional space constructed based on a collected environment image and coinciding with a current real environment, or the three-dimensional space is a VR digital three-dimensional space.

19. A user operation behavior prediction device characterized by comprising: Applied to a server, comprising: A function point directed graph construction module configured to acquire user-operable function points of a software program to be tested, and to construct a function point directed graph according to a link relationship between the function points, the directed graph including a plurality of levels of function point sets having a logical order, each level of function point set including one or more function points, and each upper-level function point corresponding to a lower-level function point set; A model construction module connected to the function point directed graph construction module and configured to construct a three-dimensional model of the function point directed graph based on the function point directed graph of the software program to be tested, wherein a plurality of function points at the same level are located in the same spatial region, and function points at different levels are located in different spatial regions; A pheromone concentration calculation module configured to calculate an initial pheromone concentration of each function point based on a possibility of the function point being operated by a user, wherein the pheromone concentration D of each function point is calculated based on the following formula: wherein, is the operation impact factor of the i-th dimension of the current function point, and n is the total number of dimensions; , is the actual impact factor value of the i-th dimension, is the weight of the actual impact factor, wherein, , k is the test case serial number, is a constant, when the current function point is predicted to be a connected function point, ; when the current function point is not predicted to be a connected function point, ; A data interaction module configured to be connected to a test terminal and a display terminal, and configured to, in response to a test start request of the test terminal, send function point directed graph data and test case data of the software program to be tested to the test terminal, and send the three-dimensional model of the function point directed graph to the display terminal to display the three-dimensional model of the function point directed graph on the display terminal; and A display marking module connected to the data interaction module and configured to receive prediction data from the test terminal from the data interaction module, and to mark corresponding connected function points and paths of the connected function points to upper-level connected function points in the three-dimensional model of the function point directed graph on the display terminal according to a marking strategy based on function point identification.

20. A user operation behavior prediction device characterized by comprising: Applied to a terminal device as a test terminal, comprising: A data acquisition module configured to acquire function point directed graph data and test case data of a software program to be tested; wherein the function point directed graph data includes a plurality of levels of function point sets having a logical order, a corresponding relationship of each function point to a lower-level function point set, and an initial pheromone concentration of each function point, and the test case data includes operation habit data; A test flow monitoring module connected to the data acquisition module and configured to start the software program to be tested, and to use a function point set at a lowest level of the logical order in the directed graph as a test level when starting the test, and to end the test process of the software program when a termination condition is met; A function point prediction module connected to the test flow monitoring module and configured to predict one or more function points in a test level based on an operation habit of a test case and a pheromone concentration of the function point, wherein the pheromone concentration D of each function point is calculated based on the following formula: wherein, is the operation impact factor of the i-th dimension of the current function point, and n is the total number of dimensions; , is the actual impact factor value of the i-th dimension, is the weight of the actual impact factor, wherein, k is the test case serial number, is a constant, when the current function point is predicted to be a connected function point, ; when the current function point is not predicted to be a connected function point, ; an analog operation module connected with the test flow monitoring module and the function point prediction module respectively, configured to implement an operation corresponding to the operation type of the connected function point in response to the connected function point predicted by the function point prediction module; correspondingly, the test flow monitoring module takes the subordinate function point set corresponding to the connected function point as a new test level; and a pheromone concentration updating module connected with the function point prediction module, configured to update the pheromone concentration of all function points in the test level based on the prediction result during the test.

21. A server, characterized by The computer readable storage medium stores computer program instructions, and the computer program instructions are executed by the processor to implement the user operation behavior prediction method applied to the service end according to any one of claims 5-10 or the user operation behavior prediction method applied to the test terminal device according to any one of claims 11-18.

22. A terminal device, comprising: The computer readable storage medium stores computer program instructions, and the computer program instructions are executed by the processor to implement the user operation behavior prediction method applied to the service end according to any one of claims 5-10 or the user operation behavior prediction method applied to the test terminal device according to any one of claims 11-18. The computer readable storage medium stores computer program instructions, and the computer program instructions are executed by the processor to implement the user operation behavior prediction method applied to the service end according to any one of claims 5-10 or the user operation behavior prediction method applied to the test terminal device according to any one of claims 11-18. The computer readable storage medium stores computer program instructions, and the computer program instructions are executed by the processor to implement the user operation behavior prediction method applied to the service end according to any one of claims 5-10 or the user operation behavior prediction method applied to the test terminal device according to any one of claims 11-18.

23. A computer-readable storage medium, characterized in that, ​ 24. A computer program product, characterised in that, ​

Images