A user behavior prediction method, device, apparatus and storage medium

Abstract

The application discloses a user behavior prediction method and device, equipment and a storage medium. The method comprises the following steps: acquiring the identification information of a user, the identification information of each control clicked by the user, the page browsing time corresponding to each control clicked by the user and the historical click frequency corresponding to each control clicked by the user; and predicting the user behavior according to the identification information of the user, the identification information of each control clicked by the user, the page browsing time corresponding to each control clicked by the user and the historical click frequency corresponding to each control clicked by the user, to obtain a target control. Through the technical scheme of the application, the operation guide in different scenes can be intelligently recommended for the user, the learning cost is reduced, the user is more conveniently and quickly interacted with, and the user experience is improved.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a user behavior prediction method, apparatus, device and storage medium. Background Technology

[0002] DevOps systems are end-to-end development systems integrating requirements gathering, development, operations, and testing. These systems are often large, complex to operate, and frequently updated, requiring a significant learning curve. Users typically learn through manuals, multiple clicks, or customer service inquiries, resulting in a poor user experience. While some systems provide user guides when new features are released, these are often general and lack personalization, wasting users' time and effort. Summary of the Invention

[0003] This invention provides a user behavior prediction method, apparatus, device, and storage medium, which solves the problems of high learning costs due to complex system operation and poor user experience due to low personalization of system recommendation information, thus wasting users' time and energy.

[0004] According to one aspect of the present invention, a user behavior prediction method is provided, comprising:

[0005] Obtain the user's identification information, the identification information of each control clicked by the user, the page view time corresponding to each control clicked by the user, and the historical click frequency corresponding to each control clicked by the user;

[0006] Based on the user's identification information, the identification information of each control clicked by the user, the page view time corresponding to each control clicked by the user, and the historical click frequency corresponding to each control clicked by the user, the user's behavior is predicted, and the target control is obtained.

[0007] According to another aspect of the present invention, a user behavior prediction device is provided, the user behavior prediction device comprising:

[0008] The first acquisition module is used to acquire the user's identification information, the identification information of each control that the user has clicked, the page browsing time corresponding to each control that the user has clicked, and the historical click frequency corresponding to each control that the user has clicked;

[0009] The first module is used to predict user behavior and obtain the target control based on the user's identification information, the identification information of each control clicked by the user, the page browsing time corresponding to each control clicked by the user, and the historical click frequency corresponding to each control clicked by the user.

[0010] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0011] At least one processor; and

[0012] A memory communicatively connected to the at least one processor; wherein,

[0013] The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the user behavior prediction method according to any embodiment of the present invention.

[0014] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement the user behavior prediction method according to any embodiment of the present invention.

[0015] This invention, through obtaining user identification information, the identification information of each control clicked by the user, the page browsing time corresponding to each clicked control, and the historical click frequency corresponding to each clicked control, predicts user behavior based on the user's identification information, the identification information of each clicked control, the page browsing time corresponding to each clicked control, and the historical click frequency corresponding to each clicked control, thereby obtaining the target control. This solves the problems of high learning costs due to the complexity of system operation and poor user experience due to the low personalization of system recommendation information, which wastes users' time and energy. It can intelligently recommend operation guides for users in different scenarios, reduce learning costs, and make it more convenient and faster to interact with users, thereby improving the user experience.

[0016] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart of a user behavior prediction method according to Embodiment 1 of the present invention;

[0019] Figure 2This is a schematic diagram of the structure of a user behavior prediction device according to Embodiment 2 of the present invention;

[0020] Figure 3 This is a schematic diagram of the structure of an electronic device according to Embodiment 3 of the present invention. Detailed Implementation

[0021] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0022] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0023] It is understood that before using the technical solutions disclosed in the various embodiments of this disclosure, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in this disclosure in an appropriate manner in accordance with relevant laws and regulations, and user authorization should be obtained.

[0024] Example 1

[0025] Figure 1 This is a flowchart of a user behavior prediction method according to Embodiment 1 of the present invention. This embodiment is applicable to situations where user behavior is predicted and users are accurately guided. This method can be executed by the user behavior prediction device in this embodiment of the present invention, which can be implemented in software and / or hardware, such as... Figure 1 As shown, the method specifically includes the following steps:

[0026] S110, obtain the user's identification information, the identification information of each control clicked by the user, the page browsing time corresponding to each control clicked by the user, and the historical click frequency corresponding to each control clicked by the user.

[0027] The user's identification information can be their login name and password, or their ID information. The user can determine which controls have been clicked based on the identification information of each control. The page view time for each clicked control is the time the user remained on the displayed page after clicking the control. The historical click frequency for each clicked control is the total number of historical clicks on each control.

[0028] Specifically, the methods for obtaining user identification information, the identification information of each control clicked by the user, the page view time corresponding to each control clicked by the user, and the historical click frequency corresponding to each control clicked by the user can be as follows: The DevOps system can collect user identification information, the identification information of each control clicked by the user, the page view time corresponding to each control clicked by the user, and the historical click frequency corresponding to each control clicked by the user within a specified period. It should be noted that the identification information of each control clicked by the user, the page view time corresponding to each control clicked by the user, and the historical click frequency corresponding to each control clicked by the user can be updated in real time according to the specified period for statistical analysis.

[0029] S120: Based on the user's identification information, the identification information of each control clicked by the user, the page browsing time corresponding to each control clicked by the user, and the historical click frequency corresponding to each control clicked by the user, predict the user's behavior and obtain the target control.

[0030] The target control can be the control that the user will click next, which is predicted based on the user behavior using a trained LSTM model.

[0031] Specifically, the method to predict user behavior and obtain target controls based on user identification information, identification information of each control clicked by the user, page browsing time corresponding to each control clicked by the user, and historical click frequency corresponding to each control clicked by the user can be as follows: input the user identification information, identification information of each control clicked by the user, page browsing time corresponding to each control clicked by the user, and historical click frequency corresponding to each control clicked by the user into a trained LSTM model to predict user behavior and obtain target controls.

[0032] Optionally, user behavior can be predicted based on the user's identification information, the identification information of each control clicked by the user, the page view time corresponding to each control clicked by the user, and the historical click frequency corresponding to each control clicked by the user, to obtain the target controls, including:

[0033] The user's identification information, the identification information of each control clicked by the user, the page browsing time corresponding to each control clicked by the user, and the historical click frequency corresponding to each control clicked by the user are input into the target model to obtain the predicted control. The target model is obtained by iteratively training an LSTM model through a target sample set, which includes: the identification information of the sample user, the identification information of each control clicked by the sample user, the page browsing time corresponding to each control clicked by the sample user, the historical click frequency corresponding to each control clicked by the sample user, and the control corresponding to the next operation of the sample user.

[0034] LSTM (Long Short Term Memory) is a time-recurrent neural network that can be used to predict a user's next intention. Predictive controls are controls that predict the user's next click.

[0035] Specifically, the method for obtaining predicted controls by inputting user identification information, identification information of each control clicked by the user, page view time corresponding to each control clicked by the user, and historical click frequency corresponding to each control clicked by the user into the target model can be as follows: The target model is obtained by training an LSTM model using the user identification information, identification information of each control clicked by the user, page view time corresponding to each control clicked by the user, historical click frequency corresponding to each control clicked by the user, and the control corresponding to the user's next action as parameters. The target model is then obtained by inputting the user identification information, identification information of each control clicked by the user, page view time corresponding to each control clicked by the user, and historical click frequency corresponding to each control clicked by the user as input parameters, and the predicted controls are output.

[0036] It's important to note that traditional front-end tracking methods can collect data, tag users, and recommend content based on those tags. However, this technology generates redundant code throughout the data collection process, consuming resources, disrupting original business logic, and leading to inaccurate data collection, thus reducing the accuracy of subsequent user behavior data collection. When predicting controls, user behavior is monitored to obtain the user's single-target content. Users may have multiple roles within the system and have performed actions according to various planned content; therefore, simply classifying or tagging users by role is insufficient for accurate guidance. By iteratively training an LSTM model using a target sample set, taking user identification information, the identification information of each control clicked by the user, the page view time corresponding to each clicked control, and the historical click frequency of each clicked control as input parameters, the LSTM model trained iteratively with the target sample set can predict the user's next intention in real time and accurately, thus obtaining the predicted control.

[0037] Optionally, the LSTM model can be iteratively trained using the target sample set, including:

[0038] The identification information of the sample users in the target sample set, the identification information of each control clicked by the sample users, the page browsing time corresponding to each control clicked by the sample users, and the historical click frequency corresponding to each control clicked by the sample users are input into the LSTM model to obtain the predicted control.

[0039] The parameters of the LSTM model are trained based on the objective function formed by the prediction control and the control corresponding to the next operation of the sample user.

[0040] The process returns to the target sample set and inputs the user identification information, the identification information of each control clicked by the user, the page browsing time corresponding to each control clicked by the user, and the historical click frequency corresponding to each control clicked by the user into the LSTM model to obtain the operation of predicting the control, until the target model is obtained.

[0041] The objective function is a function formed based on the predicted controls and the controls corresponding to the next action of the sample user. It is mainly used to train the parameters of the LSTM model. The parameters of the LSTM model may include: user identification information, identification information of each control clicked by the user, page view time corresponding to each control clicked by the user, and historical click frequency corresponding to each control clicked by the user.

[0042] Optionally, after predicting user behavior and obtaining the predicted controls based on the user's identification information, the identification information of each control clicked by the user, the page view time corresponding to each control clicked by the user, and the historical click frequency corresponding to each control clicked by the user, the method further includes:

[0043] A first menu flow is generated based on each control that the user has clicked and the predicted controls;

[0044] Obtain the similarity between the first menu stream and each menu stream in the menu stream set;

[0045] Menu flows in the menu flow set whose similarity to the first menu flow is greater than a similarity threshold are identified as target menu flows;

[0046] Control recommendations are made based on the target menu flow.

[0047] The first menu stream is generated based on each control that the user has clicked and the predicted controls obtained based on the parameters of the input target model.

[0048] The menu flow set can be a collection of historical menu flows pre-established based on the different intentions of users with different roles or habits for the next operation. For example, a menu flow could be: if a user's first operation in a DevOps system is "Create a new work item," as a developer creating a new work item to manage development data, the menu flow would be [Create new work item, select start time, select end time, associate parent work item, generate child work item…]. As a tester creating a new work item to collect defect details, the menu flow would be [Create new work item, select defect owner, select defect status, select defect level, select deadline…]. As a pipeline manager creating a new work item to create a branch management pipeline, the menu flow would be [Create new work item, associate existing branch / create branch, run pipeline…]. The menu flow set can also be a collection of menu flows formed by at least one menu flow based on the different intentions of users with different roles or habits for the next operation.

[0049] The similarity score is the feature correlation between the first menu flow and each menu flow in the menu flow set. The similarity threshold can be set according to the actual situation. The target menu flow is the menu flow predicted for the user's next action based on the menu flow set and the first menu flow.

[0050] Specifically, the method for generating the first menu flow based on each control clicked by the user and the predicted control can be as follows: after obtaining the predicted control, the first menu flow is formed according to the order in which each control clicked by the user and the predicted control are executed.

[0051] Specifically, the similarity between the first menu stream and each menu stream in the menu stream set can be obtained by associating the first menu stream with each menu stream in the menu stream set, and clustering can be used to determine the similarity between the first menu stream and each menu stream in the menu stream set.

[0052] Specifically, the method for identifying the menu flow in the menu flow set that has a similarity greater than the first menu flow as the target menu flow can be as follows: set a similarity threshold, compare the similarity between each menu flow in the menu flow set and the first menu flow with the similarity threshold, and identify the menu flow with a similarity greater than the similarity threshold as the target menu flow.

[0053] Specifically, the method for recommending controls based on the target menu flow can be as follows: After determining the target menu flow, recommend operations to the user based on the order of controls in the target menu flow. For example, if the first menu flow generated based on the user's clicked control "Create New Work Item" and the predicted control "Select Defect Responsible Person" is [Create New Work Item, Select Defect Responsible Person], and the target menu flow where the similarity between the first menu flow and each menu flow in the menu flow set is greater than the similarity threshold is [Create New Work Item, Select Defect Responsible Person, Select Defect Status, Select Defect Level, Select Deadline...], then the next control to be clicked by the user can be recommended to be "Select Defect Status" based on the order of controls in the target menu flow.

[0054] Optional, also includes:

[0055] If the similarity between the first menu stream and each menu stream in the menu stream set is less than the similarity threshold, then control recommendations are made based on the target control.

[0056] Specifically, if the similarity between the first menu stream and each menu stream in the menu stream set is less than the similarity threshold, then the method for recommending controls based on the target control can be as follows: if the similarity between the first menu stream and each menu stream in the menu stream set is less than the similarity threshold, then the target control predicted based on the target model is directly recommended to the user.

[0057] Optionally, after identifying the menu flows in the menu flow set whose similarity to the first menu flow is greater than a similarity threshold as the target menu flow, the method further includes:

[0058] Obtain the identifier information of the first control clicked by the user;

[0059] If there is no control in the target menu stream that has the same identification information as the first control, then a second menu stream is generated based on the first menu stream and the first control;

[0060] Menu flows in the menu flow set whose similarity to the second menu flow is greater than the similarity threshold are identified as the third menu flow;

[0061] Control recommendations are made based on the third menu flow.

[0062] The first control can be the next control that the user can click after clicking the predicted control. The second menu flow is a menu flow regenerated based on the first menu flow and the first control. The third menu flow is a menu flow predicting the user's next intention after clicking the first control.

[0063] Specifically, the identification information of the first control clicked by the user can be obtained through the DevOps system.

[0064] Specifically, if there is no control in the target menu stream that has the same identification information as the first control, the method for generating the second menu stream based on the first menu stream and the first control can be as follows: obtain the target menu stream; if there is no control in the target menu stream that has the same identification information as the first control, it means that the target menu stream does not match the user's next intention; then regenerate the menu stream based on the first menu stream and the first control, and determine the regenerated menu stream as the second menu stream.

[0065] Specifically, the method for identifying the menu flow in the menu flow set whose similarity to the second menu flow is greater than the similarity threshold as the third menu flow can be as follows: use clustering to determine the similarity between the second menu flow and each menu flow in the menu flow set, and identify the menu flow whose similarity is greater than the similarity threshold as the third menu flow.

[0066] Specifically, the method for recommending controls based on the third menu flow can be as follows: Recommend the next control to the user based on the controls in the third menu flow. For example, if the first menu flow is [Create new work item, select defect manager], and the obtained target menu flow is [Create new work item, select defect manager, select defect status, select defect level, select deadline…], the next control recommended to the user should be "Select defect status". However, the user did not click "Select defect status", and the first control clicked by the user was "Associate existing branch / Create new branch". Since "Associate existing branch / Create new branch" is not in the target menu flow, a second menu flow [Create new work item, select defect manager, associate existing branch / Create new branch] is generated based on the first menu flow and the first control. The similarity between the second menu flow and each menu flow in the menu flow set is compared using a clustering method. Menu flows with a similarity greater than a similarity threshold are identified as the third menu flow. The third menu flow can be [Create new work item, select defect manager, associate existing branch / Create new branch, run pipeline…]. Then, based on the control order in the third menu flow, the next control recommended to the user is "Run pipeline".

[0067] Optional, also includes:

[0068] If the similarity between the first menu flow and each menu flow in the menu flow set is less than the similarity threshold, and the page browsing time corresponding to the control that the user has clicked is greater than the time threshold, then a prompt interface will be displayed.

[0069] In response to detecting a user's editing action on the prompt interface, obtain the target information entered by the user;

[0070] The second control is determined based on the target information and the menu flow set;

[0071] Recommend controls based on the second control.

[0072] The time threshold can be set according to actual needs. The prompt interface can be an intelligent question interface, for example, it could ask: "Hello user! What questions do you have?"

[0073] The target information can be key information input by the user that allows for the next step of the operation. The second control can be the control for the user's next click, determined after analyzing the target information, or it can be the control for the user's next click, determined based on the menu flow set after analyzing the target information.

[0074] Specifically, if the similarity between the first menu stream and each menu stream in the menu stream set is less than the similarity threshold, and the page browsing time corresponding to the control clicked by the user is greater than the time threshold, the prompt interface can be displayed as follows: if the similarity between the first menu stream and each menu stream in the menu stream set is less than the similarity threshold, and the page browsing time corresponding to the control clicked by the user is greater than the time threshold, a pop-up prompt interface will be displayed, and the system's built-in intelligent question-and-answer robot will be introduced to ask questions about the user's current situation.

[0075] Specifically, in response to detecting a user's editing operation on the prompt interface, the way to obtain the target information entered by the user can be: the user enters key information about the next step's intention into the editing position of the prompt interface based on the question in the prompt interface.

[0076] Specifically, the method for determining the second control based on the target information and the menu flow set can be as follows: Identify the keywords corresponding to the target information; match the user-clicked controls, the keywords corresponding to the target information, and each menu flow in the menu flow set; determine the corresponding menu flow based on the matching results; and then determine the second control. Alternatively, the method can be as follows: directly obtain the second control corresponding to the target information by identifying the target information.

[0077] Specifically, the method for recommending controls based on the second control can be: making further recommendations to the user based on the second control.

[0078] The technical solution of this embodiment obtains the user's identification information, the identification information of each control clicked by the user, the page browsing time corresponding to each control clicked by the user, and the historical click frequency corresponding to each control clicked by the user; based on the user's identification information, the identification information of each control clicked by the user, the page browsing time corresponding to each control clicked by the user, and the historical click frequency corresponding to each control clicked by the user, it predicts user behavior and obtains the target control. This solves the problems of high learning cost due to the complexity of system operation and poor user experience due to the low personalization of system recommendation information, which wastes the user's time and energy. It can intelligently recommend operation guides for users in different scenarios, reduce learning costs, and make it more convenient and faster to interact with users, thereby improving the user experience.

[0079] Example 2

[0080] Figure 2This is a schematic diagram of a user behavior prediction device according to Embodiment 2 of the present invention. This embodiment is applicable to situations requiring the prediction of user behavior and precise user guidance. The device can be implemented using software and / or hardware, and can be integrated into any device that provides user behavior prediction functionality, such as… Figure 2 As shown, the user behavior prediction device specifically includes: a first acquisition module 210 and a first obtaining module 220.

[0081] The first acquisition module 210 is used to acquire the user's identification information, the identification information of each control that the user has clicked, the page browsing time corresponding to each control that the user has clicked, and the historical click frequency corresponding to each control that the user has clicked.

[0082] The first module 220 is used to predict user behavior and obtain the target control based on the user's identification information, the identification information of each control clicked by the user, the page browsing time corresponding to each control clicked by the user, and the historical click frequency corresponding to each control clicked by the user.

[0083] Optional, also includes:

[0084] The first generation module is used to generate a first menu stream based on each control clicked by the user and the predicted control;

[0085] The second acquisition module is used to acquire the similarity between the first menu stream and each menu stream in the menu stream set;

[0086] The first determining module is used to determine the menu flow in the menu flow set that has a similarity greater than a similarity threshold with the first menu flow as the target menu flow;

[0087] The first recommendation module is used to recommend controls based on the target menu flow.

[0088] Optional, also includes:

[0089] The second recommendation module is used to recommend controls based on the target control if the similarity between the first menu flow and each menu flow in the menu flow set is less than a similarity threshold.

[0090] Optional, also includes:

[0091] The third acquisition module is used to acquire the identification information of the first control clicked by the user;

[0092] The second generation module is used to generate a second menu stream based on the first menu stream and the first control if there is no control in the target menu stream that has the same identification information as the first control.

[0093] The second determining module is used to determine the menu flow in the menu flow set that has a similarity greater than the similarity threshold with the second menu flow as the third menu flow;

[0094] The third recommendation module is used to recommend controls based on the third menu flow.

[0095] Optional, also includes:

[0096] The first display module is configured to display a prompt interface if the similarity between the first menu stream and each menu stream in the menu stream set is less than a similarity threshold, and the page browsing time corresponding to the control clicked by the user is greater than a time threshold.

[0097] The fourth acquisition module is used to acquire the target information entered by the user in response to the detection of the user's editing operation on the prompt interface;

[0098] The third determining module is used to determine the second control based on the target information and the menu flow set;

[0099] The fourth recommendation module is used to recommend controls based on the second control.

[0100] Optionally, the first obtaining module is specifically used for:

[0101] The user's identification information, the identification information of each control clicked by the user, the page browsing time corresponding to each control clicked by the user, and the historical click frequency corresponding to each control clicked by the user are input into the target model to obtain the predicted control. The target model is obtained by iteratively training an LSTM model through a target sample set, which includes: the identification information of the sample user, the identification information of each control clicked by the sample user, the page browsing time corresponding to each control clicked by the sample user, the historical click frequency corresponding to each control clicked by the sample user, and the control corresponding to the next operation of the sample user.

[0102] Optionally, the first obtaining module is specifically used for:

[0103] The identification information of the sample users in the target sample set, the identification information of each control clicked by the sample users, the page browsing time corresponding to each control clicked by the sample users, and the historical click frequency corresponding to each control clicked by the sample users are input into the LSTM model to obtain the predicted control.

[0104] The parameters of the LSTM model are trained based on the objective function formed by the prediction control and the control corresponding to the next operation of the sample user.

[0105] The process returns to the target sample set and inputs the user identification information, the identification information of each control clicked by the user, the page browsing time corresponding to each control clicked by the user, and the historical click frequency corresponding to each control clicked by the user into the LSTM model to obtain the operation of the predicted control, until the target model is obtained.

[0106] The above-described products can perform the methods provided in any embodiment of the present invention, and have the corresponding functional modules and beneficial effects for performing the methods.

[0107] The technical solution of this embodiment obtains the user's identification information, the identification information of each control clicked by the user, the page browsing time corresponding to each control clicked by the user, and the historical click frequency corresponding to each control clicked by the user; based on the user's identification information, the identification information of each control clicked by the user, the page browsing time corresponding to each control clicked by the user, and the historical click frequency corresponding to each control clicked by the user, it predicts user behavior and obtains the target control. This solves the problems of high learning cost due to the complexity of system operation and poor user experience due to the low personalization of system recommendation information, which wastes the user's time and energy. It can intelligently recommend operation guides for users in different scenarios, reduce learning costs, and make it more convenient and faster to interact with users, thereby improving the user experience.

[0108] Example 3

[0109] Figure 3 This is a schematic diagram of an electronic device according to Embodiment 3 of the present invention. The electronic device 10 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (such as helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0110] like Figure 3As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0111] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0112] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as user behavior prediction methods.

[0113] In some embodiments, the user behavior prediction method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the user behavior prediction method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform the user behavior prediction method by any other suitable means (e.g., by means of firmware).

[0114] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0115] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0116] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0117] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0118] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0119] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0120] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0121] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A user behavior prediction method, characterized in that, include: Obtain the user's identification information, the identification information of each control clicked by the user, the page view time corresponding to each control clicked by the user, and the historical click frequency corresponding to each control clicked by the user; Based on the user's identification information, the identification information of each control clicked by the user, the page browsing time corresponding to each control clicked by the user, and the historical click frequency corresponding to each control clicked by the user, predict user behavior and obtain the target control; The method further includes: after predicting user behavior and obtaining the target control based on the user's identification information, the identification information of each control clicked by the user, the page browsing time corresponding to each control clicked by the user, and the historical click frequency corresponding to each control clicked by the user, the method further includes: Obtain the target controls, and form a first menu flow based on each control clicked by the user and the target controls in the order of clicks; Obtain the similarity between the first menu stream and each menu stream in the menu stream set; If the similarity between the first menu flow and each menu flow in the menu flow set is less than the similarity threshold, and the page browsing time corresponding to the control that the user has clicked is greater than the time threshold, then a prompt interface will be displayed. In response to detecting a user's editing action on the prompt interface, obtain the target information entered by the user; The second control is determined based on the target information and the menu flow set; Recommending controls based on the second control includes: identifying keywords corresponding to the target information based on the target information; matching the controls clicked by the user, the keywords corresponding to the target information, and each menu flow in the menu flow set; determining the corresponding menu flow based on the matching results; determining the second control; and making further recommendations to the user based on the second control.

2. The method according to claim 1, characterized in that, After predicting user behavior and obtaining the target control based on the user's identification information, the identification information of each control clicked by the user, the page view time corresponding to each control clicked by the user, and the historical click frequency corresponding to each control clicked by the user, the process also includes: Menu flows in the menu flow set whose similarity to the first menu flow is greater than a similarity threshold are identified as target menu flows; Control recommendations are made based on the target menu flow.

3. The method according to claim 2, characterized in that, Also includes: If the similarity between the first menu stream and each menu stream in the menu stream set is less than the similarity threshold, then control recommendations are made based on the target control.

4. The method according to claim 2, characterized in that, After identifying the menu flows in the menu flow set whose similarity to the first menu flow is greater than a similarity threshold as the target menu flows, the process also includes: Obtain the identifier information of the first control clicked by the user; If there is no control in the target menu stream that has the same identification information as the first control, then a second menu stream is generated based on the first menu stream and the first control; Menu flows in the menu flow set whose similarity to the second menu flow is greater than the similarity threshold are identified as the third menu flow; Control recommendations are made based on the third menu flow.

5. The method according to claim 1, characterized in that, Based on the user's identification information, the identification information of each control clicked by the user, the page view time corresponding to each control clicked by the user, and the historical click frequency corresponding to each control clicked by the user, the user's behavior is predicted, and the target controls are obtained, including: The target model is obtained by inputting the user's identification information, the identification information of each control clicked by the user, the page browsing time corresponding to each control clicked by the user, and the historical click frequency corresponding to each control clicked by the user. The target model is obtained by iteratively training an LSTM model through a target sample set, which includes: the identification information of the sample user, the identification information of each control clicked by the sample user, the page browsing time corresponding to each control clicked by the sample user, the historical click frequency corresponding to each control clicked by the sample user, and the control corresponding to the next operation of the sample user.

6. The method according to claim 5, characterized in that, The LSTM model is trained iteratively using the target sample set, including: The target control is obtained by inputting the identification information of the sample users in the target sample set, the identification information of each control clicked by the sample users, the page browsing time corresponding to each control clicked by the sample users, and the historical click frequency corresponding to each control clicked by the sample users into the LSTM model. The parameters of the LSTM model are trained based on the objective function formed by the target control and the control corresponding to the next operation of the sample user. The process returns to the point where the identifier information of the sample users in the target sample set, the identifier information of each control clicked by the sample users, the page browsing time corresponding to each control clicked by the sample users, and the historical click frequency corresponding to each control clicked by the sample users are input into the LSTM model to obtain the operation of the target control, until the target model is obtained.

7. A user behavior prediction device, characterized in that, include: The first acquisition module is used to acquire the user's identification information, the identification information of each control that the user has clicked, the page browsing time corresponding to each control that the user has clicked, and the historical click frequency corresponding to each control that the user has clicked; The first module is used to predict user behavior and obtain the target control based on the user's identification information, the identification information of each control clicked by the user, the page browsing time corresponding to each control clicked by the user, and the historical click frequency corresponding to each control clicked by the user. The device further includes: The first generation module is used to obtain target controls and form a first menu stream according to the order in which each control clicked by the user and the target controls are clicked. The second acquisition module is used to acquire the similarity between the first menu stream and each menu stream in the menu stream set; The first display module is configured to display a prompt interface if the similarity between the first menu stream and each menu stream in the menu stream set is less than a similarity threshold, and the page browsing time corresponding to the control clicked by the user is greater than a time threshold. The fourth acquisition module is used to acquire the target information entered by the user in response to the detection of the user's editing operation on the prompt interface; The third determining module is used to determine the second control based on the target information and the menu flow set; The fourth recommendation module is used to recommend controls based on the second control, including: identifying keywords corresponding to the target information based on the target information; matching the controls clicked by the user, the keywords corresponding to the target information, and each menu flow in the menu flow set; determining the corresponding menu flow based on the matching result; determining the second control; and making further recommendations to the user based on the second control.

8. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the user behavior prediction method according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the user behavior prediction method according to any one of claims 1-6.

Images