Application service pushing method and device, equipment and medium
By collecting and analyzing users' personalized operating habits, calculating the operating habit correlation coefficients between devices, classifying similar devices into the same group, pushing corresponding application services, solving the problem of mismatch of user needs in the existing application push methods, and achieving more accurate and efficient application push.
Patent Information
- Application Number
- CN202510204224.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-02-24
- Publication Date
- 2025-06-13
AI Technical Summary
The existing application push methods fail to fully consider the user's personalized operation habits, resulting in the push application not matching the user's needs and the push effect is not good.
Based on the pre-generated operation trajectory acquisition architecture, screen touch operations, screen pressure sensing data, text input behavior data and semantic changes of the text content of multiple devices are collected, and the correlation coefficient of operation habits is calculated, and devices with correlation coefficients greater than the preset value are classified into the same group, and access services for designated applications of each device in the same group are pushed to other devices.
It improves the accuracy and user experience of push, reduces invalid push, saves resources, and improves push efficiency.
Smart Images
Figure CN120144861A_ABST
Abstract
Description
Technical Field
[0001] This specification relates to the field of computer technology, and particularly to an application service push method, apparatus, device, and medium. Background Art
[0002] With the rapid development of the Internet and mobile Internet, and the popularization of mobile devices such as smart phones and tablets, the operation behaviors of users on mobile devices are becoming increasingly complex. In the mobile application market, users are faced with a vast number of application choices. How to improve user stickiness to specific applications and enhance the user experience of applications has become the focus of attention of mobile application developers.
[0003] Traditional application push methods mainly rely on user behavior data, such as geographical location, browsing records, etc., but these methods often ignore the personalized operation habits of users when using specific applications. Therefore, the existing application push methods fail to fully consider the personalized operation habits of users, resulting in a mismatch between the pushed applications and user needs, and poor push effects. Summary of the Invention
[0004] One or more embodiments of this specification provide an application service push method, apparatus, device, and medium for solving the technical problems raised in the background art.
[0005] One or more embodiments of this specification adopt the following technical solutions:
[0006] An application service push method provided by one or more embodiments of this specification, the method includes:
[0007] Based on a pre-generated operation trajectory acquisition architecture, collect the operation habits of multiple devices and the access services of a specified application, where the operation habits include screen touch operations, screen pressure sensing data, text input behavior data, and semantic changes in modified text content;
[0008] Group the devices with a correlation coefficient of the operation habits greater than a preset value into the same group;
[0009] Push the access services of the specified application of each device in the same group to other devices in the same group.
[0010] It should be noted that through the above content, the embodiments of this specification have the following beneficial effects:
[0011] Improve push accuracy: By collecting users' personalized operation habits, including screen touch operations, screen pressure sensing data, text input behavior data, and semantic changes in modified text content, this method can more deeply understand the behavior patterns of users when using specific applications, thereby achieving more accurate application service push.
[0012] Optimize the user experience: Since the pushed applications are recommended based on the user's actual operation habits, the user is more likely to be interested in the pushed applications, thereby improving user satisfaction and application usage frequency.
[0013] Reduce ineffective pushes: By grouping devices with high similarity in operation habits into the same group, ineffective pushes to irrelevant users can be reduced, saving resources and improving push efficiency.
[0014] Further, grouping the devices with the correlation coefficient of the operation habits greater than the preset value includes:
[0015] Determine the application type corresponding to the specified application;
[0016] Obtain the specific operation habits in the operation habits preset for the application type;
[0017] Group the devices with the correlation coefficient of the specific operation habits greater than the preset value into the same group.
[0018] It should be noted that the embodiments of this specification have the following beneficial effects through the above content:
[0019] Enhance personalized push: By using the correlation coefficient of operation habits as the grouping basis, it can ensure that there is a high similarity in the operation habits of specific applications among devices in the same group, thus realizing a more personalized application service push.
[0020] Improve the push effect: By presetting the application type and the corresponding specific operation habits, accurate pushes can be made according to the characteristics of different applications, improving the user's interest and acceptance of the pushed content, thereby enhancing the push effect.
[0021] Simplify the grouping process: The preset operation habits and correlation coefficient thresholds simplify the grouping process, reduce the need for real-time calculation and data processing, and improve the response speed and efficiency of the system.
[0022] Adapt to different application requirements: Setting different operation habits for different application types makes the grouping more refined and can meet the specific requirements of different applications for user behavior data.
[0023] Further, grouping the devices with the correlation coefficient of the operation habits greater than the preset value includes:
[0024] Determine the application type corresponding to the specified application;
[0025] Obtain the specific operation habits in the operation habits preset for the application type;
[0026] Set the occupancy ratio of each operation habit in the operation habits based on the specific habit;
[0027] Determine the correlation coefficient of the operation habits based on the occupancy ratio of each operation habit in the operation habits.
[0028] Group the devices with the correlation coefficient of the operation habits greater than the preset value into the same group.
[0029] It should be noted that through the above content, the embodiments of this specification have the following beneficial effects:
[0030] Fine-grained grouping: By determining the application type corresponding to the specified application and obtaining specific operation habits, fine-grained grouping of different types of applications can be achieved, thereby improving the pertinence of push.
[0031] Operation habit weight allocation: Based on specific habits, setting the occupancy ratio of each operation habit in the operation habits can more accurately reflect the main operation behaviors of users in specific applications, making the grouping more in line with the actual usage habits of users.
[0032] Dynamically adjust the push strategy: By dynamically setting the occupancy ratio of operation habits, the system can adjust the push strategy according to the changes in user behavior, improving the timeliness and relevance of push content.
[0033] Improve the accuracy of the correlation coefficient: By determining the correlation coefficient of operation habits, the similarity of operation habits between devices can be measured more precisely, thereby achieving more effective grouping.
[0034] Optimize the user experience: Grouping based on user operation habits can ensure that the pushed application services are more in line with the interests and needs of users, thereby enhancing the user experience.
[0035] Reduce ineffective push: Through accurate correlation coefficient calculation, the number of times of pushing application services to unmatched users can be reduced, the proportion of ineffective push can be lowered, and resources can be saved.
[0036] Furthermore, the screen touch operation includes the starting coordinates of touching the screen, the ending coordinates of leaving the screen, and the coordinate changes during the movement on the screen; the screen pressure sensing data is the pressure change curve during the process of touching the screen; the text input behavior data includes the input time interval for each character, the character editing behavior, and the occurrence time of the character editing behavior.
[0037] It should be noted that through the above content, the embodiments of this specification have the following beneficial effects:
[0038] Deeply understand user behavior: By collecting the starting and ending coordinates of the screen touch operation and the coordinate changes during the movement, the operation trajectory of the user on the screen can be comprehensively understood, thereby more deeply analyzing the user's behavior pattern.
[0039] Personalized service optimization: The collection of screen pressure sensing data provides detailed information on the pressure changes when the user touches the screen, which helps to optimize personalized services, such as adjusting the font size or touch sensitivity to adapt to different user operation habits.
[0040] Improved interaction experience: By analyzing text input behavior data, such as input time intervals and character editing behaviors, the interaction experience of the application can be improved, for example, by predicting user intentions to optimize the auto-complete function.
[0041] User behavior analysis: These detailed data contribute to user behavior analysis, providing data support for product design and function development, and helping developers better understand user needs.
[0042] Improving application accuracy: The data on the occurrence time of character editing behaviors can help the application more accurately identify the user's intentions, especially during text editing and input, reducing errors and inconveniences.
[0043] Furthermore, the collection of the operation habits of multiple devices and the access services of a specified application includes:
[0044] When the network condition is greater than the preset speed and the device battery level is greater than the preset battery level, collect the operation habits of multiple devices and the access services of the specified application based on the first preset frequency;
[0045] When the network condition is not greater than the preset speed or the device battery level is less than the preset battery level, collect the operation habits of multiple devices and the access services of the specified application based on the second preset frequency, where the first preset frequency is greater than the second preset frequency.
[0046] It should be noted that the embodiments of this specification have the following beneficial effects through the above content:
[0047] Optimal utilization of resources: By adjusting the data collection frequency according to different network conditions and device battery levels, it is possible to optimize the use of network and device battery while ensuring data integrity, avoiding unnecessary resource waste.
[0048] Improved network efficiency: In the case of a better network condition, increasing the data collection frequency can collect user operation habit and access service data faster, improving the efficiency of data collection.
[0049] Battery management: When the network condition is poor or the device battery level is low, reducing the data collection frequency can reduce the consumption of device battery and extend the usage time of the device.
[0050] Furthermore, the method further includes:
[0051] An anomaly detection model based on machine learning identifies invalid operation habits due to equipment failures and eliminates the invalid operation habits from the operation habits of multiple collected devices.
[0052] It should be noted that the embodiments of this specification have the following beneficial effects through the above content:
[0053] Improve data accuracy: By using machine learning algorithms to identify and eliminate invalid operation habits caused by equipment failures, it is possible to ensure that the analyzed data is more accurate and avoid the impact of equipment failures on data analysis results.
[0054] Enhance analysis efficiency: Using a machine learning model for anomaly detection can automatically screen out abnormal data, reducing the time for manual review and data elimination, and improving the efficiency of data analysis.
[0055] Enhance system robustness: By eliminating invalid operation habits, the system is more stable and reliable when facing equipment failures or abnormal behaviors, enhancing the system's robustness.
[0056] Furthermore, the operation habits also include page interaction operation habits, and the page interaction operation habits are the operation details of each element on the page for each device. Combining page jump information forms the page interaction operation habits of each device.
[0057] It should be noted that the embodiments of this specification have the following beneficial effects through the above content:
[0058] In-depth understanding of user behavior: By analyzing page interaction operation habits, including the operation details of each element on the page and page jump information, it is possible to deeply understand users' browsing habits and interests.
[0059] Optimization of personalized recommendations: Understanding users' specific operations on the page, such as clicks and swipes, helps to provide more personalized content recommendations and services for users.
[0060] Improvement of user experience: By analyzing page interaction operation habits, developers can identify pain points during users' operations, thereby optimizing page layouts and interaction designs to enhance the user experience.
[0061] Function improvement direction: The analysis of page interaction operation habits can help developers identify which functions or page elements are favored by users and which need to be improved or optimized.
[0062] An application service push device provided by one or more embodiments of this specification includes:
[0063] The acquisition unit collects the operation habits of multiple devices and the access services of a specified application based on a pre-generated operation trajectory acquisition architecture. The operation habits include screen touch operations, screen pressure sensing data, text input behavior data, and semantic changes in modified text content.
[0064] The grouping unit groups the devices with a correlation coefficient of the operation habits greater than a preset value into the same group.
[0065] The service push unit pushes the access services of the specified application of each device in the same group to other devices in the same group.
[0066] An application service push device provided by one or more embodiments of this specification includes:
[0067] At least one processor; and,
[0068] A memory communicatively connected to the at least one processor; wherein,
[0069] The memory stores instructions executable by the at least one processor. When the instructions are executed by the at least one processor, the at least one processor is enabled to:
[0070] Collect the operation habits of multiple devices and the access services of a specified application based on a pre-generated operation trajectory acquisition architecture. The operation habits include screen touch operations, screen pressure sensing data, text input behavior data, and semantic changes in modified text content.
[0071] Group the devices with a correlation coefficient of the operation habits greater than a preset value into the same group;
[0072] Push the access services of the specified application of each device in the same group to other devices in the same group.
[0073] A non-volatile computer storage medium provided by one or more embodiments of this specification stores computer-executable instructions. When the computer-executable instructions are executed by a computer, they can implement:
[0074] Collect the operation habits of multiple devices and the access services of a specified application based on a pre-generated operation trajectory acquisition architecture. The operation habits include screen touch operations, screen pressure sensing data, text input behavior data, and semantic changes in modified text content.
[0075] Group the devices with a correlation coefficient of the operation habits greater than a preset value into the same group;
[0076] Push the access services of the specified application of each device in the same group to other devices in the same group.
[0077] The above at least one technical solution adopted in the embodiments of this specification can achieve the following beneficial effects:
[0078] Improve push accuracy: By collecting users' personalized operation habits, including screen touch operations, screen pressure sensing data, text input behavior data, and semantic changes in modified text content, this method can more deeply understand users' behavior patterns when using specific applications, thereby achieving more accurate application service pushes.
[0079] Optimize the user experience: Since the pushed applications are recommended based on users' actual operation habits, users are more likely to be interested in the pushed applications, thereby improving user satisfaction and application usage frequency.
[0080] Reduce ineffective pushes: By grouping devices with high operation habit similarity into the same group, ineffective pushes to irrelevant users can be reduced, saving resources and improving push efficiency. Brief Description of the Drawings
[0081] In order to more clearly illustrate the technical solutions in the embodiments of this specification or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments recorded in this specification. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings. In the drawings:
[0082] Figure 1 It is a schematic flowchart of an application service push method provided by one or more embodiments of this specification;
[0083] Figure 2 It is a schematic flowchart of a method for obtaining user operation trajectories based on the consent of mobile users provided by one or more embodiments of this specification;
[0084] Figure 3 It is a schematic structural diagram of an application service push device provided by one or more embodiments of this specification;
[0085] Figure 4 It is a schematic structural diagram of an application service push device provided by one or more embodiments of this specification. Detailed Embodiments
[0086] The embodiments of this specification provide an application service push method, device, device, and medium.
[0087] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of this specification. Obviously, the described embodiments are only a part of the embodiments of this specification, rather than all the embodiments. Based on the embodiments of this specification, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of this specification.
[0088] Figure 1 It is a schematic flowchart of an application service push method provided for one or more embodiments of this specification. This process can be executed by an application service push system. Some input parameters or intermediate results in the process allow manual intervention and adjustment to help improve accuracy.
[0089] The method process steps of the embodiments of this specification are as follows:
[0090] S101, based on a pre-generated operation trajectory collection architecture, collect the operation habits of multiple devices and the access services of a specified application. The operation habits include screen touch operations, screen pressure sensing data, text input behavior data, and semantic changes in modifying text content.
[0091] In the embodiments of this specification, the screen touch operation includes the starting coordinates of touching the screen, the ending coordinates of leaving the screen, and the coordinate changes during the movement on the screen; the screen pressure sensing data is the pressure change curve during the process of touching the screen; the text input behavior data includes the input time interval for each character, the character editing behavior, and the occurrence time of the character editing behavior.
[0092] It should be noted that the above content of the embodiments of this specification can be implemented through the following specific implementation schemes:
[0093] Architecture design: Design a pre-generated operation trajectory collection architecture that should be able to be compatible with multiple devices and operating systems. The architecture should include a data collection module, a data processing module, a storage module, and a user interface module.
[0094] Device compatibility: Ensure that the collection architecture can be adapted to different brands and models of devices, including smartphones, tablets, etc. Adapt to different operating systems (such as Android, iOS, Windows, etc.) to ensure the consistency of data collection.
[0095] The specific data collection is as follows:
[0096] For screen touch operations, the built-in touch screen technology of the device can be utilized to collect the starting coordinates, ending coordinates of the user's touch on the screen, and the coordinate changes during the movement process. An event listener is used to record each touch event and record the corresponding coordinate data.
[0097] For screen pressure sensing data, if the device supports pressure sensing (such as some smartphones), the pressure change curve during the touch process is collected through a pressure sensor. The pressure data is recorded in real time, and the accuracy and integrity of the data are ensured.
[0098] For text input behavior data, the time interval between each character input by the user and the character editing behaviors (such as deletion, addition, replacement, etc.) are collected. A keyboard event listener is used to record the user's input behavior and record the relevant timestamps.
[0099] For the semantic changes in the modified text content, natural language processing (NLP) technology can be utilized to analyze the changes in the text content and identify the semantic changes.
[0100] It should be noted that the embodiments of this specification have the following beneficial effects through the above content:
[0101] Deep understanding of user behavior: By collecting the starting and ending coordinates of the screen touch operation and the coordinate changes during the movement process, the operation trajectory of the user on the screen can be comprehensively understood, thereby more deeply analyzing the user's behavior pattern.
[0102] Personalized service optimization: The collection of screen pressure sensing data provides detailed information on the pressure changes when the user touches the screen, which helps to optimize personalized services, such as adjusting the font size or touch sensitivity to adapt to different user operation habits.
[0103] Improved interaction experience: By analyzing text input behavior data, such as input time intervals and character editing behaviors, the interaction experience of the application can be improved, such as optimizing the auto-complete function by predicting the user's intention.
[0104] User behavior analysis: These detailed data help to conduct user behavior analysis, provide data support for product design and function development, and help developers better understand user needs.
[0105] Improving application accuracy: The occurrence time data of character editing behaviors can help the application more accurately identify the user's intention, especially during text editing and input, reducing errors and inconveniences.
[0106] Furthermore, the operation habits also include page interaction operation habits, and the page interaction operation habits are the operation details of each element in the page for each device. Combining the page jump information forms the page interaction operation habits of each device.
[0107] It should be noted that the above content of the embodiments of this specification can be implemented through the following specific implementation plans:
[0108] Define indicators for page interaction operation habits: Determine the indicators for page interaction operation habits that need to be collected, such as the number of clicks, click positions, click durations, number of scrolls, scroll distances, dwell times, number of page jumps, and frequencies, etc.
[0109] Technical preparation: Select appropriate Web front-end and back-end technology stacks to support data collection, processing, and analysis. Ensure that the technologies used are compatible with different devices and browsers.
[0110] Front-end implementation: Embed JavaScript code in the web page or use existing front-end libraries (such as Google Analytics, Mixpanel, etc.) to track user behavior. Use event listeners to capture user operations such as clicks, scrolls, and inputs. Record detailed information on user interactions with page elements, including element IDs, class names, positions, etc.
[0111] Back-end implementation: Develop a back-end service to receive user behavior data sent from the front-end. Use APIs to process and store this data, ensuring the real-time nature and security of the data.
[0112] The data collection is as follows:
[0113] For page interaction data, collect user interaction behaviors such as clicks and scrolls on each element on the page. Use JavaScript's'mousemove', 'click','scroll', etc. events to capture user operations.
[0114] For page jump information, track user jumps between different pages and record the jump paths and dwell times. Use the browser's 'popstate' event or URL changes to detect page jumps.
[0115] It should be noted that through the above content, the embodiments of this specification have the following beneficial effects:
[0116] In-depth understanding of user behavior: By analyzing page interaction operation habits, including operation details of each element on the page and page jump information, the browsing habits and interests of users can be deeply understood.
[0117] Optimization of personalized recommendations: Understanding the specific operations of users on the page, such as clicks and swipes, helps to provide more personalized content recommendations and services for users.
[0118] User experience improvement: By analyzing page interaction operation habits, developers can identify pain points during user operations, thereby optimizing page layout and interaction design to enhance the user experience.
[0119] Function improvement direction: Analyzing page interaction operation habits can help developers identify which functions or page elements are favored by users and which need to be improved or optimized.
[0120] Furthermore, when collecting the operation habits of multiple devices and the access services of a specified application, when the network condition is greater than the preset speed and the device battery level is greater than the preset battery level, collect the operation habits of multiple devices and the access services of the specified application based on a first preset frequency; when the network condition is not greater than the preset speed or the device battery level is less than the preset battery level, collect the operation habits of multiple devices and the access services of the specified application based on a second preset frequency, and the first preset frequency is greater than the second preset frequency. The preset speed, preset battery level, first preset frequency, and second preset frequency can be set based on the actual scenario. For example, the preset speed is a network signal of 4G or above, the preset battery level is 80%, the first preset frequency is 20 times per second, and the second preset frequency is 5 times per second.
[0121] It should be noted that through the above content, the embodiments of this specification have the following beneficial effects:
[0122] Optimal utilization of resources: By adjusting the data collection frequency according to different network conditions and device battery levels, while ensuring data integrity, it is possible to optimize the use of network and device battery power and avoid unnecessary resource waste.
[0123] Improved network efficiency: In the case of a good network condition, increasing the data collection frequency can collect user operation habit and access service data faster, improving the efficiency of data collection.
[0124] Battery management: When the network condition is poor or the device battery level is low, reducing the data collection frequency can reduce the consumption of device battery power and extend the usage time of the device.
[0125] S102. Group the devices with the correlation coefficient of the operation habits greater than the preset value into the same group.
[0126] In the embodiments of this specification, the application type corresponding to the specified application can be determined first; obtain the specific operation habits in the operation habits preset for the application type; group the devices with the correlation coefficient of the specific operation habits greater than the preset value into the same group.
[0127] It should be noted that the above content of the embodiments of this specification can be implemented through the following specific implementation methods:
[0128] Determine the application type: Define the characteristics of different application types, such as education, entertainment, business, social, etc. Set a set of standard characteristics for each type for easy identification and classification. Analyze the characteristics of the specified application, such as its functions, user interface, target user group, etc. Classify the specified application into the corresponding application type according to the classification system.
[0129] Obtain specific operation habits: For each application type, define a set of key operation habits.
[0130] Calculate the correlation coefficient: Select a suitable correlation coefficient method according to the data type and distribution, such as the Pearson correlation coefficient, Spearman rank correlation coefficient, or Kendall rank correlation coefficient. Calculate the correlation coefficient between the specific operation habits of each device.
[0131] Device grouping: According to business requirements or data analysis results, set a preset value for the correlation coefficient. If the correlation coefficient of the specific operation habits of multiple devices is greater than the preset value, then group them together.
[0132] It should be noted that through the above content, the embodiments of this specification have the following beneficial effects:
[0133] Enhance personalized push: By using the correlation coefficient of operation habits as the grouping basis, it can ensure that there is a high similarity in the operation habits of specific applications among devices in the same group, thus realizing a more personalized application service push.
[0134] Improve the push effect: By presetting the application type and corresponding specific operation habits, it is possible to perform precise pushes according to the characteristics of different applications, improve the user's interest and acceptance of the pushed content, and thus enhance the push effect.
[0135] Simplify the grouping process: The preset operation habits and correlation coefficient thresholds simplify the grouping process, reduce the need for real-time calculation and data processing, and improve the response speed and efficiency of the system.
[0136] Adapt to different application requirements: Setting different operation habits for different application types makes the grouping more refined and can meet the specific requirements of different applications for user behavior data.
[0137] Furthermore, when grouping devices with a correlation coefficient of operation habits greater than the preset value, it is also possible to first determine the application type corresponding to the specified application; obtain the specific operation habits in the operation habits preset for the application type; set the occupancy ratio of each operation habit in the operation habits based on the specific habits; determine the correlation coefficient of the operation habits based on the occupancy ratio of each operation habit in the operation habits; and group the devices with a correlation coefficient of operation habits greater than the preset value together.
[0138] It should be noted that the above content of the embodiments of this specification can be implemented through the following specific implementation schemes:
[0139] 1. Determine the application type of the specified application
[0140] Establish an application type classification standard: Define the characteristics of different application types, such as functions, user interfaces, target user groups, etc. Establish a classification system to clearly classify applications into the corresponding types.
[0141] Identify the specified application type: Analyze the characteristics of the specified application and determine its application type according to the classification standard.
[0142] 2. Obtain specific operation habits
[0143] Define operation habits: For each application type, determine a set of standard operation habits, such as the number of clicks, usage duration, function usage frequency, etc.
[0144] Collect specific operation habit data: Collect user operation data through in-app event tracking, logging, or third-party analysis tools. Ensure that data collection complies with privacy protection regulations.
[0145] 3. Set the occupancy ratio of each operation in the operation habits
[0146] Analyze operation habits: Analyze the collected data to determine which operation habits are the most critical for the use of the application.
[0147] Calculate the occupancy ratio: Calculate the occupancy ratio of each operation habit in the entire set of operation habits. The occupancy ratio can be relative (e.g., a certain operation habit accounts for 20% of the total operations) or absolute (e.g., the average usage duration of a certain operation habit accounts for 30% of the total usage duration).
[0148] 4. Determine the correlation coefficient of the operation habits
[0149] Select the correlation coefficient method: Select a suitable correlation coefficient method according to the data type and distribution, such as the Pearson correlation coefficient, Spearman rank correlation coefficient, etc.
[0150] Calculate the correlation coefficient: Considering the occupancy ratio of each operation habit, calculate the correlation coefficient of the operation habits between different devices.
[0151] 5. Set the preset value and group
[0152] Set the preset value: Set a preset threshold for the correlation coefficient according to business requirements and data analysis results.
[0153] Group: Compare the correlation coefficient of the operation habits with the preset value. If the correlation coefficient of the operation habits between multiple devices is greater than the preset value, then group them into the same group.
[0154] It should be noted that the embodiments of this specification have the following beneficial effects through the above content:
[0155] Fine-grained grouping: By determining the application type corresponding to a specified application and obtaining specific operation habits, fine-grained grouping of different types of applications can be achieved, thereby improving the pertinence of push.
[0156] Operation habit weight allocation: Based on specific habits, setting the occupancy ratio of each operation habit in the operation habits can more accurately reflect the main operation behaviors of users in a specific application, making the grouping more in line with the actual usage habits of users.
[0157] Dynamically adjust the push strategy: By dynamically setting the occupancy ratio of operation habits, the system can adjust the push strategy according to changes in user behavior, improving the timeliness and relevance of push content.
[0158] Improve the accuracy of the correlation coefficient: By determining the correlation coefficient of operation habits, the similarity of operation habits between devices can be measured more precisely, thereby achieving more effective grouping.
[0159] Optimize the user experience: Grouping based on user operation habits can ensure that the pushed application services are more in line with the interests and needs of users, thereby enhancing the user experience.
[0160] Reduce ineffective pushes: Through precise correlation coefficient calculation, the number of times of pushing application services to unmatched users can be reduced, the proportion of ineffective pushes can be lowered, and resources can be saved.
[0161] S103. Push the access services of the specified applications of each device in the same group to other devices in the same group.
[0162] It should be noted that when pushing the access services of the specified applications of each device in the same group to other devices in the same group, the following method can be used. For example, device A, device B, and device C are in a group. The access services of device A in the specified application include service 1, service 2, and service 3. The access services of device B in the specified application include service 4, service 5, and service 6. The access services of device C in the specified application include service 7, service 8, and service 9. When pushing, service 4, service 5, service 6, service 7, service 8, and service 9 can be pushed to device B and device C, service 1, service 2, service 3, service 7, service 8, and service 9 can be pushed to device B, and service 1, service 2, service 3, service 4, service 5, and service 6 can be pushed to device C.
[0163] Furthermore, the embodiments of this specification can identify ineffective operation habits due to device failures based on an anomaly detection model of machine learning, and eliminate the ineffective operation habits from the operation habits collected from multiple devices.
[0164] It should be noted that the above content of the embodiments of this specification can be implemented through the following specific implementation methods:
[0165] 1. Abnormal Detection Model Training
[0166] Select a model: Select a machine learning model suitable for anomaly detection, such as Isolation Forest, One-Class SVM, Autoencoders, or LSTM (Long Short-Term Memory Network).
[0167] Data annotation: Annotate data with known anomalies, or use semi-supervised or unsupervised methods to distinguish normal and abnormal operations.
[0168] Model training: Use the annotated data or unsupervised learning methods to train the anomaly detection model.
[0169] 2. Anomaly Detection and Identification of Ineffective Operation Habits
[0170] Model prediction: Use the trained model to predict new operation habit data and identify abnormal or potentially faulty data points.
[0171] Threshold setting: Set a threshold to determine which operation habits are considered abnormal. This threshold can be determined through cross-validation.
[0172] Elimination of ineffective operation habits: Identify and eliminate ineffective operation habits based on the model's prediction and the set threshold.
[0173] It should be noted that through the above content, the embodiments of this specification have the following beneficial effects:
[0174] Improve data accuracy: By using machine learning algorithms to identify and eliminate ineffective operation habits caused by equipment failures, the analyzed data can be ensured to be more accurate, avoiding the impact of equipment failures on data analysis results.
[0175] Enhance analysis efficiency: Using a machine learning model for anomaly detection can automatically screen out abnormal data, reducing the time for manual review and data elimination, and improving the efficiency of data analysis.
[0176] Enhance system robustness: By eliminating ineffective operation habits, the system is more stable and reliable in the face of equipment failures or abnormal behaviors, enhancing the system's robustness.
[0177] The present invention discloses a method for obtaining user operation trajectories based on the consent of mobile users, which relates to the field of mobile Internet technology. Aiming at the problems of unclear user consent and superficial collection in the current acquisition of mobile operation trajectories, the present invention constructs a multi-level visual consent guidance mechanism. When the APP is started for the first time, it explains to the user in detail the purpose, content, storage time limit and privacy protection measures of operation trajectory collection in a form of pictures and texts. After the user gives consent through intuitive operations, the function is enabled, and the authorization can be adjusted at any time. The refined operation trajectory acquisition module relying on the underlying interface of the mobile terminal can accurately capture the details of touch, text input, and page interaction operations, and also has an intelligent adaptive strategy to adjust the acquisition frequency according to the network and battery conditions. The intelligent data preprocessing and storage system uses machine learning to eliminate invalid data, mixes and compresses to reduce costs, and stores in a three-dimensional manner. The in-depth data analysis and application components use clustering and association rule mining to provide portraits and optimization suggestions for developers, and push personalized services to users. The present invention effectively protects user privacy, helps the optimization of APPs and the implementation of personalized services, and ensures the quality and security of data.
[0178] The present invention relates to the field of mobile Internet technology, and particularly to technical means for accurately collecting user operation trajectories in the environment of application programs (APPs) running on various mobile terminal devices such as smart phones and tablet computers in a legal, compliant and user-willing manner.
[0179] With the explosive growth of mobile applications, developers increasingly expect to deeply understand user behavior patterns in order to optimize product design, provide accurate personalized services, and achieve efficient marketing. However, there are significant deficiencies in the current acquisition of mobile user operation trajectories. On the one hand, when many APPs are started for the first time or updated, the privacy policies and user consent terms are often hidden in long and complex texts, making it difficult for users to clearly understand the scope, purpose, and subsequent processing methods of operation trajectory collection in a short time. As a result, many users passively consent without knowing the truth, which undoubtedly violates the privacy rights of users. On the other hand, existing operation trajectory acquisition technologies mostly stay at a relatively superficial level, usually only able to record simple page access sequences and rough click counts, and cannot capture in detail the behavioral details of users during the operation process, such as the speed change when the finger slides on the screen to browse content, the pause and hesitation time in a specific area, the modification idea during text input, and the time interval between different operations. Such a rough information collection is neither conducive to truly helping developers accurately optimize products nor enabling users to actually benefit from personalized services.
[0180] The main objective of the present invention is to construct a method for obtaining user operation trajectories based on the consent of mobile users, effectively safeguarding users' right to know and right of independent choice, enabling users to voluntarily decide whether to authorize the collection of operation trajectories on the basis of full understanding; meanwhile, after obtaining users' consent, innovative technologies are used to comprehensively and highly accurately collect users' operation trajectories, providing valuable data support for APP developers, and then realizing the organic unity of user rights protection, APP quality improvement, and precise delivery of personalized services.
[0181] To achieve the above objective, the present invention provides a method for obtaining user operation trajectories based on the consent of mobile users, and the specific process is as follows:
[0182] I. Multi-level visual consent guidance mechanism:
[0183] Design a visual consent guidance process with distinct levels and easy to understand. When a user first opens the APP, the first interface presented is not the main application interface, but a carefully designed "Explanation of Operation Trajectory Collection" guidance page. At the top of this page, a prominent large-font headline states "Your privacy, we protect - Regarding the collection of operation trajectories", and the main body below is divided into multiple regions with vivid pictures and texts. First, use vivid picture examples combined with short texts to explain why operation trajectories are collected. For example, show a picture of a user finding a favorite product based on the accurate recommendations of the APP, and caption it with "By understanding your operation habits, we can recommend better products that suit your taste"; then, list in detail the specific content collected, covering "every sliding path of your finger on the screen (accurate to the pixel coordinate sequence), the start and end time and position of long-press actions, every character of text input and the modification process (including the time nodes of deletion, replacement, insertion and the previous and subsequent texts), the jump logic between each page element (accurate to the millisecond jump time and source and target element identifiers)", etc.; then, clearly state the data storage time limit, such as "The collected operation trajectory data will be stored on the company's professional-level security server for 3 months and will be automatically destroyed after the expiration"; finally, introduce the privacy protection mechanism with vivid encryption icons and concise texts, "Adopt bank-level encryption algorithms to ensure that data is not leaked, and only authorized professional engineers can access it when optimizing the APP".
[0184] Users need to express their consent or dissent through a series of intuitive operations. For example, a slidable slider is set at the bottom of the page. The initial position is at the "dissent" end, and the words "I consent to collection" are marked on the slider. After the user reads the information, they slide the slider to the "consent" end. At this time, a confirmation pop-up window appears on the page, showing the key information again and asking the user if they are sure. After the user clicks the confirmation button, the APP officially enables the operation trajectory acquisition function. At the same time, users can always re-check and adjust the authorization status in the "Privacy" sub-item of the APP settings menu, and the system updates the display status icon in real time. A green tick is shown for consent, and a red cross is shown for dissent.
[0185] II. Refined operation trajectory collection module:
[0186] Relying on the underlying interfaces of the mobile operating system, a set of deep and refined operation trajectory collection architectures is built. For touch operations, high-precision touch event monitoring technology is used to not only accurately record the starting coordinates (accurate to pixels) where the finger touches the screen and the ending coordinates where the finger leaves the screen, but also to track in real time the coordinate changes during the finger's movement on the screen. The sampling frequency is as high as 20 times per second, generating a continuous sequence of coordinate points to restore the sliding path. At the same time, combined with the device pressure sensing function (if supported), the pressure change curve during the touch process is recorded to analyze the user's concentration and emotional state during browsing. For example, a continuous increase in pressure may indicate that the user is hesitating or focusing on key content. For text input, the underlying framework of the input method is embedded. In addition to recording each character input and the input time interval, it deeply captures editing behaviors such as character modification, deletion, copy and paste, and the time when they occur, and analyzes the semantic changes in the text before and after each modification to gain insights into the user's thinking adjustment process. When it comes to page interaction, the APP page element architecture is deeply analyzed, and element recognition and event monitoring technology are used to record the operation details of each element such as buttons, menus, dropdown boxes, etc., including clicks, selections, expansions, and foldings by the user. Combining page jump information, the user's behavior logic within the APP is completely outlined. For example, the time when the user clicks the "Product Details" button on the home page to enter the details page, the button ID, and the time when the details page finishes loading are recorded.
[0187] Considering the network environment and power consumption issues of mobile devices, the collection module is built with an intelligent adaptive strategy. When the network condition is good (such as 4G or above network, strong WiFi signal) and the device battery is sufficient (such as the remaining battery power is 80% or above), high-frequency and high-precision collection is maintained, such as 20 samples per second as mentioned above. When the network fluctuates (such as 3G network, weak WiFi signal) or the device battery is low (such as the remaining battery power is 20% or below), the sampling frequency is automatically reduced to 5 times per second to reduce the data transmission volume, giving priority to ensuring the user's basic usage experience, and then restoring high-frequency collection after the network and battery conditions improve.
[0188] III. Intelligent Data Preprocessing and Storage System:
[0189] The massive operation trace data collected first enters the intelligent preprocessing link. Here, an anomaly detection model based on machine learning is used to identify and eliminate invalid data caused by reasons such as equipment failures (such as screen flashing and abnormal clicks caused by accidental touches) and network anomalies (such as false operation records caused by momentary packet loss); a hybrid compression technology is adopted. For continuous data such as coordinate sequences, a lossless compression algorithm is used, and for text data, an efficient lossy compression algorithm is used (to reduce the storage volume while ensuring that key information is not lost), significantly reducing the storage cost.
[0190] The preprocessed data is stored in a three-dimensional manner according to user personalization identifiers (such as user ID, device unique identification code, etc.), APP classification, and time period. The storage medium selects a high-performance and high-security solid-state drive array adapted to the mobile end, combined with a distributed storage architecture, which not only ensures fast data reading and writing but also prevents data loss caused by single-point failures; customized storage solutions are provided according to the needs of different APP developers. For example, e-commerce APPs can focus on storing the operation traces during the user's purchase decision-making process, while social APPs focus on storing the operation traces related to user interaction and communication.
[0191] IV. Deep Data Analysis and Application Components:
[0192] Relying on a professional data analyst team and cutting-edge machine learning algorithms, the stored data is deeply mined. Through clustering analysis, mobile users with similar operation habits are grouped into one group, providing APP developers with a precise target audience portrait to facilitate the targeted development of new functions and optimization of old functions; using association rule mining technology, the potential connections between user operations and different results within the application (such as purchase behavior, content sharing, level improvement, etc.) are found to guide developers to optimize the product process to guide users to achieve expected behaviors. For example, if it is found that users are likely to place an order after frequently browsing food pictures and searching for nearby restaurants, then the navigation can be optimized between the browsing and searching links to shorten the conversion path.
[0193] The analysis results are fed back to APP developers in the form of visual charts (such as bar charts showing the operation frequency differences of different user groups and line charts reflecting the changing trend of user operation habits over time) and data reports (including detailed user behavior insights and improvement suggestions). At the same time, based on the analysis of the user's personal operation traces, personalized services are pushed to the user, such as "You often view aerobic courses in the fitness APP recently. It is recommended that you participate in the newly launched aerobic boxing course this week", so that users can truly feel the benefits brought by the collection of operation traces.
[0194] Effectively protect user privacy: Through a multi-level visual consent guidance mechanism, users can clearly and thoroughly understand the collection details of operation tracks, independently control the flow of personal data, effectively avoid the risk of privacy leakage, and greatly enhance user trust in the APP.
[0195] Assist in deep optimization of APP: The high-precision data provided by the refined operation trajectory collection module opens a window for developers to deeply understand user behavior, enabling them to accurately locate product pain points, tap potential needs, and promote continuous iteration and upgrading of APP.
[0196] Accurately implement personalized services: Based on in-depth analysis and application of user operation trajectories, personalized services tailored for users are more in line with actual needs, significantly improving user satisfaction with the APP and enhancing user stickiness.
[0197] Ensure data quality and security: Intelligent data preprocessing and advanced storage and analysis technologies ensure data reliability and availability, laying a solid foundation for long-term data-driven decision-making.
[0198] Figure 2 A flowchart of a method for obtaining user operation traces based on the consent of a mobile user is provided for one or more embodiments of this specification, specifically:
[0199] Design a multi-level visual consent guidance mechanism to clearly express the meaning of data collection and make user authorization more intuitive. Refine the operation trajectory collection module and build a deep and refined operation trajectory collection architecture. Intelligent data preprocessing and storage system to eliminate useless data and customize storage solutions. Deep data analysis and application components to deeply mine and analyze stored data.
[0200] Figure 3 A schematic diagram of the structure of an application service push device provided for one or more embodiments of this specification includes: a collection unit 301, a grouping unit 302 and a service push unit 303.
[0201] The collection unit 301 collects the operation habits of multiple devices and the access services of designated applications based on a pre-generated operation track collection framework, wherein the operation habits include screen touch operations, screen pressure sensing data, text input behavior data, and semantic changes of modified text content;
[0202] A grouping unit 302, grouping devices whose correlation coefficient of the operation habits is greater than a preset value into the same group;
[0203] The service push unit 303 pushes the access service of the designated application of each device in the same group to other devices in the same group.
[0204] Figure 4A schematic structural diagram of an application service push device provided for one or more embodiments of this specification, including:
[0205] At least one processor; and,
[0206] A memory communicatively connected to the at least one processor; wherein,
[0207] The memory stores instructions executable by the at least one processor, and when the instructions are executed by the at least one processor, the at least one processor is enabled to:
[0208] Collect the operation habits of multiple devices and the access services of a specified application based on a pre-generated operation trajectory collection architecture, where the operation habits include screen touch operations, screen pressure sensing data, text input behavior data, and semantic changes in modified text content;
[0209] Group the devices with a correlation coefficient of the operation habits greater than a preset value into the same group;
[0210] Push the access services of the specified application of each device in the same group to other devices in the same group.
[0211] A non-volatile computer storage medium provided for one or more embodiments of this specification, storing computer-executable instructions, and when the computer-executable instructions are executed by a computer, they can implement:
[0212] Collect the operation habits of multiple devices and the access services of a specified application based on a pre-generated operation trajectory collection architecture, where the operation habits include screen touch operations, screen pressure sensing data, text input behavior data, and semantic changes in modified text content;
[0213] Group the devices with a correlation coefficient of the operation habits greater than a preset value into the same group;
[0214] Push the access services of the specified application of each device in the same group to other devices in the same group.
[0215] Each embodiment in this specification is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other. Each embodiment focuses on the differences from other embodiments. In particular, for the embodiments of the device, equipment, and non-volatile computer storage medium, since they are basically similar to the method embodiments, the description is relatively simple, and for the relevant parts, reference can be made to the partial description of the method embodiments.
[0216] Each embodiment in this specification is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other, and the differences between each embodiment and other embodiments are emphasized. In particular, for the device embodiments, since they are basically similar to the method embodiments, the descriptions are relatively simple, and reference can be made to the relevant parts of the method embodiments for the relevant content.
[0217] Those of ordinary skill in the art can realize that the units and algorithm steps of the examples described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.
[0218] In the embodiments provided in this application, it should be understood that the disclosed devices / network devices and methods can be implemented in other ways. For example, the device / network device embodiments described above are merely illustrative. For example, the division of the modules or units is only a logical function division, and there can be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces, and the indirect couplings or communication connections of the devices or units can be in electrical, mechanical or other forms.
[0219] The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they can be located in one place, or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0220] In addition, the functional units in each embodiment of this application can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above units can be implemented in the form of hardware or software.
[0221] When the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, to implement all or part of the processes in the above-described embodiment methods of this application, it can also be completed by a computer program instructing relevant hardware. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above-described various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disc, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electrical carrier signal, telecommunication signal, and software distribution medium, etc. It should be noted that the content included in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electrical carrier signals and telecommunication signals.
[0222] The above-described embodiments are only used to illustrate the technical solutions of this application, rather than to limit them; although this application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of this application, and should all be included within the protection scope of this application.
Claims
1. A method for pushing an application service, characterized in that: The method comprises: Based on the pre-generated operation track collection framework, the operation habits of multiple devices and the access services of designated applications are collected, and the operation habits include screen touch operations, screen pressure sensing data, text input behavior data and semantic changes of modified text content; Grouping devices whose correlation coefficients of the operation habits are greater than a preset value into the same group; The access service of the designated application of each device in the same group is pushed to other devices in the same group.
2. The method according to claim 1, characterized in that The step of grouping the devices whose correlation coefficients of the operation habits are greater than a preset value into the same group includes: Determine an application type corresponding to the specified application; Acquire a specific operating habit among the operating habits pre-set for the application type; Devices whose correlation coefficients of the specific operating habits are greater than a preset value are grouped together.
3. The method according to claim 1, characterized in that The step of grouping the devices whose correlation coefficients of the operation habits are greater than a preset value into the same group includes: Determine an application type corresponding to the specified application; Acquire a specific operating habit among the operating habits pre-set for the application type; Setting the occupation ratio of each operating habit in the operating habits based on the specific habit; determining a correlation coefficient of the operating habits based on an occupation ratio of each operating habit in the operating habits; Devices whose correlation coefficients of the operating habits are greater than a preset value are grouped together.
4. The method according to claim 1, characterized in that: The screen touch operation includes the starting coordinates of contacting the screen, the ending coordinates of leaving the screen, and the coordinate changes during movement on the screen; the screen pressure sensing data is the pressure change curve during the touch screen process; the text input behavior data includes the input time interval for each character, the character editing behavior, and the occurrence time of the character editing behavior.
5. The method according to claim 1, characterized in that The collecting of the operation habits of multiple devices and the access services of designated applications includes: When the network condition is greater than a preset speed and the device power is greater than a preset power, collecting the operation habits of multiple devices and access services of designated applications based on a first preset frequency; When the network condition is not greater than the preset speed or the device power is less than the preset power, the operation habits of multiple devices and the access services of the designated application are collected based on the second preset frequency, and the first preset frequency is greater than the second preset frequency.
6. The method according to claim 1, characterized in that The method further comprises: The anomaly detection model based on machine learning identifies invalid operating habits due to equipment failure and eliminates invalid operating habits from the collected operating habits of multiple devices.
7. The method according to claim 1, characterized in that The operation habits also include page interaction operation habits, which are operation details of each device on each element in the page, combined with page jump information to form the page interaction operation habits of each device.
8. An application service push device, characterized in that: include: A collection unit, based on a pre-generated operation track collection architecture, collects the operation habits of multiple devices and access services of designated applications, wherein the operation habits include screen touch operations, screen pressure sensing data, text input behavior data, and semantic changes of modified text content; A grouping unit, grouping devices whose correlation coefficient of the operation habits is greater than a preset value into the same group; The service push unit pushes the access service of the designated application of each device in the same group to other devices in the same group.
9. An application service push device, characterized in that: include: at least one processor; as well as, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to: Based on the pre-generated operation track collection framework, the operation habits of multiple devices and the access services of designated applications are collected, and the operation habits include screen touch operations, screen pressure sensing data, text input behavior data and semantic changes of modified text content; Grouping devices whose correlation coefficients of the operation habits are greater than a preset value into the same group; The access service of the designated application of each device in the same group is pushed to other devices in the same group.
10. A non-volatile computer storage medium, characterized in that: Computer executable instructions are stored, and when the computer executable instructions are executed by a computer, the following can be achieved: Based on the pre-generated operation track collection framework, the operation habits of multiple devices and the access services of designated applications are collected, and the operation habits include screen touch operations, screen pressure sensing data, text input behavior data and semantic changes of modified text content; Grouping devices whose correlation coefficients of the operation habits are greater than a preset value into the same group; The access service of the designated application of each device in the same group is pushed to other devices in the same group.