Operation interface switching method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention provides an operation interface switching method and device, electronic equipment and a storage medium. When address information, behavior habit information and use time period information sent by client side equipment are received, the operation interface is switched; based on the address information, the behavior habit information and the use time period information, generating a user category judgment result for the user through the user category recognition model; when it is determined that the user is an old person based on the user category determination result, a control signal is sent to the client device; and the client device is configured to switch the operation interface to the old people mode in response to the received control signal, so that the situation that a user actively operates to enter the old people mode is avoided, and the switching efficiency of the operation interface is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of operation interface switching, and in particular to an operation interface switching method, an operation interface switching device, an electronic device and a computer-readable storage medium. Background Art

[0002] Smartphones have become an indispensable part of life, but for the elderly, complex mobile phone operations are often daunting. In order to allow the elderly to easily enjoy the convenience brought by technology, mobile phone manufacturers have launched "senior mode". This mode usually simplifies the mobile phone interface, enlarges the font, and highlights commonly used functions to adapt to the visual and operating habits of the elderly.

[0003] However, the elderly mode of most mobile phones in related technologies needs to be turned on manually by users, which undoubtedly increases the usage threshold for elderly people who are not familiar with mobile phone operations. Summary of the invention

[0004] The embodiments of the present invention provide an operation interface switching method, device, electronic device and computer-readable storage medium to overcome the above problems or at least partially solve the above problems.

[0005] An embodiment of the present invention discloses an operation interface switching method, which is applied to a cloud server, wherein the cloud server is configured with a corresponding client device, and the client device is provided with an operation interface, including:

[0006] Acquire an initial training set, wherein the initial training set includes address training information, behavior habit training information, and usage period training information;

[0007] Generate a user category recognition model for identifying user categories based on the address training information, the behavior habit training information and the usage period training information; the client device is configured to: obtain the user's address information, behavior habit information and usage period information; send the address information, the behavior habit information and the usage period information to the cloud server;

[0008] When receiving the address information, the behavior habit information and the usage period information sent by the client device, generating a user category determination result for the user through the user category recognition model based on the address information, the behavior habit information and the usage period information;

[0009] When the user is determined to be an elderly person based on the user category determination result, a control signal is sent to the client device; the client device is configured to: in response to receiving the control signal, switch the operation interface to the elderly mode.

[0010] Optionally, before the step of generating a user category identification model for identifying user categories based on the address training information, the behavior habit training information and the usage period training information, the method further includes:

[0011] For incomplete data, abnormal data and duplicate data in the initial training set;

[0012] Performing a filling operation on the incomplete data;

[0013] Performing a removal operation on the abnormal data and the duplicate data;

[0014] A normalization operation is performed on the initial training set.

[0015] Optionally, before the step of generating a user category identification model for identifying user categories based on the address training information, the behavior habit training information and the usage period training information, the method further includes:

[0016] Determine a convolutional neural network as an initial model; the initial model includes a convolution kernel;

[0017] Features are extracted from the initial training set by using the convolution kernel.

[0018] Optionally, the step of generating a user category identification model for identifying user categories based on the address training information, the behavior habit training information and the usage period training information includes:

[0019] Determining refined features for behavioral habit training information from the features extracted by the convolution kernel;

[0020] Calculating the importance reference value of the refined feature through the gradient in the back propagation process; the importance reference value is used to characterize the importance of the refined feature;

[0021] A user category identification model for identifying user categories is generated by using the address training information, the refined features, the importance reference values ​​for the refined features, and the usage period training information.

[0022] Optionally, the user category determination result includes user feature information for characterizing user features, and further includes:

[0023] Constructing a client device control instruction corresponding to the user characteristic information;

[0024] Sending the client device control instruction to the client device;

[0025] The client device is configured to, in response to receiving the client device control instruction, execute the client device control instruction.

[0026] Optionally, the user characteristic information is used to characterize the user's visual impairment, and the client device control instruction corresponding to the user characteristic information is a zoom instruction; the client device is configured to, in response to receiving the zoom instruction, enlarge the interface information of the operation interface.

[0027] Optionally, the user characteristic information is used to characterize the hearing impairment of the user, and the client device control instruction corresponding to the user characteristic information is a volume control instruction; the client device is configured to increase the volume in response to receiving the volume control instruction.

[0028] The embodiment of the present invention further discloses an operation interface switching method, which is applied to a client device, wherein the client device is configured with a corresponding cloud server, the client device is provided with an operation interface, and the cloud server is configured to: obtain an initial training set, wherein the initial training set includes address training information, behavior habit training information, and usage period training information; generate a user category recognition model for identifying user categories based on the address training information, the behavior habit training information, and the usage period training information; the method includes:

[0029] Obtain the user's address information, behavior information, and usage time information;

[0030] sending the address information, the behavior habit information and the usage period information to the cloud server; the cloud server is configured to: upon receiving the address information, the behavior habit information and the usage period information sent by the client device, generate a user category determination result for the user through the user category recognition model based on the address information, the behavior habit information and the usage period information; and when the user is determined to be an elderly person based on the user category determination result, send a control signal to the client device;

[0031] In response to receiving the control signal, the operation interface is switched to the elderly mode.

[0032] The embodiment of the present invention further discloses an operation interface switching device, which is applied to a cloud server, the cloud server is configured with a corresponding client device, and the client device is provided with an operation interface, including:

[0033] An initial training set acquisition module, used to acquire an initial training set, wherein the initial training set includes address training information, behavior habit training information, and usage period training information;

[0034] A user category identification model generation module is used to generate a user category identification model for identifying user categories based on the address training information, the behavior habit training information and the usage period training information; the client device is configured to: obtain the user's address information, behavior habit information and usage period information; send the address information, the behavior habit information and the usage period information to the cloud server;

[0035] A user category determination result generating module, configured to generate a user category determination result for the user through the user category recognition model based on the address information, the behavior habit information and the usage period information sent by the client device when receiving the address information, the behavior habit information and the usage period information;

[0036] A control signal sending module is used to send a control signal to the client device when the user is determined to be an elderly person based on the user category determination result; the client device is configured to: in response to receiving the control signal, switch the operation interface to the elderly mode.

[0037] The embodiment of the present invention further discloses an operation interface switching device, which is applied to a client device, wherein the client device is configured with a corresponding cloud server, and the client device is provided with an operation interface, and the cloud server is configured to: obtain an initial training set, wherein the initial training set includes address training information, behavior habit training information, and usage period training information; generate a user category recognition model for identifying user categories based on the address training information, the behavior habit training information, and the usage period training information; the device includes:

[0038] Information acquisition module, used to obtain the user's address information, behavior habit information and usage time information;

[0039] an information sending module, for sending the address information, the behavior habit information and the usage period information to the cloud server; the cloud server is configured to: upon receiving the address information, the behavior habit information and the usage period information sent by the client device, generate a user category determination result for the user through the user category recognition model based on the address information, the behavior habit information and the usage period information; and when the user is determined to be an elderly person based on the user category determination result, send a control signal to the client device;

[0040] An operation interface switching module is used to switch the operation interface to the elderly mode in response to receiving the control signal.

[0041] The embodiment of the present invention further discloses an electronic device, comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus;

[0042] The memory is used to store computer programs;

[0043] The processor is used to implement the method described in the embodiment of the present invention when executing the program stored in the memory.

[0044] The embodiment of the present invention further discloses a computer-readable storage medium having instructions stored thereon, which, when executed by one or more processors, enables the processors to execute the method described in the embodiment of the present invention.

[0045] The embodiments of the present invention include the following advantages:

[0046] In an embodiment of the present invention, an initial training set is obtained, and the initial training set includes address training information, behavior habit training information, and usage period training information; a user category recognition model for identifying user categories is generated based on the address training information, the behavior habit training information, and the usage period training information; the client device is configured to: obtain the user's address information, behavior habit information, and usage period information; send the address information, the behavior habit information, and the usage period information to the cloud server; when the address information, the behavior habit information, and the usage period information sent by the client device are received, a user category determination result for the user is generated through the user category recognition model based on the address information, the behavior habit information, and the usage period information; when the user is determined to be an elderly person based on the user category determination result, a control signal is sent to the client device; the client device is configured to: in response to receiving the control signal, switch the operation interface to the elderly mode, thereby achieving the following beneficial effects:

[0047] Intelligence: Through machine learning technology, the system can automatically identify user categories without the need for manual settings by the user.

[0048] Personalization: Based on the user category, the system can provide targeted services, such as elderly mode.

[0049] Convenience: Elderly users do not need complicated operations, the system can automatically switch to the appropriate mode.

[0050] Improve user experience: simplify the operation interface, reduce learning costs, and make it easier for the elderly to use smart devices.

[0051] Through machine learning technology, intelligent identification of user categories is achieved, and personalized services are provided to elderly users based on the identification results, effectively solving the problem of elderly people using smart devices. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 is a flowchart of a method for switching an operation interface provided in an embodiment of the present invention;

[0053] Figure 2 It is a schematic diagram of a flow chart of an operation interface switching method provided in an embodiment of the present invention;

[0054] Figure 3 is a flowchart of another method for switching operation interfaces provided in an embodiment of the present invention;

[0055] Figure 4 is a structural block diagram of an operation interface switching device provided in an embodiment of the present invention;

[0056] Figure 5 is a structural block diagram of another operation interface switching device provided in an embodiment of the present invention;

[0057] Figure 6 is a hardware structure block diagram of an electronic device provided in an embodiment of the present invention;

[0058] Figure 7 is a schematic diagram of a computer-readable medium provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0059] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0060] Reference Figure 1 , shows a flowchart of a method for switching an operation interface provided in an embodiment of the present invention, which may specifically include the following steps:

[0061] Step 101, obtaining an initial training set, wherein the initial training set includes address training information, behavior habit training information, and usage period training information;

[0062] Step 102, generating a user category recognition model for identifying user categories based on the address training information, the behavior habit training information and the usage period training information; the client device is configured to: obtain the user's address information, behavior habit information and usage period information; send the address information, the behavior habit information and the usage period information to the cloud server;

[0063] Step 103, when the address information, the behavior habit information and the usage period information sent by the client device are received, a user category determination result for the user is generated by the user category recognition model based on the address information, the behavior habit information and the usage period information;

[0064] Step 104, when the user is determined to be an elderly person based on the user category determination result, a control signal is sent to the client device; the client device is configured to: in response to receiving the control signal, switch the operation interface to the elderly mode.

[0065] The embodiment of the present invention is applied to a cloud server, the cloud server is configured with a corresponding client device, and the client device is provided with an operation interface.

[0066] In practical applications, client devices refer to various electronic devices that can connect to the Internet and exchange data with cloud servers. These devices have various forms and functions, but they all have the following common features:

[0067] Terminality: As a device that users directly contact, client devices usually have a human-computer interaction interface, through which users can input instructions to the system or obtain information.

[0068] Network connectivity: The client device needs to have network connectivity to exchange data with the cloud server.

[0069] Application software: Client devices usually have various application software installed, which is responsible for interacting with users and sending requests or receiving data to cloud servers.

[0070] Common client device types may include, but are not limited to:

[0071] Smartphone: The most common client device. Almost all modern smartphones have the ability to interact with cloud servers.

[0072] Tablet: Similar to a smartphone, a tablet is also a mobile device, but with a larger screen size, making it more suitable for browsing the web, watching videos, etc.

[0073] Personal computers: including desktops and laptops, are traditional computing devices that interact with cloud servers through browsers or specialized software.

[0074] IoT devices: such as smart home devices, wearable devices, etc., usually have weak computing power and mainly rely on cloud servers to complete complex data processing and storage.

[0075] On-board equipment: The on-board system on a smart car can interact with cloud servers to achieve functions such as remote control and real-time navigation.

[0076] Data interaction between client devices and cloud servers:

[0077] Data interaction between client devices and cloud servers is mainly based on the following methods:

[0078] HTTP protocol: This is the most commonly used method of data interaction. Through the HTTP protocol, the client can send a request to the server, and the server returns the corresponding response.

[0079] WebSocket Protocol: The WebSocket protocol is a full-duplex communication protocol on a single TCP connection, which enables the server to push messages to the client, thereby achieving real-time communication.

[0080] MQTT protocol: MQTT protocol is a lightweight publish / subscribe protocol suitable for IoT devices, which can achieve low-power and high-reliability data transmission.

[0081] gRPC protocol: gRPC is a modern RPC framework that enables efficient and reliable remote procedure calls.

[0082] Data interaction process:

[0083] Client initiates request: The application on the client device sends a request to the cloud server based on the user's operation.

[0084] Server processing request: After receiving the request, the server processes it according to the content of the request and generates a corresponding response.

[0085] Server returns response: The server returns the processing result to the client.

[0086] Client displays results: After receiving the response, the client displays the result to the user, or performs subsequent operations based on the response.

[0087] In order to ensure the security of data transmission, the following measures are usually adopted:

[0088] HTTPS protocol: Use HTTPS protocol to encrypt data transmission to prevent data from being eavesdropped.

[0089] API Key: Each client application is assigned a unique API key for authentication purposes.

[0090] Access control: Set different access rights for different client devices and users.

[0091] Example: Interaction between mobile APP and cloud server

[0092] When a mobile app is started, it sends a request to the cloud server to obtain user information, latest news and other data. When the user performs a search operation, the app sends the search keyword to the server, the server searches the database for relevant information, and returns the result to the app, which then displays the result to the user.

[0093] Data interaction between client devices and cloud servers is the foundation of modern Internet applications. Through various protocols and technologies, client devices can access cloud resources anytime and anywhere to achieve a variety of functions.

[0094] The embodiment of the present invention can obtain an initial training set to establish a basic data set containing various user data to provide data support for subsequent model training.

[0095] Beneficial effect: By collecting enough representative user data, the generalization ability of the model can be improved, making it better adapted to the characteristics of different users.

[0096] 1. Address training information:

[0097] Content: Geographic location information of the user's device, including detailed address, longitude and latitude, etc.

[0098] Purpose: Identify nursing homes: By comparing address information with the geographic location database of nursing homes, hospitals, parks and other places, determine whether the user lives or frequently moves in these places.

[0099] Analyze the surrounding environment: Based on the address information, analyze whether there are medical institutions, community service centers and other elderly care-related infrastructure around the user, providing a basis for subsequent personalized services.

[0100] Analysis of regional characteristics: Elderly care habits and needs may vary in different regions. Address information can be used to understand the characteristics of the user's region and provide a more accurate basis for the model.

[0101] 2. Behavioral habit training information:

[0102] Content: Various behavioral data of users when using the device, including operation speed, sliding speed, click frequency, dwell time, font size, volume setting, etc.

[0103] Purpose: Identify elderly users: Elderly users usually operate their phones more slowly than young people, click less frequently, and prefer large fonts and high volume. By analyzing these behavioral characteristics, we can preliminarily determine whether the user is elderly.

[0104] Analyze usage habits: Users of different age groups have different habits when using mobile phones. By analyzing user behavior habits, we can provide elderly users with interfaces and functions that are more in line with their habits.

[0105] Identify potential health issues: Some behavioral traits may be related to the user’s health status. For example, if the user frequently makes mistakes when clicking, they may have vision problems.

[0106] 3. Use period training information:

[0107] Content: The time period during which the user uses the device, including daily active hours, weekend usage, etc.

[0108] Purpose:

[0109] Identify elderly users: Elderly people have relatively regular living habits and usually use their phones at fixed time periods. By analyzing the user's active time periods, you can determine whether the user is elderly.

[0110] Analyze usage scenarios: User behavior may be different at different time periods. For example, the elderly may check the news in the morning and listen to music in the evening. By analyzing usage time periods, we can understand the needs of users at different time periods and provide a basis for personalized services.

[0111] The overall purpose of the initial training set:

[0112] Create user portraits: By comprehensively analyzing information such as user addresses, behavioral habits, and usage time, a detailed portrait is created for each user, including age, health status, interests and hobbies, etc.

[0113] Identify elderly users: Based on user portraits, accurately identify elderly users and provide them with more suitable applications and services.

[0114] Provide data foundation for personalized services: provide data support for subsequent personalized recommendations, intelligent reminders and other functions.

[0115] By building an initial training set containing rich user data, the system can better understand users and provide more accurate and humanized services for elderly users. This data can not only be used to identify elderly users, but also provide data support for subsequent personalized recommendations, health monitoring and other functions.

[0116] The embodiment of the present invention can generate a user category recognition model to train a user category recognition model based on an initial training set, which can accurately identify user categories according to information such as user addresses, behavioral habits and usage time periods, and build an elderly recognition system based on the user category recognition model.

[0117] Beneficial effects: This model can provide a basis for subsequent intelligent judgment and provide more accurate services for elderly users.

[0118] The embodiment of the present invention can generate a user category determination result to analyze the real-time data of new users or existing users to determine whether they are elderly users.

[0119] Beneficial effects: Determine the user's category in real time and provide a basis for subsequent personalized services.

[0120] The embodiment of the present invention can send a control signal to switch to the elderly mode, so that when the system determines that the user is an elderly person, it sends an instruction to the client device to switch it to the elderly mode.

[0121] Beneficial effects: Provide elderly users with a simpler and easier-to-operate interface, improving user experience.

[0122] In an embodiment of the present invention, an initial training set is obtained, and the initial training set includes address training information, behavior habit training information, and usage period training information; a user category recognition model for identifying user categories is generated based on the address training information, the behavior habit training information, and the usage period training information; the client device is configured to: obtain the user's address information, behavior habit information, and usage period information; send the address information, the behavior habit information, and the usage period information to the cloud server; when the address information, the behavior habit information, and the usage period information sent by the client device are received, a user category determination result for the user is generated through the user category recognition model based on the address information, the behavior habit information, and the usage period information; when the user is determined to be an elderly person based on the user category determination result, a control signal is sent to the client device; the client device is configured to: in response to receiving the control signal, switch the operation interface to the elderly mode, thereby achieving the following beneficial effects:

[0123] Intelligence: Through machine learning technology, the system can automatically identify user categories without the need for manual settings by the user.

[0124] Personalization: Based on the user category, the system can provide targeted services, such as elderly mode.

[0125] Convenience: Elderly users do not need complicated operations, the system can automatically switch to the appropriate mode.

[0126] Improve user experience: simplify the operation interface, reduce learning costs, and make it easier for the elderly to use smart devices.

[0127] Through machine learning technology, intelligent identification of user categories is achieved, and personalized services are provided to elderly users based on the identification results, effectively solving the problem of elderly people using smart devices.

[0128] Based on the above embodiment, a variant embodiment of the above embodiment is proposed. It should be noted that in order to make the description concise, only the differences from the above embodiment are described in the variant embodiment.

[0129] In an optional embodiment of the present invention, before the step of generating a user category identification model for identifying user categories based on the address training information, the behavior habit training information and the usage period training information, the step further includes:

[0130] For incomplete data, abnormal data and duplicate data in the initial training set;

[0131] Performing a filling operation on the incomplete data;

[0132] Performing a removal operation on the abnormal data and the duplicate data;

[0133] A normalization operation is performed on the initial training set.

[0134] In the specific implementation, during the data cleaning stage, the embodiment of the present invention not only removes invalid and incomplete data, but also normalizes the data to meet the requirements of model training. The quality and consistency of the data are ensured by filling missing values, removing outliers, etc.

[0135] The specific steps of data preprocessing are as follows:

[0136] 1. Data cleaning:

[0137] Identify and handle incomplete data:

[0138] Cause: The presence of incomplete data will affect the training effect of the model and even cause the model to fail to run normally.

[0139] Treatment method:

[0140] Delete: If there are too many missing values ​​or the data value is low, you can delete the record directly.

[0141] Filling: Use statistics such as mean, median, mode, etc. to fill in missing numerical values; for categorical features, you can use the mode to fill or introduce a new category "unknown".

[0142] Prediction: Use machine learning models to predict missing values, but this method is computationally expensive and is usually used for features that are sensitive to missing values.

[0143] Identify and remove anomalous data:

[0144] Cause: Abnormal data is often caused by data collection errors, equipment failures, etc., which will interfere with model training.

[0145] Treatment method:

[0146] Statistical-based methods: Use box plots, Z-score, and other methods to identify outliers.

[0147] Based on domain knowledge: judge abnormal data based on business background and common sense.

[0148] To remove duplicate data:

[0149] Reason: Duplicate data will increase the amount of calculation and reduce the generalization ability of the model.

[0150] Treatment method:

[0151] Exact match: directly delete identical records.

[0152] Fuzzy matching: Use similarity measurement methods to delete records with high similarity.

[0153] 2. Data normalization:

[0154] Reason: The value ranges of different features vary greatly. If normalization is not performed, features with larger values ​​will dominate the model training and affect the role of other features.

[0155] method:

[0156] Min-Max Normalization: Scale the features to between [0, 1].

[0157] Z-score standardization: transform the features into a standard normal distribution.

[0158] Other normalization methods: Select an appropriate normalization method based on the distribution characteristics of the data.

[0159] 3. Feature Engineering (optional)

[0160] Feature selection: Select the most useful features for predicting the target variable from the original features, reduce the complexity of the model, and improve the generalization ability of the model.

[0161] Feature construction: By combining original features, new features are constructed to better express the inherent laws of the data.

[0162] Feature dimensionality reduction: When the feature dimension is too high, methods such as principal component analysis (PCA) and linear discriminant analysis (LDA) can be used to reduce the dimension.

[0163] Through the above data preprocessing steps, the quality of the data can be effectively improved, laying a solid foundation for subsequent model training. The specific preprocessing method needs to be selected according to the characteristics of the data and the requirements of the model.

[0164] Benefits of data preprocessing:

[0165] Improve model accuracy: reduce the impact of noise data on the model and improve the generalization ability of the model.

[0166] Accelerate model training: Normalized data can speed up the convergence of the model.

[0167] Enhance model interpretability: Feature engineering can help to better understand the data and thus better explain the results of the model.

[0168] Optionally, the embodiment of the present invention may also perform the following preprocessing operations on the initial training set.

[0169] Data integrity check: First, all data in the initial training set are thoroughly checked to ensure that the data items are complete and there are no empty or missing values.

[0170] Purpose: To ensure that the data is complete and to avoid model training errors or inaccurate results due to missing values.

[0171] Effect: Ensure the quality of model training data and prevent noise from being introduced due to missing values.

[0172] Unified data types: To ensure the accuracy of model training, all numerical data need to be unified into the same numerical type, such as converting floating-point numbers into integers to ensure the consistency of data types.

[0173] Purpose: To make data types consistent and facilitate unified processing of the model.

[0174] Effect: Avoid calculation errors or model training difficulties caused by inconsistent data types.

[0175] Outlier processing: Identify and process outliers in data through statistical analysis and other methods. Common outlier processing methods include deleting outliers, replacing them with the mean or median, or treating them as missing values.

[0176] Purpose: To remove or correct outliers in the data to prevent them from misleading the model.

[0177] Effect: Improve the robustness of the model and make it less sensitive to abnormal data.

[0178] Data consistency check: Check the logical relationship in the data to ensure that there are no contradictions or inconsistencies between the data. For example, the user's date of birth cannot be later than the current date.

[0179] Purpose: To ensure the logical consistency of data and avoid contradictory data.

[0180] Effect: Improve the reliability of data and enhance the interpretability of the model.

[0181] Deduplication: Use appropriate deduplication algorithms to remove duplicate data in the training set to ensure that each sample is unique.

[0182] Purpose: To prevent duplicate data from having too great an impact on model training.

[0183] Effect: Improve the generalization ability of the model and prevent overfitting.

[0184] Data normalization: Normalize data in different ranges to make them in the same numerical range, such as scaling numerical features to between 0 and 1 to facilitate model training.

[0185] Purpose: Scale data from different ranges to the same range so that different features have a more balanced influence on the model.

[0186] Effect: Accelerate model convergence and improve model stability.

[0187] Feature engineering: Based on business needs and model characteristics, feature engineering is performed on the original data to extract features that are useful to the model. For example, a detailed address can be parsed into more fine-grained features such as province, city, and district.

[0188] Purpose: Extract features that are useful to the model, reduce feature dimensions, and improve the expressiveness of the model.

[0189] Effect: Improve the accuracy and generalization ability of the model.

[0190] Data balancing: If the number of samples of different categories in the training data varies greatly, data balancing is required, such as balancing the number of samples through oversampling, undersampling, or SMOTE.

[0191] Purpose: To solve the problem of data category imbalance and prevent the model from being biased towards the majority class.

[0192] Effect: Improve the model's ability to identify minority classes.

[0193] Data label verification: Carefully check the accuracy of data labels to ensure that the labels correspond to the actual data.

[0194] Purpose: Ensure the accuracy of data labels and prevent model training errors caused by label errors.

[0195] Effect: Improve the accuracy of the model.

[0196] Stratified data sampling: When dividing the training set, validation set, and test set, it is necessary to ensure that the sample distribution of each data set is consistent with the distribution of the original data set to avoid data leakage and overfitting problems.

[0197] Purpose: To ensure that the data distribution of the training set, validation set, and test set is consistent to avoid data leakage.

[0198] Effect: Improve the generalization ability of the model and avoid overfitting.

[0199] Through the above data cleaning steps, high-quality training data can be obtained, providing a solid foundation for subsequent model training. These steps can not only improve the accuracy of the model, but also enhance the robustness, generalization ability and interpretability of the model. Therefore, data cleaning is an indispensable part of machine learning projects.

[0200] In an optional embodiment of the present invention, before the step of generating a user category identification model for identifying user categories based on the address training information, the behavior habit training information and the usage period training information, the step further includes:

[0201] Determine a convolutional neural network as an initial model; the initial model includes a convolution kernel;

[0202] Features are extracted from the initial training set by using the convolution kernel.

[0203] Convolutional Neural Network (CNN) is a deep learning model that is particularly good at processing data such as images and videos. Its core advantage is that it can automatically learn useful features from data without manually designing features.

[0204] The role of the convolution kernel:

[0205] Feature Extractor: The convolution kernel can be thought of as a filter that slides over the image to extract specific features. These features can be edges, textures, shapes, etc.

[0206] Weight update: During the training process, the weights of the convolution kernel are continuously updated to better extract features that can distinguish different categories.

[0207] Vectorization: Through convolution operations, the image is converted into a series of feature vectors, which can be used as input for subsequent classifiers.

[0208] Convolutional layer and pooling layer:

[0209] Convolutional Layer:

[0210] Function: Extract local features of the image.

[0211] Process: The convolution kernel slides on the image and performs convolution operation with the pixel values ​​in the local area to obtain a new feature map.

[0212] Features: Different convolution kernels can extract different features, such as vertical edges, horizontal edges, color channels, etc.

[0213] Pooling layer:

[0214] Function: Reduce the dimension of the feature map and the number of parameters while retaining the most important features.

[0215] Method: Commonly used pooling methods include maximum pooling and average pooling. Maximum pooling selects the maximum value in a local area and retains the most significant features; average pooling calculates the average value in a local area and retains the average features.

[0216] Effect: The pooling layer can make the model invariant to slight position changes and improve the robustness of the model.

[0217] Feature Mapping:

[0218] Concept: The feature map output by the convolution layer, each feature map represents a feature.

[0219] Function: The feature map contains rich information about the image, which will be passed to subsequent layers for classification or other tasks.

[0220] Dimension: The dimension of the feature map depends on parameters such as the size of the convolution kernel, stride, and padding.

[0221] Back propagation and weight update:

[0222] Back propagation: Update the weights of the convolution kernel by calculating the gradient of the loss function with respect to the convolution kernel.

[0223] Goal: Minimize the error between the prediction results output by the network and the actual label.

[0224] Process: The error is propagated layer by layer from the output layer to the input layer, the gradient is calculated at each layer, and the parameters are updated.

[0225] The embodiment of the present invention uses CNN as the initial model, and realizes automatic extraction of image features through convolutional layers and pooling layers. The convolution kernel, as a feature extractor, continuously learns during the back propagation process to extract the vector that best represents the image features. The feature map, as the output of the convolutional layer, contains rich information about the image, providing a basis for subsequent classification or other tasks.

[0226] In an optional embodiment of the present invention, the step of generating a user category identification model for identifying user categories based on the address training information, the behavior habit training information and the usage period training information includes:

[0227] Determining refined features for behavioral habit training information from the features extracted by the convolution kernel;

[0228] Calculating the importance reference value of the refined feature through the gradient in the back propagation process; the importance reference value is used to characterize the importance of the refined feature;

[0229] A user category identification model for identifying user categories is generated by using the address training information, the refined features, the importance reference values ​​for the refined features, and the usage period training information.

[0230] In a specific implementation, the embodiment of the present invention can find more specific features that can represent user behavior habits from the features extracted by the convolution kernel.

[0231] For example, the convolution kernels in the trained CNN model can be visualized and analyzed. Convolution kernels usually learn some specific patterns, such as edges, textures, etc. By observing these patterns, it can be inferred which convolution kernels correspond to which behavioral habits. Map the features extracted by the convolution kernel to specific behavioral habits. For example, if a convolution kernel mainly responds to the click behavior of a game app, then it can be considered that the features extracted by this convolution kernel represent the behavioral habit of "game hobby". Perform cluster analysis on the features extracted by the convolution kernel and cluster similar features into one category. In this way, some more abstract and high-level features can be obtained.

[0232] In the embodiment of the present invention, the influence of each refined feature on the model prediction result can also be calculated through the back propagation algorithm to obtain an importance reference value.

[0233] For example, the gradient of the weight corresponding to each feature can be calculated during the back propagation process. The larger the absolute value of the gradient, the greater the impact of the feature on the model output, and the more important it is. The absolute value of the gradient is used as the importance score of the feature. The importance score is normalized to fall between 0 and 1 for easy comparison.

[0234] In the embodiment of the present invention, information such as address, refined features, feature importance, and usage period can also be used to train a new user category recognition model.

[0235] Exemplarily, model training can be implemented in the following manner.

[0236] Feature selection: Select the most important refinement features based on feature importance scores.

[0237] Data preparation: The original data (address, usage period, etc.) and the selected refined features and corresponding importance scores are used as new input data.

[0238] Model training: Use the new dataset to train a new user category recognition model. You can use the same CNN structure as before, or try other types of models, such as XGBoost, LightGBM, etc.

[0239] Model evaluation: Use the test set to evaluate the performance of the new model and compare the accuracy, recall and other indicators of the new and old models.

[0240] The output of the model is a probability value, which indicates the possibility that the user is elderly. If this probability value is high, it can be considered that the user is elderly. In order to train this model, the back propagation algorithm can be used to continuously adjust the parameters of the model so that the model's prediction results are closer and closer to the actual situation.

[0241] Suppose a CNN model is trained to predict whether a user is an elderly person. The input features of the model include:

[0242] Walking speed, hand shaking frequency, mobile phone volume, page dwell time and mobile phone font size.

[0243] Model training process:

[0244] Data preparation: We collected a large number of users’ relevant data, including their age, walking speed, etc.

[0245] Model training: These data are input into the CNN model for training. The model continuously adjusts internal parameters to minimize the error between the predicted age and the actual age.

[0246] Feature importance assessment: After training is completed, the model is analyzed to calculate the impact of each input feature (such as walking speed and hand tremor frequency) on the model output.

[0247] For example:

[0248] Suppose that the gradient value corresponding to the feature "walking speed" is very large. This means that when the walking speed of a sample is slightly changed, the output of the model (i.e. the predicted age) will also change significantly. This shows that "walking speed" has a strong ability to distinguish whether a person is elderly.

[0249] If there is a set of data in which the elderly generally walk slower, while the young walk faster, when the model discovers this pattern during training, it will assign a larger weight to the feature of "walking speed". In this way, when the model encounters a new sample, it will determine which age group the sample belongs to based on the sample's walking speed.

[0250] The purpose of feature importance assessment is to: some target features have an important impact on the result prediction, taking walking speed as an example:

[0251] Biological characteristics: Aging affects a person's physical functions, causing walking speed to slow down.

[0252] Lifestyle habits: Elderly people may prefer to walk slowly due to physical reasons.

[0253] By analyzing the importance of features, the following conclusions can be drawn:

[0254] The decision basis of the model: Understand the features that the model uses to make judgments.

[0255] Optimization of feature engineering: More important features can be collected and processed in a targeted manner to improve the accuracy of the model.

[0256] Model interpretability: Better understanding of how the model works, thus increasing trust in the model.

[0257] By training and analyzing the CNN model, we can find the features that have the greatest impact on the prediction results. These features can help us better understand the behavioral characteristics of the elderly and provide data support for subsequent personalized services.

[0258] In an optional embodiment of the present invention, the user category determination result includes user feature information for characterizing user features, and further includes:

[0259] Constructing a client device control instruction corresponding to the user characteristic information;

[0260] Sending the client device control instruction to the client device;

[0261] The client device is configured to, in response to receiving the client device control instruction, execute the client device control instruction.

[0262] 1. Meaning of user category determination results: Through model analysis, users are classified into categories such as the elderly, young people, children, etc.

[0263] Contains information: In addition to basic category labels (such as “elderly people”), it also contains more fine-grained user feature information.

[0264] 2. Examples of user characteristic information:

[0265] The elderly: poor eyesight, poor hearing, limited mobility, unfamiliarity with new technologies, etc.

[0266] Young people: sensitive to new technologies, socially active, and like personalized settings, etc.

[0267] Children: They are very curious, have short attention span, and need to protect their eyesight, etc.

[0268] 3. Client device control instructions can be used to generate corresponding control instructions for client devices (such as smart phones, smart home devices) based on user feature information.

[0269] Example:

[0270] Poor eyesight: Increase font size, improve contrast, and enable night mode.

[0271] Hard of hearing: Increase the volume, turn on subtitles, use voice assistant.

[0272] People with limited mobility: Simplify the operation steps and provide one-key operation function.

[0273] 4. Method of sending instructions to the client device: Send the instructions to the client device through the network or local communication protocol.

[0274] Protocol: It can be a standard communication protocol (such as MQTT, HTTP) or a custom protocol.

[0275] 5. The client device executes instructions as follows

[0276] Configuration: The client device needs to be pre-configured so that it can receive and execute corresponding control instructions.

[0277] Response: After receiving the instruction, the client device will make corresponding adjustments based on the content of the instruction, for example:

[0278] Smartphones: adjust system settings, launch specific apps, play audio, etc.

[0279] Smart home devices: adjust light brightness, adjust air conditioning temperature, turn on / off home appliances, etc.

[0280] For example, suppose an elderly user is judged to have "poor eyesight" and "poor hearing". The system will generate the following control instructions and send them to the user's smartphone:

[0281] System settings:

[0282] Set the font size to maximum.

[0283] Switch the system theme to high contrast mode.

[0284] Turn on screen reading.

[0285] Application Settings:

[0286] Turn on subtitles in your video player.

[0287] Set the font size of social apps to the maximum.

[0288] Smart Home:

[0289] Turn up the brightness of the living room lights.

[0290] Turn up the volume on your TV.

[0291] In this way, the system can provide personalized services to users based on their characteristics and improve user experience.

[0292] By constructing a client device control instruction having a corresponding relationship with the user characteristic information; sending the client device control instruction to the client device; and the client device being configured to, in response to receiving the client device control instruction, execute the client device control instruction, the following beneficial effects can be achieved:

[0293] Personalization: Provide customized services based on the specific characteristics of users.

[0294] Intelligence: The system can automatically adapt to user changes and provide a more humane experience.

[0295] Universality: can be applied to various smart devices and scenarios.

[0296] In an optional embodiment of the present invention, the user characteristic information is used to characterize the user's visual impairment, and the client device control instruction corresponding to the user characteristic information is a zoom instruction; the client device is configured to, in response to receiving the zoom instruction, enlarge the interface information of the operation interface.

[0297] Exemplarily, the execution process is as follows:

[0298] User feature identification: Through user behavior data (such as click error rate, reading time, etc.) and device usage habits (such as font size setting, screen brightness adjustment, etc.), the user category identification model determines whether the user has visual impairment. Alternatively, the user can actively declare their vision condition during the initial setup.

[0299] Generate scaling instructions:

[0300] Once the model confirms that the user has a visual impairment, the system will automatically generate a "zoom instruction" to instruct the device to enlarge the screen content, including icon enlargement and text enlargement.

[0301] The scaling can be adjusted based on the user's level of visual impairment.

[0302] Instruction execution:

[0303] After the client device receives the zoom instruction, it will execute it immediately and enlarge all the content on the screen to a size suitable for the user to read.

[0304] Zooming can be done for the entire screen or for a specific app or page.

[0305] Specific implementation method:

[0306] The format of the zoom command can be a simple value indicating the zoom ratio, or a more complex command containing information such as the zoom center and zoom speed.

[0307] Scaling algorithm: The client device can use different scaling algorithms, such as linear scaling, smooth scaling, etc.

[0308] User interaction: Users can adjust the zoom ratio manually or control it through gestures or voice commands.

[0309] Adaptive zoom: The system can automatically adjust the zoom ratio according to different application scenarios and content. For example, when reading an e-book, the font can be enlarged to the optimal reading size.

[0310] Technical implementation details:

[0311] Model training: Training a model that can accurately identify users’ visual impairments requires a large amount of labeled data.

[0312] Instruction sending: Instructions can be sent through network protocols (such as HTTP, MQTT) or APIs provided by the operating system.

[0313] Client implementation: The client device needs to implement a command parsing module that can parse the received commands and perform corresponding operations.

[0314] By configuring the client device to, in response to receiving the zoom instruction, enlarge the interface information of the operation interface, the advantages are:

[0315] Improve user experience: For visually impaired users, enlarging screen content can significantly improve the convenience of reading and operation.

[0316] Personalized service: The system can provide customized services based on the user's specific circumstances.

[0317] Automation: The entire process can be automated without manual user operation.

[0318] In an optional embodiment of the present invention, the user characteristic information is used to characterize the hearing impairment of the user, and the client device control instruction corresponding to the user characteristic information is a volume control instruction; the client device is configured to increase the volume in response to receiving the volume control instruction.

[0319] Exemplarily, the execution process is as follows:

[0320] User hearing impairment identification:

[0321] Direct feedback: Users can proactively declare their hearing status in device settings.

[0322] Behavior analysis: The user category recognition model determines whether the user may have hearing impairment by analyzing the user's operating behavior on the device (such as frequently adjusting the volume, using subtitles, etc.).

[0323] Speech recognition: The system determines hearing status by analyzing the characteristics of the user's speech (such as volume, timbre, clarity, etc.).

[0324] Generate volume control instructions:

[0325] Once the system confirms that the user has hearing impairment, it will automatically generate an instruction to "increase the volume".

[0326] The volume boost can be adjusted according to the degree of hearing impairment of the user.

[0327] Instruction execution:

[0328] When the client device receives the volume control command, it will immediately execute it and increase the volume of the device to a level suitable for the user's hearing.

[0329] Volume adjustment can be for the system global volume, specific application volume or media volume.

[0330] Specific implementation method:

[0331] Volume control command format: It can be a simple value indicating the volume level; or it can be a more complex command containing information such as volume increase and volume upper limit.

[0332] Volume adjustment algorithm: The client device can use different volume adjustment algorithms such as linear adjustment and logarithmic adjustment.

[0333] User interaction: Users can adjust the volume manually or through voice commands or gestures.

[0334] Adaptive adjustment: The system can automatically adjust the volume according to different audio content and ambient noise to provide the best listening experience.

[0335] Technical implementation details:

[0336] Model training: Training a model that can accurately identify a user’s hearing impairment requires a large amount of labeled data.

[0337] Instruction sending: Instructions can be sent through network protocols (such as HTTP, MQTT) or APIs provided by the operating system.

[0338] Client implementation: The client device needs to implement a command parsing module that can parse the received commands and perform corresponding operations.

[0339] By configuring the client device to increase the volume in response to receiving the volume control instruction, the following beneficial effects can be achieved.

[0340] Improve user experience: For hearing-impaired users, increasing the volume can improve the hearing experience and make it easier for them to interact with the device.

[0341] Personalized service: The user category recognition model can provide customized services based on the user's specific hearing conditions.

[0342] Automation: The entire process can be automated without manual user operation.

[0343] Application scenarios for different client devices:

[0344] Smartphone: calls, music, videos, etc.

[0345] Smart speakers: voice interaction, listening to news, listening to music, etc.

[0346] Smart home devices: voice control, reminders, etc.

[0347] Summarize:

[0348] By combining user feature information with client device control instructions, we can provide hearing-impaired users with a friendlier and more convenient digital experience. This personalized service can not only improve the quality of life of users, but also reflect the care of technology for vulnerable groups in society.

[0349] In order to enable those skilled in the art to better understand the embodiments of the present invention, a complete example is used for illustration below.

[0350] refer to Figure 2 , Figure 2 It is a schematic diagram of a flow chart of an operation interface switching method provided in an embodiment of the present invention;

[0351] 1. Users need to be able to quickly switch to senior mode. A core challenge is how to ensure that the user is elderly. To this end, we have adopted a series of strategies to confirm whether the user is elderly. In the first step, we will collect the user's address, user behavior, user usage habits, and active time periods. After collecting the data, a pop-up window with a clear font and prominent UI will pop up when the system is started for the first time to ensure that people of all ages can see it clearly. We will use this pop-up window to apply for authorization to ensure that the user allows us to upload and analyze the data. Finally, we will also obtain some common behaviors of the elderly through other forms.

[0352] 2. User address analysis: whether the user is located in a nursing home, retirement home, park, hospital, hospital department, etc. with complete elderly care and infrastructure;

[0353] User behavior and user usage habits mainly analyze the user's operation speed, user sliding speed, user click frequency, user stay time, user font size and user external sound volume, etc.

[0354] The active time period is mainly used to analyze the active time of users. The elderly usually use it in the morning and evening.

[0355] 3. In the process of building the elderly recognition system, we deeply explored the technical means of data cleaning, feature extraction, model training and optimization to improve the accuracy of recognition and the complexity of the system.

[0356] First, during the data cleaning phase, we not only removed invalid and incomplete data, but also normalized the data to make it meet the requirements of model training. We ensured the quality and consistency of the data by filling missing values ​​and removing outliers.

[0357] In addition, we extract useful features from the raw data, such as the postal code of the address, the type and number of surrounding facilities, residential density, the number of user operations, operation speed, user sliding time, sliding speed, click frequency, and user stay time.

[0358] In terms of feature extraction, we use convolution kernels in convolutional neural networks (CNN) to extract features. These convolution kernels are continuously updated during the back-propagation process to approximate the vector set that best represents the data features. CNN extracts image features through convolution layers and pooling layers, where the convolution layer is responsible for extracting features, while the pooling layer extracts the maximum value of the local block to form a feature map.

[0359] During the model training phase, we built a CNN model consisting of an input layer, a hidden layer, and an output layer. The input layer receives the preprocessed feature vector, the hidden layer is responsible for extracting higher-level features, and the output layer uses a sigmoid activation function to output the probability that the user belongs to the elderly group. To optimize the model, we used hyperparameter optimization techniques to find the best model parameters through methods such as Bayesian optimization. In terms of model optimization, we evaluated the importance of each input feature by analyzing the gradient during backpropagation.

[0360] During the recognition process, the input data includes the elderly’s walking speed, hand shaking frequency, mobile phone volume, page dwell time, and mobile phone font size. These data pass through each layer of the network, and the neurons in each layer will be weighted and summed with the input data, and passed through the activation function to finally generate a predicted output to predict whether the user is an elderly person. The difference between the predicted output and the actual output is measured by the loss function. This difference reflects the current network’s predictive performance for elderly recognition. Starting from the output layer, the error of each layer is back-propagated to the previous layer. The goal of this step is to calculate the partial derivative (i.e., gradient) of the loss function with respect to each weight. In the elderly recognition system, this involves calculating the impact of each feature (such as walking speed and hand shaking frequency, etc.) on the loss function. During the back-propagation process, we first calculate the partial derivative of the loss function with respect to the output of the output layer neuron. Then, using the chain rule, the partial derivative of the output layer is propagated back to the hidden layer. For each neuron in the hidden layer, the partial derivative of its output with respect to the input of the next layer neuron is calculated and multiplied with the partial derivative returned by the next layer to accumulate the total partial derivative of the neuron with respect to the loss function. This means we can assess the extent to which features such as walking speed and hand tremor frequency influence the model’s prediction of elderly identity.

[0361] For example, if the partial derivative of walking speed is large, this indicates that walking speed is a more important feature for distinguishing between elderly and non-elderly people. Finally, we use these partial derivatives to update the weights of the network. The weights are updated in the opposite direction of the partial derivatives because we want to reduce the value of the loss function. Through this process, we can understand the extent to which each input feature (such as walking speed, hand shake frequency) affects the model's prediction results, and then evaluate its importance.

[0362] This method can help us understand which input features contribute more to the output changes, and accordingly screen out the features with greater contributions for retraining to improve the accuracy of the model in predicting the probability of being elderly.

[0363] In CNN, the convolution kernel is updated through the back-propagation algorithm. In each iteration, the output error of the network is back-propagated to each layer of the network, the gradient of each convolution kernel is calculated, and then these gradients are used to update the weights of the convolution kernel. This process involves calculating the dot product of the gradient and the input data, and then updating the weights according to the gradient descent or other optimization algorithms to minimize the loss function of the network. In addition, we also consider personalized services and dynamic adjustments.

[0364] By analyzing user behavior data, we can provide customized services and dynamically adjust service content according to changes in user behavior to improve user experience. For example, customized services are provided based on the health of the elderly; for example, for elderly people who are not very sensitive to sound, the sound will be played outward and the volume will be increased when playing audio and video; for elderly people with less flexible fingers, the UI will be enlarged; for elderly people with blurred vision, the UI will also be enlarged, and the font size, color, etc. will be adjusted.

[0365] Finally, to improve the robustness of the model, we used ensemble learning methods, gradient boosting machines, which improve the accuracy and robustness of predictions by building an ensemble of multiple models. We also considered transfer learning, a method of applying knowledge learned from one domain to another, especially in cases where data is scarce, to improve the adaptability and robustness of the model. Through the comprehensive application of these technical means, our system is able to accurately identify the elderly population.

[0366] 4. After successfully training the elderly recognition model, we applied it to actual scenarios and analyzed the collected user data to predict whether the user belongs to the elderly group. We will distinguish different elderly groups based on health and physical changes, such as vision problems, hearing loss, stiff finger joints caused by motor function loss, decreased finger flexibility, etc. When the elderly encounter these problems, we will enter different elderly interfaces according to different problems, such as enlarging fonts, enhancing external sound, simplifying operation procedures, voice control and high-contrast interfaces. For healthy elderly people, we do not enter the elderly mode.

[0367] Reference Figure 3 , shows a flowchart of another method for switching operation interfaces provided in an embodiment of the present invention, which may specifically include the following steps:

[0368] Step 301, obtaining the user's address information, behavior habit information and usage time period information;

[0369] Step 302, sending the address information, the behavior habit information and the usage period information to the cloud server; the cloud server is configured to: when receiving the address information, the behavior habit information and the usage period information sent by the client device, generate a user category determination result for the user through the user category recognition model based on the address information, the behavior habit information and the usage period information; when the user is determined to be an elderly person based on the user category determination result, send a control signal to the client device;

[0370] Step 303: In response to receiving the control signal, switch the operation interface to the elderly mode.

[0371] An embodiment of the present invention is applied to a client device, wherein the client device is configured with a corresponding cloud server, wherein the client device is provided with an operation interface, and wherein the cloud server is configured to: obtain an initial training set, wherein the initial training set includes address training information, behavior habit training information, and usage period training information; and generate a user category recognition model for identifying user categories based on the address training information, the behavior habit training information, and the usage period training information.

[0372] As for the client device embodiment, since it is basically similar to the cloud server embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the cloud server embodiment.

[0373] It should be noted that, for the sake of simplicity, the method embodiments are described as a series of action combinations, but those skilled in the art should be aware that the embodiments of the present invention are not limited by the order of the actions described, because according to the embodiments of the present invention, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily required by the embodiments of the present invention.

[0374] Reference Figure 4 , shows a structural block diagram of an operation interface switching device provided in an embodiment of the present invention, which may specifically include the following modules:

[0375] An initial training set acquisition module 401 is used to acquire an initial training set, wherein the initial training set includes address training information, behavior habit training information, and usage period training information;

[0376] The user category identification model generation module 402 is used to generate a user category identification model for identifying user categories based on the address training information, the behavior habit training information and the usage period training information; the client device is configured to: obtain the user's address information, behavior habit information and usage period information; send the address information, the behavior habit information and the usage period information to the cloud server;

[0377] A user category determination result generating module 403 is used to generate a user category determination result for the user through the user category recognition model based on the address information, the behavior habit information and the usage period information sent by the client device when receiving the address information, the behavior habit information and the usage period information;

[0378] The control signal sending module 404 is used to send a control signal to the client device when the user is determined to be an elderly person based on the user category determination result; the client device is configured to: in response to receiving the control signal, switch the operation interface to the elderly mode.

[0379] Reference Figure 5 , shows a structural block diagram of another operation interface switching device provided in an embodiment of the present invention, which may specifically include the following modules:

[0380] Information acquisition module 501, used to acquire the user's address information, behavior habit information and usage period information;

[0381] The information sending module 502 is used to send the address information, the behavior habit information and the usage period information to the cloud server; the cloud server is configured to: when receiving the address information, the behavior habit information and the usage period information sent by the client device, generate a user category determination result for the user through the user category recognition model based on the address information, the behavior habit information and the usage period information; when the user is determined to be an elderly person based on the user category determination result, send a control signal to the client device;

[0382] The operation interface switching module 503 is used to switch the operation interface to the elderly mode in response to receiving the control signal.

[0383] As for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0384] In addition, an embodiment of the present invention further provides an electronic device, comprising: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, the various processes of the above-mentioned operation interface switching method embodiment are implemented, and the same technical effect can be achieved. To avoid repetition, it will not be described here.

[0385] The embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, each process of the above-mentioned operation interface switching method embodiment is implemented, and the same technical effect can be achieved. To avoid repetition, it is not repeated here. The computer-readable storage medium is, for example, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0386] Figure 6 A schematic diagram of the hardware structure of an electronic device for implementing various embodiments of the present invention.

[0387] The electronic device 600 includes but is not limited to: a radio frequency unit 601, a network module 602, an audio output unit 603, an input unit 604, a sensor 605, a display unit 606, a user input unit 607, an interface unit 608, a memory 609, a processor 610, and a power supply 611. Those skilled in the art will appreciate that Figure 6 The electronic device structure shown in the figure does not constitute a limitation on the electronic device, and the electronic device may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently. In the embodiments of the present invention, the electronic device includes but is not limited to a mobile phone, a tablet computer, a laptop computer, a PDA, a vehicle-mounted terminal, a wearable device, and a pedometer.

[0388] It should be understood that in the embodiment of the present invention, the radio frequency unit 601 can be used for receiving and sending signals during information transmission or communication. Specifically, after receiving downlink data from the base station, it is sent to the processor 610 for processing; in addition, uplink data is sent to the base station. Generally, the radio frequency unit 601 includes but is not limited to an antenna, at least one amplifier, a transceiver, a coupler, a low noise amplifier, a duplexer, etc. In addition, the radio frequency unit 601 can also communicate with the network and other devices through a wireless communication system.

[0389] The electronic device provides users with wireless broadband Internet access through the network module 602, such as helping users to send and receive emails, browse web pages, and access streaming media.

[0390] The audio output unit 603 can convert the audio data received by the RF unit 601 or the network module 602 or stored in the memory 609 into an audio signal and output it as sound. Moreover, the audio output unit 603 can also provide audio output related to a specific function performed by the electronic device 600 (for example, a call signal reception sound, a message reception sound, etc.). The audio output unit 603 includes a speaker, a buzzer, a receiver, etc.

[0391] The input unit 604 is used to receive audio or video signals. The input unit 604 may include a graphics processor (GPU) 6041 and a microphone 6042, and the graphics processor 6041 processes the image data of a static picture or video obtained by an image capture device (such as a camera) in a video capture mode or an image capture mode. The processed image frame can be displayed on the display unit 606. The image frame processed by the graphics processor 6041 can be stored in the memory 609 (or other storage medium) or sent via the radio frequency unit 601 or the network module 602. The microphone 6042 can receive sound and can process such sound into audio data. The processed audio data can be converted into a format output that can be sent to a mobile communication base station via the radio frequency unit 601 in the case of a telephone call mode.

[0392] The electronic device 600 also includes at least one sensor 605, such as a light sensor, a motion sensor, and other sensors. Specifically, the light sensor includes an ambient light sensor and a proximity sensor, wherein the ambient light sensor can adjust the brightness of the display panel 6061 according to the brightness of the ambient light, and the proximity sensor can turn off the display panel 6061 and / or the backlight when the electronic device 600 is moved to the ear. As a type of motion sensor, the accelerometer sensor can detect the magnitude of acceleration in each direction (generally three axes), and can detect the magnitude and direction of gravity when stationary, which can be used to identify the posture of the electronic device (such as horizontal and vertical screen switching, related games, magnetometer posture calibration), vibration recognition related functions (such as pedometer, tapping), etc.; the sensor 605 can also include a fingerprint sensor, a pressure sensor, an iris sensor, a molecular sensor, a gyroscope, a barometer, a hygrometer, a thermometer, an infrared sensor, etc., which will not be repeated here.

[0393] The display unit 606 is used to display information input by the user or information provided to the user. The display unit 606 may include a display panel 6061, which may be configured in the form of a liquid crystal display (LCD), an organic light-emitting diode (OLED), or the like.

[0394] The user input unit 607 can be used to receive input digital or character information, and to generate key signal input related to user settings and function control of the electronic device. Specifically, the user input unit 607 includes a touch panel 6071 and other input devices 6072. The touch panel 6071, also known as a touch screen, can collect the user's touch operation on or near it (such as the user's operation on the touch panel 6071 or near the touch panel 6071 using any suitable object or accessory such as a finger, stylus, etc.). The touch panel 6071 may include two parts: a touch detection device and a touch controller. Among them, the touch detection device detects the user's touch orientation, detects the signal brought by the touch operation, and transmits the signal to the touch controller; the touch controller receives the touch information from the touch detection device, converts it into contact coordinates, and then sends it to the processor 610, receives the command sent by the processor 610 and executes it. In addition, the touch panel 6071 can be implemented in various types such as resistive, capacitive, infrared and surface acoustic waves. In addition to the touch panel 6071, the user input unit 607 may also include other input devices 6072. Specifically, other input devices 6072 may include but are not limited to a physical keyboard, function keys (such as volume control keys, switch keys, etc.), a trackball, a mouse, and a joystick, which will not be described in detail here.

[0395] Furthermore, the touch panel 6071 may be overlaid on the display panel 6061. When the touch panel 6071 detects a touch operation on or near it, it is transmitted to the processor 610 to determine the type of touch event. Then, the processor 610 provides corresponding visual output on the display panel 6061 according to the type of touch event. Figure 6 In the figure, the touch panel 6071 and the display panel 6061 are used as two independent components to realize the input and output functions of the electronic device. However, in some embodiments, the touch panel 6071 and the display panel 6061 can be integrated to realize the input and output functions of the electronic device, which is not limited here.

[0396] The interface unit 608 is an interface for connecting an external device to the electronic device 600. For example, the external device may include a wired or wireless headset port, an external power supply (or battery charger) port, a wired or wireless data port, a memory card port, a port for connecting a device with an identification module, an audio input / output (I / O) port, a video I / O port, a headphone port, etc. The interface unit 608 may be used to receive input (e.g., data information, power, etc.) from an external device and transmit the received input to one or more elements within the electronic device 600 or may be used to transmit data between the electronic device 600 and an external device.

[0397] The memory 609 can be used to store software programs and various data. The memory 609 can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), etc. In addition, the memory 609 can include a high-speed random access memory, and can also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0398] The processor 610 is the control center of the electronic device. It uses various interfaces and lines to connect various parts of the entire electronic device. By running or executing software programs and / or modules stored in the memory 609 and calling data stored in the memory 609, it performs various functions of the electronic device and processes data, thereby monitoring the electronic device as a whole. The processor 610 may include one or more processing units; preferably, the processor 610 may integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user interface, and application programs, and the modem processor mainly processes wireless communications. It is understandable that the above-mentioned modem processor may not be integrated into the processor 610.

[0399] The electronic device 600 may also include a power supply 611 (such as a battery) for supplying power to each component. Preferably, the power supply 611 may be logically connected to the processor 610 through a power management system, thereby implementing functions such as charging, discharging, and power consumption management through the power management system.

[0400] In addition, the electronic device 600 includes some functional modules not shown, which will not be described in detail here.

[0401] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0402] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for enabling a terminal (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0403] like Figure 7 As shown, in another embodiment provided by the present invention, a computer-readable storage medium 701 is also provided, in which instructions are stored. When the computer-readable storage medium is run on a computer, the computer executes the operation interface switching method described in the above embodiment.

[0404] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation modes, which are merely illustrative rather than restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the present invention and the claims, all of which are within the protection of the present invention.

[0405] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed in the embodiments of the present invention can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0406] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0407] In the embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as 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 mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0408] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0409] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0410] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, ROM, RAM, magnetic disks, or optical disks.

[0411] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. An operation interface switching method, characterized in that: The method is applied to a cloud server, the cloud server is configured with a corresponding client device, and the client device is provided with an operation interface, including: Acquire an initial training set, wherein the initial training set includes address training information, behavior habit training information, and usage period training information; Generate a user category recognition model for identifying user categories based on the address training information, the behavior habit training information and the usage period training information; the client device is configured to: obtain the user's address information, behavior habit information and usage period information; send the address information, the behavior habit information and the usage period information to the cloud server; When receiving the address information, the behavior habit information and the usage period information sent by the client device, generating a user category determination result for the user through the user category recognition model based on the address information, the behavior habit information and the usage period information; When the user is determined to be an elderly person based on the user category determination result, a control signal is sent to the client device; the client device is configured to: in response to receiving the control signal, switch the operation interface to the elderly mode.

2. The method according to claim 1, characterized in that Before the step of generating a user category identification model for identifying user categories based on the address training information, the behavior habit training information and the usage period training information, the step further includes: For incomplete data, abnormal data and duplicate data in the initial training set; Performing a filling operation on the incomplete data; Performing a removal operation on the abnormal data and the duplicate data; A normalization operation is performed on the initial training set.

3. The method according to claim 1 or 2, characterized in that: Before the step of generating a user category identification model for identifying user categories based on the address training information, the behavior habit training information and the usage period training information, the step further includes: Determine a convolutional neural network as an initial model; the initial model includes a convolution kernel; Features are extracted from the initial training set by using the convolution kernel.

4. The method according to claim 3, characterized in that The step of generating a user category identification model for identifying user categories based on the address training information, the behavior habit training information and the usage period training information comprises: Determining refined features for behavioral habit training information from the features extracted by the convolution kernel; Calculating the importance reference value of the refined feature through the gradient in the back propagation process; the importance reference value is used to characterize the importance of the refined feature; A user category identification model for identifying user categories is generated by using the address training information, the refined features, the importance reference values ​​for the refined features, and the usage period training information.

5. The method according to claim 4, characterized in that The user category determination result includes user feature information used to characterize user features, and also includes: Constructing a client device control instruction corresponding to the user characteristic information; Sending the client device control instruction to the client device; The client device is configured to, in response to receiving the client device control instruction, execute the client device control instruction.

6. The method according to claim 5, characterized in that The user characteristic information is used to characterize the user's visual impairment, and the client device control instruction corresponding to the user characteristic information is a zoom instruction; the client device is configured to, in response to receiving the zoom instruction, enlarge the interface information of the operation interface.

7. The method according to claim 5, characterized in that The user characteristic information is used to characterize the hearing impairment of the user, and the client device control instruction corresponding to the user characteristic information is a volume control instruction; the client device is configured to increase the volume in response to receiving the volume control instruction.

8. An operation interface switching method, characterized in that: The method is applied to a client device, the client device is configured with a corresponding cloud server, the client device is provided with an operation interface, and the cloud server is configured to: obtain an initial training set, the initial training set includes address training information, behavior habit training information and usage period training information; Generate a user category recognition model for identifying user categories based on the address training information, the behavior habit training information and the usage period training information; the method includes: Obtain the user's address information, behavior information, and usage time information; sending the address information, the behavior habit information and the usage period information to the cloud server; the cloud server is configured to: upon receiving the address information, the behavior habit information and the usage period information sent by the client device, generate a user category determination result for the user through the user category recognition model based on the address information, the behavior habit information and the usage period information; and when the user is determined to be an elderly person based on the user category determination result, send a control signal to the client device; In response to receiving the control signal, the operation interface is switched to the elderly mode.

9. An operation interface switching device, characterized in that: The device is applied to a cloud server, the cloud server is configured with a corresponding client device, and the client device is provided with an operation interface, including: An initial training set acquisition module, used to acquire an initial training set, wherein the initial training set includes address training information, behavior habit training information, and usage period training information; A user category identification model generation module is used to generate a user category identification model for identifying user categories based on the address training information, the behavior habit training information and the usage period training information; the client device is configured to: obtain the user's address information, behavior habit information and usage period information; send the address information, the behavior habit information and the usage period information to the cloud server; A user category determination result generating module, configured to generate a user category determination result for the user through the user category recognition model based on the address information, the behavior habit information and the usage period information sent by the client device when receiving the address information, the behavior habit information and the usage period information; A control signal sending module is used to send a control signal to the client device when the user is determined to be an elderly person based on the user category determination result; the client device is configured to: in response to receiving the control signal, switch the operation interface to the elderly mode.

10. An operation interface switching device, characterized in that: The device is applied to a client device, the client device is configured with a corresponding cloud server, the client device is provided with an operation interface, and the cloud server is configured to: obtain an initial training set, the initial training set includes address training information, behavior habit training information and usage period training information; Generate a user category recognition model for identifying user categories based on the address training information, the behavior habit training information and the usage period training information; the device includes: Information acquisition module, used to obtain the user's address information, behavior habit information and usage time information; an information sending module, for sending the address information, the behavior habit information and the usage period information to the cloud server; the cloud server is configured to: upon receiving the address information, the behavior habit information and the usage period information sent by the client device, generate a user category determination result for the user through the user category recognition model based on the address information, the behavior habit information and the usage period information; and when the user is determined to be an elderly person based on the user category determination result, send a control signal to the client device; An operation interface switching module is used to switch the operation interface to the elderly mode in response to receiving the control signal.

11. An electronic device, characterized in that: It includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus; The memory is used to store computer programs; The processor is used to implement the method according to any one of claims 1-7 or 8 when executing the program stored in the memory.

12. A computer-readable storage medium having instructions stored thereon, which, when executed by one or more processors, cause the processors to perform the method according to any one of claims 1 to 7 or 8.

Images