Application recommendation system, method, storage medium, and vehicle
By acquiring driving suggestion data and health data, and using application-based modeling to automatically recommend applications, the problem of the in-vehicle interface being unable to intelligently match user needs has been solved, reducing user operation time and improving user experience.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- FAURECIA (CHINA) HOLDING CO LTD
- Filing Date
- 2025-01-26
- Publication Date
- 2026-07-28
AI Technical Summary
The existing in-vehicle infotainment interface cannot intelligently match the user's application needs, causing users to have to perform cumbersome operations to find the required application, which takes a long time.
By acquiring driving advice data and the health data of the target occupants in the vehicle, the system uses an application identification model to determine recommended applications and displays recommended application icons, reducing manual operation by the user.
Automatically identify and display recommended apps based on users' different states, reducing user search and interaction time and improving user experience.
Smart Images
Figure CN122470802A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of automotive technology, and more particularly to an application recommendation system, method, storage medium, and vehicle. Background Technology
[0002] As vehicles become increasingly intelligent, they can offer users more services during driving, such as music, navigation, and communication. These services are primarily provided through applications, meaning that the corresponding applications need to be installed in the vehicle before using any of these services.
[0003] Currently, different users have different application needs for in-vehicle infotainment systems, but existing interfaces often fail to intelligently match these needs, requiring users to perform cumbersome operations to find the desired applications. Clearly, the current application installation method is time-consuming. Summary of the Invention
[0004] This application provides an application recommendation system, method, storage medium, and vehicle that can reduce the time users spend searching for and interacting with applications.
[0005] To achieve the above objectives, this application adopts the following technical solution:
[0006] Firstly, an application recommendation system for an automotive application store is provided, comprising: an acquisition module for acquiring application determination data. The application determination data includes driving suggestion data and health data of target occupants within the vehicle; the driving suggestion data is generated based on the driver's driving behavior data. A processing module for determining recommended applications based on the application determination data. A display module for displaying the recommended applications.
[0007] In one possible implementation, the processing module is specifically used to: determine multiple groups of candidate applications based on application determination data, and perform weighted processing on the initial application scores of each candidate application in the multiple groups of candidate applications to obtain a target application score for each candidate application. Candidate applications whose target application scores meet the application determination criteria are determined as recommended applications.
[0008] In one possible implementation, the processing module is specifically used to: input application-determining data into each of multiple application-determining models to obtain multiple sets of candidate applications. One application-determining model outputs a set of candidate applications, and the multiple application-determining models are trained based on different training data.
[0009] In one possible implementation, the application-determined data includes numerical data and character data. The processing module is specifically used to: determine the first feature value of the numerical data and the second feature value of the character data. The first feature value is extracted after normalizing and standardizing the numerical data, and the second feature value is the binary feature value obtained by one-hot encoding the character data. The first and second feature values are input into each of the multiple application-determining models to obtain multiple sets of candidate applications. In another possible implementation, the acquisition module is specifically used to: acquire the health data of the target personnel through a cockpit monitoring system. The health data includes one or more of the following: gender, age, heart rate, respiratory rate, blood oxygen saturation, and mood.
[0010] In one possible implementation, the application-determined data may also include one or more of the following: driving behavior data of the driver in the vehicle; applications installed in the vehicle; application data of applications not installed in the vehicle on the application platform; the vehicle's location and weather data for that location.
[0011] Secondly, a method for recommending applications in an automotive app store is provided. The method includes: acquiring application determination data. The application determination data includes driving suggestion data and health data of target occupants within the vehicle. The driving suggestion data is generated based on the driver's driving behavior data. Recommended applications are determined based on the application determination data. The recommended applications are then displayed. In one possible implementation, determining recommended applications based on the application determination data includes: identifying multiple groups of candidate applications based on the application determination data, and weighting the initial application scores of each candidate application in the multiple groups of candidate applications to obtain a target application score for each candidate application. Candidate applications whose target application scores meet the application determination criteria are identified as recommended applications.
[0012] In one possible implementation, multiple sets of candidate applications are determined based on application-determining data, including: inputting the application-determining data into each of multiple application-determining models to obtain multiple sets of candidate applications. One application-determining model outputs a set of candidate applications, and the multiple application-determining models are trained based on different training data.
[0013] In one possible implementation, the application determination data includes numerical data and character data. Multiple candidate applications are determined based on this data, including: determining a first feature value for the numerical data and a second feature value for the character data. The first feature value is extracted after normalizing and standardizing the numerical data, and the second feature value is a binary feature value obtained by one-hot encoding the character data. The first and second feature values are then input into each of the multiple application determination models to obtain multiple candidate applications.
[0014] Thirdly, an application recommendation device for an automotive application store is provided, the device comprising: an acquisition unit, a determination unit, and a display unit.
[0015] The acquisition unit is used to acquire application determination data. The application determination data includes driving suggestion data and health data of the target occupants in the vehicle. The driving suggestion data is generated based on the driver's driving behavior data. The determination unit is used to determine recommended applications based on the application determination data. The display unit is used to display the recommended applications. In one possible implementation, the determination unit is specifically used to: determine multiple groups of candidate applications based on the application determination data, and perform weighted processing on the initial application scores of each candidate application in the multiple groups of candidate applications to obtain a target application score for each candidate application. Candidate applications whose target application scores meet the application determination criteria are determined as recommended applications.
[0016] In one possible implementation, the determining unit is specifically used to: input application determining data into each of multiple application determining models to obtain multiple sets of candidate applications. One application determining model outputs a set of candidate applications, and the multiple application determining models are trained based on different training data.
[0017] In one possible implementation, the application determination data includes numerical data and character data. The determination unit is specifically used to: determine a first feature value for the numerical data and a second feature value for the character data. The first feature value is extracted after normalizing and standardizing the numerical data, and the second feature value is a binary feature value obtained by one-hot encoding the character data. The first and second feature values are input into each of the multiple application determination models to obtain multiple sets of candidate applications. One application determination model outputs a set of candidate applications, and any two application determination models are trained based on different training data.
[0018] Fourthly, an electronic device is provided, including a memory and a processor. The memory and the processor are coupled. The memory is used to store computer program code, which includes computer instructions. When the processor executes the computer instructions, it causes the electronic device to perform the method as described in the second aspect.
[0019] Fifthly, a computer-readable storage medium is provided, which stores computer instructions. When the computer instructions are executed on an electronic device, they cause the electronic device to perform the method as described in the second aspect.
[0020] This application provides an application recommendation system for an automotive app store, which offers at least the following benefits: It acquires driving advice data and health data of target occupants within the vehicle. Furthermore, it determines recommended applications based on driving behavior and the target occupants' health data. This allows for the identification and display of corresponding recommended applications based on the user's different statuses, eliminating the need for manual user intervention and reducing the time spent searching for and interacting with applications. Attached Figure Description
[0021] Figure 1 This is a schematic diagram of an application recommendation system architecture provided by an embodiment of this application;
[0022] Figure 2 This is a schematic diagram illustrating the collection of application-specific data provided in an embodiment of this application;
[0023] Figure 3 This is one of the schematic diagrams showing a recommended application provided in an embodiment of this application;
[0024] Figure 4 This is a second schematic diagram showing a recommended application provided in an embodiment of this application;
[0025] Figure 5 This is an application recommendation diagram provided by an embodiment of this application;
[0026] Figure 6 This is a schematic diagram of the download process for the recommended application provided in the embodiments of this application;
[0027] Figure 7 This is one of the schematic diagrams illustrating the process of determining the recommended application provided in the embodiments of this application;
[0028] Figure 8 This is a schematic diagram illustrating the model training process of an exemplary embodiment provided in this application;
[0029] Figure 9 This is a second schematic diagram illustrating the process of determining a recommended application, provided by an embodiment of this application.
[0030] Figure 10 This is the third schematic diagram illustrating the process of determining a recommended application, provided by an embodiment of this application.
[0031] Figure 11 This is a block diagram illustrating an application recommendation device as provided in an exemplary embodiment of this application;
[0032] Figure 12 This is a block diagram illustrating an electronic device provided by an embodiment of this application. Detailed Implementation
[0033] The technical solutions in the embodiments of this application will now be described with reference to the accompanying drawings.
[0034] In the description of this application, unless otherwise stated, " / " means "or". For example, A / B can mean A or B. "And / or" in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, and B alone. Furthermore, "at least one" and "multiple" refer to two or more. The terms "first," "second," etc., do not limit the quantity or order of execution, and "first," "second," etc., do not necessarily imply differences.
[0035] The inventive concept of this application is described below:
[0036] As the background technology points out, different users have different application needs for in-vehicle systems, but existing in-vehicle system interfaces often fail to intelligently match user needs, requiring users to perform cumbersome operations to find the required applications. Therefore, the current application installation method is time-consuming.
[0037] To address the aforementioned technical problems, this application provides an application recommendation system for an automotive application store. The system includes: an acquisition module for acquiring application determination data, which includes driving suggestion data and health data of target occupants within the vehicle; the driving suggestion data is generated based on the behavioral parameters of the occupants driving the vehicle; a processing module for determining recommended applications based on the application determination data; and a display module for displaying the recommended applications.
[0038] In this way, driving suggestion data and health data of the target occupants in the vehicle are obtained. Furthermore, based on driving behavior and the target occupants' health data, recommended applications are determined. This allows for the identification and display of corresponding recommended applications based on the user's different statuses, eliminating the need for manual user operation and reducing the time spent searching for and interacting with applications.
[0039] Based on the above inventive concept, such as Figure 1 As shown, Figure 1 This application illustrates an application recommendation system 100 for an automotive application marketplace, comprising: an acquisition module 101, a processing module 102, and a display module 103. The processing module 102 is communicatively connected to both the acquisition module 101 and the display module 103.
[0040] In some embodiments, the application recommendation system 100 is applied to an in-vehicle terminal. The in-vehicle terminal can be an electronic device.
[0041] The acquisition module 101 is used to acquire application-defined data.
[0042] The application identifies data including driving advice data and health data of the target occupants in the vehicle. The driving advice data is generated based on the behavioral data of the person driving the vehicle.
[0043] In some embodiments, the acquisition module 101 is used to acquire application-determined data while the vehicle is in operation.
[0044] In other embodiments, the acquisition module 101 is used to acquire application determination data in response to an application recommendation operation.
[0045] In some embodiments, the acquisition module 101 is used to acquire health data of a target person through a cockpit monitoring system. The health data includes one or more of the following: gender, age, heart rate, respiratory rate, blood oxygen saturation, and mood.
[0046] For example, if the target person is a driver, the acquisition module 101 is used to acquire the driver's health data through a driver monitoring system (DMS). As another example, if the target person is a passenger, the acquisition module 101 is used to acquire the passenger's health data through an occupancy monitoring system (OMS).
[0047] In some embodiments, the acquisition module 101 is used to acquire health data of target individuals from a health database.
[0048] In this embodiment of the application, emotions can be text-based data. For example, emotions can be anger, disgust, fear, happiness, sadness, or surprise.
[0049] In this embodiment of the application, the driving suggestion data may include vehicle driving suggestions and / or mood relief suggestions.
[0050] In some embodiments, the processing module 102 is used to generate driving suggestion data based on the vehicle's acceleration information, driving speed information, and braking information. For example, when the vehicle's acceleration is greater than a preset acceleration, the processing module 102 generates a vehicle driving suggestion text of "accelerate slowly". As another example, when the vehicle's acceleration is greater than a preset acceleration and the number of sudden braking events is greater than a preset number, the processing module 102 generates an emotion-soothing suggestion text of "calm down".
[0051] It should be noted that the preset acceleration and preset number of times are pre-configured by the maintenance personnel, and can also be determined according to the driving environment of the vehicle. This application embodiment does not limit this.
[0052] In another scenario, processing module 102 is used to generate driving suggestion data based on the health data of the target person in the health database.
[0053] In this embodiment of the application, the application determination data may also include one or more of the following: driving behavior data of the driver in the vehicle, application data of applications installed in the vehicle, application data of applications not installed in the vehicle on the application platform, the location of the vehicle and the meteorological data of the location.
[0054] Driving behavior data may include one or more of the following: steering frequency, steering angle, accelerator pedal opening, brake pedal opening, or driving speed. Data for installed applications may include one or more of the following: online time, application type, usage frequency, or installation duration. Data for applications not installed may include one or more of the following: download count, application rating, application category, or application name.
[0055] In some embodiments, such as Figure 2 As shown, Figure 2 A schematic diagram illustrating the collection of data for an application is shown. Figure 2 In the process, the acquisition module 101 acquires application data for uninstalled applications sent by the application platform server: download count, application rating, application category, and application name. The DMS stores the received health data in a health database. This health data includes the driver's gender, age, heart rate, respiratory rate, blood oxygen saturation, and mood. Furthermore, the health database generates driving suggestions based on the target person's health data and sends these suggestions to the acquisition module 101. The acquisition module 101 also acquires data collected from the vehicle's in-vehicle system, including the online duration of installed applications, current driving behavior data, vehicle status, historical behavior data, location information, and weather information.
[0056] In some embodiments, the acquisition module 101 is used to acquire raw vehicle data and obtain application-determined data based on the raw vehicle data. For example, the acquisition module 101 is used for vehicle positioning data, vehicle DMS data, and health data from a health database.
[0057] For example, module 101 is used to input vehicle positioning data into the location determination model and output the vehicle's location. As another example, module 101 is used to input vehicle DMS data into the information determination model and output age and gender.
[0058] It should be noted that the location determination model and information determination model are pre-configured by the operations and maintenance personnel.
[0059] Understandably, module 101 is used to acquire information from five dimensions: user profile, user habits and preferences, user health status, user driving habits, and application market information.
[0060] Processing module 102 is used to determine recommended applications based on application determination data.
[0061] In some embodiments, the processing module 102 is configured to determine the recommended application type based on the application determination data, and to determine the recommended application based on the recommended application type.
[0062] In some embodiments, the processing module 102 is configured to determine multiple groups of candidate applications based on application determination data, and to perform weighted processing on each candidate application in the multiple groups of candidate applications to obtain an application score for each candidate application. The processing module 102 is further configured to determine candidate applications whose application scores meet the application determination conditions as recommended applications.
[0063] In some embodiments, the processing module 102 is used to input application determination data into each of the multiple application determination models to obtain multiple sets of candidate applications. Each application determination model outputs a set of candidate applications, and any two application determination models are trained based on different training data.
[0064] In some embodiments, processing module 102 is configured to normalize and standardize the numerical data in the application determination data to obtain processed numerical data, and determine a first feature value of the processed numerical data. Further, processing module 102 is configured to perform one-hot encoding of the character data in the application determination data, and determine a second feature value of the one-hot encoding. Subsequently, processing module 102 is further configured to input the first feature value and the second feature value into each of the multiple application determination models to obtain multiple sets of candidate applications.
[0065] In some embodiments, the processing module 102 is further configured to preprocess the data in the application-determined data. For example, the processing module 102 is further configured to input the application-determined data into a preprocessing model to obtain application-determined data with missing values and outliers removed.
[0066] It should be noted that the preprocessing model is pre-configured by the operations and maintenance personnel.
[0067] Display module 103 is used to display recommended applications.
[0068] In this embodiment, the recommended application can be an identifier for the recommended application. The identifier is used to launch or install the recommended application. For example, the identifier may include an application icon, link information, etc.
[0069] In some embodiments, processing module 102, upon determining a recommended application, retrieves the download link of the recommended application on the application platform based on the application's name, and generates an identifier for the recommended application based on the download link on the application platform. Further, processing module 102 instructs display module 103 to display the identifier of the recommended application. Correspondingly, display module 103 displays the identifier of the recommended application. Thus, the application recommendation system can respond to a user clicking the identifier of a recommended application, downloading and installing the recommended application for user convenience, thereby improving the user experience.
[0070] For example, such as Figure 3 As shown, the display module 103 is used to display the identifiers of application A, application B, and application C, as well as the installation prompt message "The following applications have been recommended for you. Click to install and use them".
[0071] In some embodiments, processing module 102 is configured to, upon determining a recommended application, download and install the recommended application from the application platform according to its name. Further, processing module 102 is configured to instruct display module 103 to display the identifier of the recommended application. Correspondingly, display module 103 is configured to display the identifier of the recommended application. Thus, the application recommendation system can respond to clicking the identifier of the recommended application to launch the recommended application, facilitating user access and improving the user experience.
[0072] For example, such as Figure 4 As shown, the display module 103 is used to display the identifiers of application A, application B, and application C, as well as the installation prompt message "The following applications have been matched and installed for you. Click to use them".
[0073] To better understand the application recommendation process of the application recommendation system 100 provided in this application embodiment, such as Figure 5 As shown, in Figure 5The diagram illustrates an application recommendation system: A health database collects user data including: gender: male, age: 22, and heart rate: 120. This user data is obtained through a DMS camera. Further, based on the user data in the health database, driving advice text is generated: "It is recommended to relax, listen to some light music, and play games during rest stops." Multiple application types are retrieved from the application server: Application Type 1: Music Application 1 and Music Application 2; Application Type 2: Game Application 1 and Social Application 1; Application Type 3: Office Application 1 and Office Application 2, etc. The following vehicle information is obtained from the in-vehicle system: Installed applications: Map Application 1 and Video Playback Application 1; Current driving status: Parked; Current location: Parking lot. Finally, the application recommendation system 100 recommends three applications based on the vehicle information, the multiple application types, and the driving advice text: Music Application 1, Music Application 2, and Game Application 1. These three applications were selected by the application recommendation system 100 through a voting process. The icons of the three recommended applications are displayed, and each application is associated with its corresponding download link, allowing users to install and use them by clicking to download.
[0074] In some embodiments of this application, the recommended application download process is as follows: Figure 6 As shown, it includes: S601-S603.
[0075] S601. Obtain the identifier of the recommended application.
[0076] S602. Search for the download link of the recommended application from the application store database based on the application's identifier.
[0077] S603. Download the recommended application using the download link provided.
[0078] The application recommendation system 100 provided in this application embodiment offers at least the following beneficial effects: It determines recommended applications based on driving behavior and the target user's health data. This allows for the identification of corresponding recommended applications based on the user's different states, eliminating the need for manual user operation and reducing the time spent searching for and interacting with applications.
[0079] In some embodiments, to obtain more accurate applications, the processing module is specifically configured to: determine multiple groups of candidate applications based on application determination data, and perform weighted processing on the initial application scores of each candidate application in the multiple groups of candidate applications to obtain a target application score for each candidate application. Further, candidate applications whose target application scores meet the application determination criteria are determined as recommended applications.
[0080] In this embodiment, the initial application score is a pre-configured value. For example, the initial application score can be 6 or 10; this embodiment does not specifically limit this. Different candidate applications can have the same or different initial application scores; this embodiment does not limit this either.
[0081] In this embodiment of the application, the application determination condition can be that the application score is greater than or equal to a preset application score, or that the application score ranking is greater than or equal to a preset threshold. This embodiment of the application does not limit the specific criteria.
[0082] It should be noted that the default application rating is pre-configured by the operations and maintenance personnel.
[0083] In some embodiments, the processing module 102 is configured to: input application determination data into each of a plurality of application determination models to obtain a plurality of candidate applications. One application determination model outputs a set of candidate applications, and the plurality of application determination models are trained based on different training data.
[0084] In some embodiments, the processing module 102 is configured to: apply feature values of the determined data and input the feature values of the determined data into each of the multiple application determination models to obtain multiple sets of candidate applications. Further, the processing module 102 is configured to perform weighted processing on the initial application scores of each candidate application in the multiple sets of candidate applications to obtain a target application score for each candidate application. Further, the processing module 102 is configured to determine candidate applications whose application scores meet the application determination conditions as recommended applications.
[0085] In some embodiments, the processing module 102 is configured to: normalize and standardize the numerical data in the application determination data to obtain processed numerical data, and determine a first feature value of the processed numerical data. Further, the processing module 102 is configured to perform one-hot encoding on the character data to obtain a second feature value; the second feature value is a binary feature value. Subsequently, the processing module 102 is configured to input the first feature value and the second feature value into each of the multiple application determination models to obtain multiple sets of candidate applications. It should be noted that the weights corresponding to each set of candidate applications can be the same or different, and this embodiment of the application does not limit this.
[0086] Understandably, multiple candidate applications are identified based on the application-specific data. The initial application scores of each candidate application are then weighted, and a recommended application is derived based on the target application scores of the candidate applications. This method of identifying recommended applications from multiple candidates leads to a more accurate selection of recommended applications.
[0087] In one design, in order to determine the matching recommended application, the processing module 102 is specifically used to: input application determination data into each of the multiple application determination models to obtain multiple sets of candidate applications.
[0088] The application determination model for the application determination data input in this embodiment is a trained application determination model.
[0089] In some embodiments, the applied deterministic model can be a neural network model, a combined prediction algorithm model, or a support vector machine model, etc., and this application does not specifically limit this. For example, the applied deterministic model can be a nonlinear multi-module decision model.
[0090] In some embodiments, multiple applications may determine that the model can be a model of the same type or a model of different types, and this application embodiment does not limit this.
[0091] In some embodiments, the number of application-determined models is not specifically limited in this application; it can be 3, 5, or 2. Furthermore, the number of applications in each group of candidate applications is not limited; it can be 3, 4, or 2.
[0092] For example, such as Figure 7 As shown, Figure 7 The example demonstrates three application determination models. Application determination data is input into each of these three models: Application Determination Model 1, Application Determination Model 2, and Application Determination Model 3, resulting in three candidate application groups: Candidate Application Group 1, Candidate Application Group 2, and Candidate Application Group 3. Furthermore, each candidate application in these three groups is weighted, and a vote is taken among the weighted candidate applications to obtain the recommended application.
[0093] Before inputting the application determination into multiple application determination models, the application determination initial model is trained in this embodiment to obtain multiple application determination models.
[0094] Specifically, such as Figure 8 As shown, Figure 8 A schematic diagram of a model training process is shown, including: S701-S704.
[0095] S701. Obtain the sample dataset.
[0096] In some embodiments, each sample data in the sample dataset includes: age, gender, health data, driving behavior data, application data of applications installed on the vehicle, application data of applications not installed on the vehicle in the application platform, location, weather, and calibrated applications.
[0097] S702. Group the sample dataset.
[0098] In some embodiments, the sample dataset is grouped according to the number of initial models determined by the application. For example, if the application determines that the number of initial models is 3, the sample dataset is divided into 3 groups. Or, for example, if the application determines that the number of models is 4, the sample dataset is divided into 4 groups.
[0099] Furthermore, the grouped sample data is categorized into training sample data groups and validation sample data groups. For example, two sets of sample data are obtained: sample data group 1 and sample data group 2. Further, sample data group 1 is configured as the training sample data group, and sample data group 2 is configured as the validation sample data group. Thus, sample data group 1 is used to train the initial model 1 for a given application, and sample data group 2 is used to validate the trained initial model 1. Conversely, sample data group 2 is configured as the training sample data group, and sample data group 1 is configured as the validation sample data group. Thus, sample data group 2 is used to train the initial model 2 for a given application, and sample data group 1 is used to validate the trained initial model 2.
[0100] It should be noted that the number of sample data in each sample data group can be the same or different. Therefore, this application embodiment does not limit the number of sample data in each sample data group.
[0101] S703. Based on the grouped sample dataset, train the initial model for the application.
[0102] Specifically, the initial model is trained using the training sample data set, and its parameters are optimized using gradient descent and a loss function. Subsequently, the initial model is validated using the validation sample data set; if the loss value of the loss function stabilizes, module optimization is stopped.
[0103] For example, let's take the application of determining the initial model as a nonlinear initial model. With three nonlinear initial models configured: nonlinear initial model 1, nonlinear initial model 2, and nonlinear initial model 3, each nonlinear initial model is trained and validated based on its training sample data set and validation sample data set, respectively, to obtain the trained nonlinear model and its corresponding weights.
[0104] S704. Determine the application model.
[0105] Understandably, by inputting application-specific data into different models to obtain multiple sets of candidate applications, and then determining the recommended application from these multiple sets of candidate applications, the accuracy of application recommendations can be improved.
[0106] In one design, to quickly obtain recommendation results, the processing module 102 in this embodiment is specifically used for: normalizing and standardizing the numerical data in the application determination data to obtain processed numerical data, and determining a first feature value of the processed numerical data; performing one-hot encoding on the character data in the application determination data, and determining a second feature value of the one-hot encoding; and inputting the first feature value and the second feature value into each of the multiple application determination models to obtain multiple sets of candidate applications.
[0107] In some embodiments, the processing module 102 classifies the application-defined data to obtain numerical data and character data. Further, the processing module 102 normalizes and standardizes the numerical data, and extracts features from the processed numerical data to obtain a first feature value.
[0108] For example, normalize and standardize numerical data such as age, heart rate, respiratory rate, and temperature in the application's defined data.
[0109] In this embodiment of the application, in order to avoid the influence of a certain value on the application of a certain model, the deviation Min-Max standardization formula is used to scale the data to the same scale. The deviation standardization formula is shown in Formula 1 below.
[0110]
[0111] Where X′ is the normalized data of X, X is any data in the numerical data before normalization, min(X) is the smallest value in the numerical data, and max(X) is the largest value in the numerical data.
[0112] In this embodiment of the application, the data is also standardized to facilitate feature extraction. For example, the data is converted into a standard normal distribution with a mean of 0 and a standard deviation of 1. The standardization formula is shown in Formula 2 below.
[0113]
[0114] Where X′ represents the standardized data, X represents any data point in the unnormalized numerical data, σ represents the standard deviation of all data in the numerical data, and μ represents the mean of all data in the numerical data.
[0115] In some embodiments, the processing module 102 is configured to input the preprocessed application determination data into a preset feature extraction model and output binary data features. Further, the processing module 102 is configured to input the binary data features into the application determination model, output the recommended application number, and obtain the link to the recommended application based on the recommended application number.
[0116] Additionally, the processing module 102 is used to determine the one-hot encoding of character data and to determine the second feature value of the one-hot encoding. Thus, for non-numerical data, it is converted into a binary vector using one-hot encoding for use in the model, accelerating data processing and leading to faster recommendation applications.
[0117] For example, determining one-hot encoding for character data such as emotion, location, driving suggestion text, and application type in the application's defined data.
[0118] In some embodiments, before normalizing and standardizing the application determination data, the processing module 102 is also used to preprocess the original application determination data.
[0119] For example, preprocessing includes missing data handling and exception handling.
[0120] For missing data handling: If the processing module 102 detects that a certain data in the application-determined data is missing, the corresponding preset data will be used to fill the missing data.
[0121] In some embodiments, when the processing module 102 detects that the original data corresponding to a certain parameter value is missing, it deletes the original data corresponding to the parameter value and determines the preset value corresponding to the parameter value as the target value corresponding to the parameter value.
[0122] For anomaly handling: When the processing module 102 detects that a certain data in the application-determined data is an anomaly, a nonlinear regression algorithm is used to process outliers that are too large or too small, so as to obtain reasonable data.
[0123] Understandably, data preprocessing is used to clean and transform raw data in order to improve data quality.
[0124] In one design, to reduce the time users spend searching for and interacting with applications, this application provides a flowchart illustrating an application recommendation method for an automotive application store, as shown below. Figure 9 As shown, the method includes: S801-S803.
[0125] S801, Obtain application-defined data.
[0126] The application determines data including driving suggestion data and health data of the target occupants in the vehicle. The driving suggestion data is generated based on the behavioral parameters of the driving vehicle.
[0127] In some embodiments, the electronic device obtains application determination data in response to an application recommendation instruction. The method by which the application determination data parameters are obtained by the aforementioned acquisition module 101 is not described further here.
[0128] S802. Determine recommended applications based on application-specific data.
[0129] As one possible approach, electronic devices determine recommended applications based on application-specific data and application-specific models.
[0130] In some embodiments, when an electronic device acquires application-determining data, it preprocesses the data to obtain preprocessed data and extracts data features from the preprocessed data. Further, the electronic device inputs these data features into an application-determining model to obtain recommended applications.
[0131] In some embodiments, the electronic device determines multiple groups of candidate applications based on application determination data. Each candidate application in the multiple groups is weighted to obtain an application score for each candidate application. Further, the electronic device determines candidate applications whose application scores meet the application determination criteria as recommended applications.
[0132] In some embodiments, the electronic device inputs application determination data into each of a plurality of application determination models to obtain multiple sets of candidate applications.
[0133] In some embodiments, the electronic device normalizes and standardizes the numerical data in the application determination data to obtain processed numerical data, and determines a first feature value of the processed numerical data. Further, it determines the one-hot encoding of the character data in the application determination data and determines a second feature value of the one-hot encoding. Subsequently, the electronic device inputs the first feature value and the second feature value into each of multiple application determination models to obtain multiple sets of candidate applications.
[0134] For details on data preprocessing, feature extraction, and determination of recommended applications in this step, please refer to the data processing method in the above-mentioned processing module 102, which will not be repeated here.
[0135] In another scenario, the application determines whether the data includes health data and mood. The electronic device uses the target person's health data and mood to determine their current state, and then recommends applications based on this current state and the corresponding target relationship. The target relationship includes multiple states and the recommended application for each state.
[0136] S803, Display recommended applications.
[0137] For details on the recommended display methods in this step, please refer to the display module 102 mentioned above. They will not be repeated here.
[0138] In one design, the above S802 includes: S8021-S8023.
[0139] S8021. Determine multiple candidate applications based on application-specific data.
[0140] S8022. The initial application scores of each candidate application in multiple candidate applications are weighted to obtain the target application score of each candidate application.
[0141] S8023. Candidate applications whose target application scores meet the application determination criteria are identified as recommended applications.
[0142] In one design, the above S8021 includes: S901.
[0143] S901. Input the application determination data into each of the multiple application determination models to obtain multiple sets of candidate applications.
[0144] In one design, the above S8021 includes: S902-S905.
[0145] S902. Normalize and standardize the numerical data in the application-defined data to obtain the processed numerical data.
[0146] S903. Determine the first characteristic value of the processed numerical data.
[0147] S904. Perform one-hot encoding on the character data to obtain the second feature value.
[0148] The second characteristic value is a binary characteristic value.
[0149] S905. Input the first feature value and the second feature value into each of the multiple application determination models to obtain multiple sets of candidate applications.
[0150] To better understand the application recommendation methods provided in the embodiments of this application, such as Figure 9 As shown, a flowchart of an application recommendation method is illustrated, including: S1001-S1005.
[0151] S1001, Obtain application-defined data.
[0152] Specifically, such as Figure 10 As shown, Figure 10 The data used to identify applications includes: online time spent with installed applications, a list of installed applications, age, gender, mood, suggested text, and daily behavior data. Figure 10 In the data, age and gender are obtained from the in-vehicle DMS; suggested text is obtained from health data in the health database; and daily behavior data is obtained from game applications, work applications, music applications, map applications, and communication applications.
[0153] S1002. Preprocess the data determined by the application.
[0154] S1003. Extract the data features of the data determined by the application.
[0155] S1004. Input the data feature data into the nonlinear decision model to obtain multiple recommendation applications.
[0156] S1005: Displays multiple recommended applications.
[0157] The foregoing mainly describes the solutions provided by the embodiments of this application from a methodological perspective. To achieve the above functions, the control device includes hardware structures and / or software modules corresponding to the execution of each function. Those skilled in the art should readily recognize that, based on the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein, this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed by hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.
[0158] Figure 11 This is an application recommendation device for an automotive application store, illustrated according to an exemplary embodiment. (Refer to...) Figure 11 The application recommendation device 110 includes: an acquisition unit 1101, a determination unit 1102, and a display unit 1103.
[0159] Get unit 1101, get application-defined data.
[0160] The application determines data including driving suggestion data and health data of the target occupants in the vehicle. The driving suggestion data is generated based on the behavioral parameters of the driving vehicle.
[0161] Unit 1102 determines the recommended application based on the application determination data.
[0162] Display unit 1103 displays recommended applications.
[0163] The recommended application is used to launch or install recommended applications.
[0164] In one possible implementation, the determining unit 1102 is specifically used to: determine multiple groups of candidate applications based on application determination data, and perform weighted processing on the initial application scores of each candidate application in the multiple groups of candidate applications to obtain a target application score for each candidate application. Candidate applications whose target application scores meet the application determination conditions are determined as recommended applications.
[0165] In one possible implementation, the determining unit 1102 is specifically used to: input application determining data into each of the multiple application determining models to obtain multiple sets of candidate applications; one application determining model is used to output a set of candidate applications, and the multiple application determining models are trained based on different training data.
[0166] In one possible implementation, the application determination data includes numerical data and character data. The determination unit 1102 is specifically used to: determine the first feature value of the numerical data and the second feature value of the character data. The first feature value and the second feature value are input into each of the multiple application determination models to obtain multiple sets of candidate applications; one application determination model is used to output a set of candidate applications, and any two application determination models in the multiple application determination models are trained based on different training data.
[0167] In one possible implementation, the determining unit 1102 is specifically used to: perform one-hot encoding on character data to obtain a second feature value; the second feature value is a binary feature value.
[0168] Figure 12 This is a block diagram illustrating an electronic device according to an exemplary embodiment. Figure 12 As shown, the electronic device includes, but is not limited to, a processor 1201 and a memory 1202.
[0169] The memory 1202 described above is used to store the executable instructions of the processor 1201. It is understood that the processor 1201 is configured to execute instructions to implement the model training method and SOC estimation method in the above embodiments.
[0170] It should be noted that those skilled in the art will understand that Figure 12 The electronic device structure shown does not constitute a limitation on the electronic device; the electronic device may include, but is not limited to, other electronic devices. Figure 12 This may indicate more or fewer components, or combinations of certain components, or different component arrangements.
[0171] Processor 1201 is the control center of the electronic device. It connects various parts of the electronic device via various interfaces and lines. By running or executing software programs and / or modules stored in memory 1202, and by calling data stored in memory 1202, it performs various functions and processes data, thereby providing overall monitoring of the electronic device. Processor 1201 may include one or more processing units. Optionally, processor 1201 may integrate an application processor and a modem processor. The application processor mainly handles the operating system, user interface, and applications, while the modem processor mainly handles wireless communication. It is understood that the modem processor may not be integrated into processor 1201.
[0172] The memory 1202 can be used to store software programs and various data. The memory 1202 may primarily include a program storage area and a data storage area. The program storage area may store the operating system, application programs required by at least one functional module (such as a determination unit, processing unit, etc.), etc. Furthermore, the memory 1202 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device.
[0173] In an exemplary embodiment, a computer-readable storage medium including instructions is also provided, such as a memory 1202 including instructions, which can be executed by a processor 1201 of an electronic device to implement the methods in the above embodiments.
[0174] Optionally, the computer-readable storage medium may be a non-transitory computer-readable storage medium, such as a read-only memory (ROM), random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device.
[0175] In an exemplary embodiment, this application also provides a computer program product including one or more instructions, which can be executed by the processor 1201 of the processing device to perform the methods described above.
[0176] It should be noted that when one or more instructions in the computer-readable storage medium or computer program product are executed by the processor of the processing device, they implement the various processes of the above method embodiments and achieve the same technical effect as the above method. To avoid repetition, they will not be described again here.
[0177] Through the above description of the embodiments, those skilled in the art can clearly understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.
[0178] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another apparatus, or some features may be ignored or not executed. Furthermore, the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.
[0179] The units described as separate components may or may not be physically separate. A component shown as a unit can be one or more physical units; that is, it can be located in one place or distributed in multiple different locations. Some or all of the classified units can be selected to achieve the purpose of this embodiment, depending on actual needs.
[0180] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.
[0181] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a readable storage medium. Based on this understanding, the technical solution of the embodiments of this application, essentially, or the part that contributes to the prior art, or a complete or partial classification of the technical solution, can be embodied in the form of a software product. This software product is stored in a storage medium and includes several instructions to cause a device (which may be a microcontroller, chip, etc.) or processor to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, ROM, RAM, magnetic disks, or optical disks.
[0182] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.
Claims
1. An application recommendation system for an automotive application marketplace, characterized in that, include: The acquisition module is used to acquire data determined by the application. The application determines data including driving suggestion data and health data of the target occupants in the vehicle, wherein the driving suggestion data is generated from the driving behavior data of the driver of the vehicle. The processing module is used to determine recommended applications based on the application determination data. The display module is used to display the recommended applications.
2. The system according to claim 1, characterized in that, The processing module is specifically used for: Based on the application determination data, multiple groups of candidate applications are determined, and the initial application scores of each candidate application in the multiple groups of candidate applications are weighted to obtain the target application score of each candidate application. Candidate applications whose target application scores meet the application determination criteria are identified as recommended applications.
3. The system according to claim 2, characterized in that, The processing module is specifically used for: The application determination data is input into each of the multiple application determination models to obtain the multiple sets of candidate applications; one application determination model is used to output a set of candidate applications, and the multiple application determination models are trained based on different training data.
4. The system according to claim 3, characterized in that, The application determines that the data includes numerical data and character data, and the processing module is specifically used for: Determine a first feature value for the numerical data and a second feature value for the character data; the first feature value is extracted after normalizing and standardizing the numerical data, and the second feature value is a binary feature value obtained by one-hot encoding the character data. The first feature value and the second feature value are input into each of the multiple application determination models to obtain the multiple sets of candidate applications.
5. The system according to any one of claims 1-4, characterized in that, The acquisition module is specifically used for: The health data of the target personnel is acquired through the cockpit monitoring system; the health data includes one or more of the following: gender, age, heart rate, respiratory rate, blood oxygen saturation, and mood.
6. The system according to any one of claims 1-4, characterized in that, The application-defined data also includes one or more of the following: The driving behavior data of the driver inside the vehicle; The vehicle has the following applications installed; Application data of vehicles without installed applications as described in the application platform; The location of the vehicle and the meteorological data for that location.
7. An application recommendation method for automotive application stores, characterized in that, The method includes: Acquire application-determined data; the application-determined data includes driving suggestion data and health data of target occupants in the vehicle, wherein the driving suggestion data is generated based on the driving behavior data of the vehicle's driver; Recommended applications are determined based on the application data. Display the identifier of the recommended application.
8. The method according to claim 7, characterized in that, The step of determining recommended applications based on the application determination data includes: Based on the application determination data, multiple groups of candidate applications are determined, and the initial application scores of each candidate application in the multiple groups of candidate applications are weighted to obtain the target application score of each candidate application. Candidate applications whose target application scores meet the application determination criteria are identified as recommended applications.
9. A computer-readable storage medium, characterized in that, When the computer-executable instructions stored in the computer-readable storage medium are executed by the processor of the processing device, the processing device is capable of performing the method as described in claim 7 or 8.
10. An electronic device, characterized in that, include: Processor and memory; The memory stores instructions that the processor can execute; When the processor is configured to execute the instructions, the electronic device performs the method as described in claim 7 or 8.