CN111797878A - Data processing method, data processing device, storage medium and electronic equipment - Google Patents
Data processing method, data processing device, storage medium and electronic equipment Download PDFInfo
- Publication number
- CN111797878A CN111797878A CN201910282467.2A CN201910282467A CN111797878A CN 111797878 A CN111797878 A CN 111797878A CN 201910282467 A CN201910282467 A CN 201910282467A CN 111797878 A CN111797878 A CN 111797878A
- Authority
- CN
- China
- Prior art keywords
- user
- data
- users
- similarity
- target
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
- G06F18/2155—Generating training patterns; Bootstrap methods, e.g. bagging or boosting characterised by the incorporation of unlabelled data, e.g. multiple instance learning [MIL], semi-supervised techniques using expectation-maximisation [EM] or naïve labelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Software Systems (AREA)
- Medical Informatics (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The application discloses a data processing method, a data processing device, a storage medium and electronic equipment. The method comprises the following steps: determining a target task type of machine learning; acquiring similarity information between users corresponding to preset user sample data, wherein the similarity information between the users is used for representing the similarity between different users; determining a similarity threshold according to the target task type; selecting data of a target user from preset user sample data according to the inter-user similarity information and the similarity threshold; and performing machine learning of the target task according to the data of the target user. The embodiment can improve the adaptability between the sample data for learning training acquired by the electronic equipment and the machine learning training task.
Description
Technical Field
The present application belongs to the technical field of electronic devices, and in particular, to a data processing method, apparatus, storage medium, and electronic device.
Background
With the development of artificial intelligence technology, machine learning can be performed on electronic equipment, so that a trained algorithm model is obtained. In the related art, when performing machine learning, the electronic device needs to acquire a large amount of sample data and perform machine learning according to the sample data. However, in the related art, when the electronic device acquires sample data for performing machine learning, the acquired sample data has poor adaptability to a machine learning training task.
Disclosure of Invention
The embodiment of the application provides a data processing method and device, a storage medium and electronic equipment, which can improve the adaptability between the sample data acquired by the electronic equipment and used for learning training and a machine learning training task.
An embodiment of the present application provides a data processing method, including:
determining a target task type of machine learning;
acquiring similarity information between users corresponding to preset user sample data, wherein the similarity information between the users is used for representing the similarity between different users;
determining a similarity threshold according to the target task type;
selecting data of a target user from preset user sample data according to the inter-user similarity information and the similarity threshold;
and performing machine learning of the target task according to the data of the target user.
An embodiment of the present application provides a data processing apparatus, including:
the first determination module is used for determining a target task type of machine learning;
the device comprises an acquisition module, a processing module and a display module, wherein the acquisition module is used for acquiring similarity information between users corresponding to preset user sample data, and the similarity information between the users is used for representing the similarity between different users;
the second determining module is used for determining a similarity threshold according to the target task type;
the selecting module is used for selecting the data of the target user from preset user sample data according to the similarity information among the users and the similarity threshold;
and the learning module is used for performing machine learning of the target task according to the data of the target user.
The embodiment of the application provides a storage medium, wherein a computer program is stored on the storage medium, and when the computer program is executed on a computer, the computer is enabled to execute the data processing method provided by the embodiment of the application.
The embodiment of the present application further provides an electronic device, which includes a memory and a processor, where the processor is configured to execute the data processing method provided in the embodiment of the present application by calling the computer program stored in the memory.
In this embodiment, when the electronic device performs a task of machine learning, data of a target user may be selected from preset sample data according to the type of the task and the similarity information between users in the sample, and machine learning of the task may be performed according to the data of the target user. Therefore, compared with a mode of randomly selecting sample data for machine learning in the related art, the embodiment of the application can select the data for machine learning according to the task type of machine learning and the similarity information between users in the sample, so that the adaptability between the sample data for learning training acquired by the electronic device and the machine learning training task can be improved.
Drawings
The technical solutions and advantages of the present application will become apparent from the following detailed description of specific embodiments of the present application when taken in conjunction with the accompanying drawings.
Fig. 1 is a schematic diagram of a panoramic sensing architecture of an electronic device provided in an embodiment of the present application.
Fig. 2 is a schematic flowchart of a data processing method according to an embodiment of the present application.
Fig. 3 is another schematic flow chart of a data processing method according to an embodiment of the present application.
Fig. 4 to fig. 5 are schematic scene diagrams of a data processing method according to an embodiment of the present application.
Fig. 6 is a schematic structural diagram of a data processing apparatus according to an embodiment of the present application.
Fig. 7 is a schematic structural diagram of an electronic device provided in an embodiment of the present application.
Fig. 8 is another schematic structural diagram of an electronic device according to an embodiment of the present application.
Detailed Description
Referring to the drawings, wherein like reference numbers refer to like elements, the principles of the present application are illustrated as being implemented in a suitable computing environment. The following description is based on illustrated embodiments of the application and should not be taken as limiting the application with respect to other embodiments that are not detailed herein.
Referring to fig. 1, fig. 1 is a schematic view of an application scenario of a data processing method according to an embodiment of the present application. The data processing method can be applied to electronic equipment. A panoramic perception framework is arranged in the electronic equipment. The panoramic sensing architecture is an integration of hardware and software for implementing the data processing method in an electronic device.
The panoramic perception architecture comprises an information perception layer, a data processing layer, a feature extraction layer, a scene modeling layer and an intelligent service layer.
The information perception layer is used for acquiring information of the electronic equipment or information in an external environment. The information-perceiving layer may include a plurality of sensors. For example, the information sensing layer includes a plurality of sensors such as a distance sensor, a magnetic field sensor, a light sensor, an acceleration sensor, a fingerprint sensor, a hall sensor, a position sensor, a gyroscope, an inertial sensor, an attitude sensor, a barometer, and a heart rate sensor.
Among other things, a distance sensor may be used to detect a distance between the electronic device and an external object. The magnetic field sensor may be used to detect magnetic field information of the environment in which the electronic device is located. The light sensor can be used for detecting light information of the environment where the electronic equipment is located. The acceleration sensor may be used to detect acceleration data of the electronic device. The fingerprint sensor may be used to collect fingerprint information of a user. The Hall sensor is a magnetic field sensor manufactured according to the Hall effect, and can be used for realizing automatic control of electronic equipment. The location sensor may be used to detect the geographic location where the electronic device is currently located. Gyroscopes may be used to detect angular velocity of an electronic device in various directions. Inertial sensors may be used to detect motion data of an electronic device. The gesture sensor may be used to sense gesture information of the electronic device. A barometer may be used to detect the barometric pressure of the environment in which the electronic device is located. The heart rate sensor may be used to detect heart rate information of the user.
And the data processing layer is used for processing the data acquired by the information perception layer. For example, the data processing layer may perform data cleaning, data integration, data transformation, data reduction, and the like on the data acquired by the information sensing layer.
The data cleaning refers to cleaning a large amount of data acquired by the information sensing layer to remove invalid data and repeated data. The data integration refers to integrating a plurality of single-dimensional data acquired by the information perception layer into a higher or more abstract dimension so as to comprehensively process the data of the plurality of single dimensions. The data transformation refers to performing data type conversion or format conversion on the data acquired by the information sensing layer so that the transformed data can meet the processing requirement. The data reduction means that the data volume is reduced to the maximum extent on the premise of keeping the original appearance of the data as much as possible.
The characteristic extraction layer is used for extracting characteristics of the data processed by the data processing layer so as to extract the characteristics included in the data. The extracted features may reflect the state of the electronic device itself or the state of the user or the environmental state of the environment in which the electronic device is located, etc.
The feature extraction layer may extract features or process the extracted features by a method such as a filtering method, a packing method, or an integration method.
The filtering method is to filter the extracted features to remove redundant feature data. Packaging methods are used to screen the extracted features. The integration method is to integrate a plurality of feature extraction methods together to construct a more efficient and more accurate feature extraction method for extracting features.
The scene modeling layer is used for building a model according to the features extracted by the feature extraction layer, and the obtained model can be used for representing the state of the electronic equipment, the state of a user, the environment state and the like. For example, the scenario modeling layer may construct a key value model, a pattern identification model, a graph model, an entity relation model, an object-oriented model, and the like according to the features extracted by the feature extraction layer.
The intelligent service layer is used for providing intelligent services for the user according to the model constructed by the scene modeling layer. For example, the intelligent service layer can provide basic application services for users, perform system intelligent optimization for electronic equipment, and provide personalized intelligent services for users.
In addition, a plurality of algorithms can be included in the panoramic perception architecture, each algorithm can be used for analyzing and processing data, and the plurality of algorithms can form an algorithm library. For example, the algorithm library may include algorithms such as a markov algorithm, a hidden dirichlet distribution algorithm, a bayesian classification algorithm, a support vector machine, a K-means clustering algorithm, a K-nearest neighbor algorithm, a conditional random field, a residual error network, a long-short term memory network, a convolutional neural network, and a cyclic neural network.
Referring to fig. 2, fig. 2 is a schematic flow chart of a data processing method according to an embodiment of the present application, where the flow chart may include:
in 101, a target task type for machine learning is determined.
With the development of artificial intelligence technology, machine learning can be performed on electronic equipment, so that a trained algorithm model is obtained. In the related art, when performing machine learning, the electronic device needs to acquire a large amount of sample data and perform machine learning according to the sample data. However, in the related art, when the electronic device acquires sample data for performing machine learning, the acquired sample data has poor adaptability to a machine learning training task. For example, when machine learning training is required, the electronic device randomly selects a part of training samples from all user sample data to perform learning training, or performs learning training by using all user sample data as training samples.
The data processing method provided by the embodiment of the present application may be applied to electronic devices such as servers or mobile terminals.
In 101 of the embodiment of the present application, for example, the electronic device may first determine the type of the machine-learned task (i.e., the target task).
In 102, inter-user similarity information corresponding to preset user sample data is obtained, and the inter-user similarity information is used for representing similarities between different users.
For example, after the type of the target task of machine learning is determined, the electronic device may obtain the inter-user similarity information corresponding to preset user sample data. Wherein, the inter-user similarity information can be used to represent the similarity between different users in the sample. For example, the similarity between the user a and the user B in the sample can be represented by a numerical value, and the magnitude of the numerical value can be used to represent the similarity between the user a and the user B. For example, the numeric value used to indicate the magnitude of the similarity may be in the range of [0,1 ]. A larger value indicates a higher degree of similarity between two users, and a smaller value indicates a lower degree of similarity between two users (i.e., a larger difference between two users).
In 103, a similarity threshold is determined according to the target task type.
For example, after determining the type of the target task to be performed by machine learning, the electronic device may determine a similarity threshold according to the type of the target task.
At 104, data of the target user is selected from preset user sample data according to the inter-user similarity information and the similarity threshold.
For example, after the similarity threshold is determined, the electronic device may select the data of the target user from the preset user sample data according to the inter-user similarity information corresponding to the preset user sample data and the similarity threshold.
At 105, machine learning of the target task is performed based on the data of the target user.
For example, after selecting the data of the target user, the electronic device may perform machine learning of the target task according to the data of the target user. That is, the electronic device may use the data of the target user as sample data for performing machine learning of the target task.
For example, the preset user sample data includes data of the user A, B, C, D, E, F, G, H, I, J, K, L. It can be understood that, in this embodiment, the preset user sample data includes the data of the 12 users as an example, and the preset user sample data may include data of a large number of users in a specific application, which is not limited in this embodiment. When the electronic equipment needs to perform machine learning, the electronic equipment determines that the type of the target task of the machine learning is a first type. And then, the electronic equipment can acquire the similarity information between the users corresponding to the preset user sample data. The inter-user similarity information may be used to indicate the similarity between different users in the sample. Then, the electronic device may determine a similarity threshold according to the first type, for example, the similarity threshold is a first threshold, and select data of the target user from preset user sample data according to the first threshold and inter-user similarity information corresponding to the preset sample data. For example, the data of the user A, B, C, D, G, H, I is selected from preset user sample data. The electronic device may then perform machine learning of the target task based on the data of the user A, B, C, D, G, H, I.
For another example, the electronic device determines that the type of the target task of machine learning that needs to be performed is the second type. And then, the electronic equipment can acquire the similarity information between the users corresponding to the preset user sample data. Then, the electronic device may determine a similarity threshold according to the second type, for example, the similarity threshold is a second threshold, and select data of the target user from preset user sample data according to the second threshold and inter-user similarity information corresponding to the preset sample data. For example, the data of the user A, B, E, F, J, K, L is selected from preset user sample data. The electronic device may then perform machine learning of the target task based on the data of the user A, B, E, F, J, K, L.
It can be understood that, in this embodiment, when the electronic device performs a task of machine learning, data of a target user may be selected from preset sample data according to the type of the task and the inter-user similarity information in the sample, and machine learning of the task may be performed according to the data of the target user. Therefore, compared with a mode of randomly selecting sample data for machine learning in the related art, the embodiment of the application can select the data for machine learning according to the task type of machine learning and the similarity information between users in the sample, so that the adaptability between the sample data for learning training acquired by the electronic device and the machine learning training task can be improved.
It should be noted that the data processing method provided by this embodiment may be applied to the data processing layer in the panoramic sensing architecture shown in fig. 1. The data processing method provided by the embodiment can enable the electronic equipment to learn the sample data acquired by the information perception layer. By using the data processing method provided by the embodiment to perform different types of model training tasks, the machine learning task can be completed more accurately and efficiently, and the model trained by learning is obtained. The learning-trained model can be applied to an intelligent service layer, and the intelligent service layer can provide intelligent service for users of electronic equipment by using the learned model, for example, information suitable for the current situation is pushed for the users.
Referring to fig. 3, fig. 3 is another schematic flow chart of a data processing method according to an embodiment of the present application, where the flow chart may include:
in 201, the electronic device acquires data of a plurality of users and determines the data as preset user sample data.
For example, the electronic device (e.g., a server) may collect data of a plurality of users, and determine preset user sample data from the data of the users.
At 202, the electronic device constructs user characterization vectors for each corresponding user based on the data for each user.
For example, after obtaining preset user sample data, the electronic device may construct a user characterization vector corresponding to each user according to data of each user included in the preset user sample data.
It should be noted that the user characterization vector may be a vector, and components included in the vector may be used to represent features possessed by the user.
In an embodiment, the above 202, the process of constructing, by the electronic device, a user characterization vector of each corresponding user according to the data of each user may include:
according to the data of each user, the electronic equipment constructs a user characterization vector corresponding to each user, wherein the user characterization vector at least comprises a first component, a second component and a third component, the first component is used for representing the degree of autonomy of the user, the second component is used for representing the degree of health of the user, and the third component is used for representing the social role label of the user.
For example, based on the user's data, the electronic device can construct a user characterization vector for the user, e.g., the user characterization vector can be represented as (x)i,yi,zi). Wherein x isiCan be used to indicate the degree of autonomy of the user, yiCan be used to indicate the health level of the user, ziSocial role labels that may be used to represent the user. For example, based on the data of user A, the electronic device can construct a user characterization vector for user A, which can be represented as (x)1,y1,z1). Wherein x is1Can be used to indicate the degree of autonomy, y, of user A1Can be used to indicate the health level of the user A, z1May be used to represent the social role label of user a. For another example, based on the data of user B, the electronic device can construct a user characterization vector of user B, which can be represented as (x)2,y2,z2). Wherein x is2Can be used to indicate the degree of autonomy, y, of user B2May be used to represent the health level of user B, z2Social role labels that may be used to represent user B, and so on.
In an embodiment, the process of acquiring data of multiple users by the electronic device 201 may include: the method comprises the steps of obtaining data of a plurality of users, wherein the data of the users at least comprise user behavior data, electronic equipment operation data and environment data.
Then, in the above step 202, the process of the electronic device constructing the user characterization vector of each corresponding user according to the data of each user may include:
according to user behavior data of each user and electronic equipment operation data, the electronic equipment determines a first component and a second component;
and determining the third component by the electronic equipment according to the user behavior data, the electronic equipment operation data and the environment data of each user.
For example, the user behavior data may include information such as the duration of the user's daily game play, the duration of office-like software usage, and the like. The electronic device operational data may include, for example, the length of time the electronic device is operating each day. Then, the electronic device may quantify the degree of autonomy of the user according to the data such as the game duration of each day, the usage duration of office software, the operation duration of each day of the electronic device, and the like. For example, the electronic device may classify the degree of autonomy of the user into three levels, i.e., high, medium, and low, according to the numerical range of the data, such as the daily game duration of the user, the daily use duration of office software, and the daily operation duration of the electronic device. Different levels may be represented quantitatively with different values. For example, a high degree of self-discipline can be represented by a numerical value of 3, a medium degree of self-discipline can be represented by a numerical value of 2, and a low degree of self-discipline can be represented by a numerical value of 1.
The user behavior data may also include data such as a schedule, a user's alarm time, and the like, and the electronic device operational data may also include data such as a night screen-break duration of the electronic device. The electronic device may then quantify the health of the user based on data such as the schedule, the user's alarm time, the electronic device's night time to screen, and so on. For example, the electronic device may classify the health degree of the user into three grades of health, general and unhealthy according to the data range of the schedule, the alarm time of the user, the night screen-off duration of the electronic device, and the like. Different levels may be represented quantitatively with different values. For example, health may be represented by a value of 3, and generally by a value of 2, while unhealthy may be represented by a value of 1.
The environmental data may include, for example, ambient light data, GPS data, and the like. The electronic device may then determine the social character tag of the user based on ambient light data, GPS data, user behavior data, electronic device operational data, and the like. For example, social character tags may include such things as work arrivals, travel arrivals, game arrivals, and the like. Different social role labels can be represented quantitatively with different numerical values. For example, a work player may be represented by a value of 3, a travel player may be represented by a value of 2, a game player may be represented by a value of 1, and so on.
After determining the first component, the second component, and the third component for constructing the user characterization vector of each corresponding user according to the data of each user, the electronic device may construct the user characterization vector according to the components. For example, for user a, the degree of autonomy is quantified as high, represented by the value 3; its health level is quantified as healthy, represented by the value 3; its social role label is determined to be a tourist, represented by the value 2. Then the user characterization vector for user a can be represented as (3,3, 2). For another example, for user B, the degree of autonomy is quantified as low, and is represented by a value of 1; the degree of health is quantified as normal, and is represented by the value 2; its social character tag is determined as the game player, indicated by the value 1. Then, the user characterization vector for user a may be represented as (1,2, 1).
In 203, according to the user characterization vectors of the users, the electronic device obtains distances between different users to obtain inter-user distance information, and determines the inter-user distance information as inter-user similarity information.
For example, after user characterization vectors of all users included in the sample are constructed, the electronic device may obtain distances between different users in the sample according to the user characterization vectors of the users, so as to obtain distance information between the users corresponding to preset user sample data, and determine the distance information between the users as similarity information between the users.
For example, a vector (x) is drawn from the user A's user profile1,y1,z1) User characterization vector (x) of user B2,y2,z2) The electronic device may obtain the distance between the user a and the user B, thereby obtaining the distance information between the users a and B, and determine the distance information between the users a and B as the similarity information between the users a and B.
In one embodiment, the inter-user distance information may be used to indicate the similarity of different users to each other in the sample. For example, the distance information between the user a and the user B in the sample may be a numerical value representing the magnitude of the distance, and the magnitude of the numerical value may be used to represent the similarity between the user a and the user B. For example, the numeric value used to indicate the magnitude of the distance may be in the range of [0,1 ]. A larger value indicates a higher degree of similarity between two users, and a smaller value indicates a lower degree of similarity between two users (i.e., a larger difference between two users).
In an embodiment, the process of obtaining, by the electronic device, the distance between different users according to the user characterization vector of each user in the process 203, and obtaining the distance information between the users may include:
according to the user characterization vectors of the users, the electronic equipment obtains the distance between different users by using an Euclidean distance formula to obtain the distance information between the users.
For example, after constructing the user characterization vectors of the users, the electronic device may use the euclidean distance formula
(Euclidean Distance) to obtain the Distance between different users, thereby obtaining the Distance information between the users.
Of course, in other embodiments, other distance calculation methods may be used to obtain the distance between different users. For example, the electronic device may further obtain the distances between the different users by using a Mahalanobis Distance (Mahalanobis Distance), a Manhattan Distance (Manhattan Distance), a Chebyshev Distance (Chebyshev Distance), a Minkowski Distance (Minkowski Distance), and a Hamming Distance (Hamming Distance) equidistance calculation method, where a numerical value of the distances is used to represent the similarity between the different users.
The above-mentioned flows 201 to 203 are flows in which the electronic device determines sample data and the inter-user similarity information corresponding to the sample data in advance. When the electronic device needs to perform machine learning subsequently, the electronic device can directly utilize the predetermined sample data and the inter-user similarity information corresponding to the sample data.
At 204, the electronic device determines a machine-learned target task type, wherein the type of the machine-learned task includes at least a first type and a second type.
For example, when an electronic device needs to perform a machine-learned task, the electronic device may first determine the type of the machine-learned task (i.e., the target task). Wherein the types of the machine-learned task may include at least a first type and a second type.
In one embodiment, the first type may be a specialization task type and the second type may be a general task type. Wherein, the specialization task type can be a type of machine learning task aiming at a certain subdivision field. For example, when it is necessary to learn the behavior habits of users who enjoy playing games, the machine learning task to be performed by the electronic device is directed to the group of users who enjoy playing games. Such a machine learning task may be of a specialized task type. Or, when the shopping behavior habit of the user who frequently shops on the internet needs to be learned, the machine learning task to be performed by the electronic device at this time is directed to the user group who frequently shops on the internet. This machine learning task may also be of a specialized task type. In contrast, a generalized task type may be a type of machine learning task for all groups, rather than a type of machine learning task for some very specific, refined group. For example, when shopping behavior habits of all users need to be learned, machine learning tasks to be performed by the electronic device at this time are directed to all user groups shopping online. Such a machine learning task may be of a generic task type.
In 205, the electronic device obtains inter-user similarity information corresponding to preset user sample data, where the inter-user similarity information is used to indicate similarities between different users.
For example, after the type of the target task of machine learning is determined, the electronic device may obtain the inter-user similarity information corresponding to preset user sample data.
At 206, the electronic device determines a similarity threshold based on the target task type.
For example, after determining the type of the target task to be performed by machine learning, the electronic device may further determine a similarity threshold according to the type of the target task. For example, in the case where the inter-user similarity information is represented by a numerical value, and the numeric value has a value range of [0,1], the similarity threshold value may be a numerical value such as 0.5 or 0.6 or 0.55. A larger value indicates a higher degree of similarity between two users, and a smaller value indicates a lower degree of similarity between two users (i.e., a larger difference between two users).
In 207, according to the similarity information between the users and the similarity threshold, the electronic device selects the data of the target user from preset user sample data, wherein if the type of the target task is the first type, the data of the target user is selected from the preset user sample data, and the similarity between every two target users is greater than or equal to the similarity threshold; and if the type of the target task is the second type, selecting data of the target user from preset user sample data, wherein the similarity between every two target users is smaller than a similarity threshold value.
For example, after the similarity threshold is determined, the electronic device may select data of the target user from preset user sample data according to inter-user similarity information corresponding to the preset user sample data and the similarity threshold. If the type of the target task determined in the process 204 is the first type, when the electronic device selects data of the target user from preset user sample data, the target user meets the following conditions: the similarity between every two target users is larger than or equal to the similarity threshold. If the type of the target task determined in the process 204 is the second type, when the electronic device selects data of the target user from preset user sample data, the target user meets the following conditions: the similarity between every two target users is smaller than a similarity threshold value.
For example, the similarity threshold is 0.5. The first type is a customized task type, and when the electronic device can select data of a target user from preset user sample data, the target user satisfies the following conditions: the similarity between every two target users is greater than or equal to the similarity threshold value of 0.5. The similarity between the target users is greater than or equal to the similarity threshold value of 0.5, which indicates that the similarity between the target users is high. It can be understood that, when the type of the target task learned by the machine is the customized task type, the data of the user with high similarity needs to be selected, so that the finally learned model can have good specificity. Therefore, in the present embodiment, the similarity between the target users may be set to be greater than or equal to the similarity threshold value 0.5.
As another example, the similarity threshold is 0.5. The second type is a universal task type, and when the electronic device can select data of a target user from preset user sample data, the target user satisfies the following conditions: the similarity between every two target users is less than the similarity threshold value of 0.5. The similarity between every two target users is less than the similarity threshold value of 0.5, which means that the similarity between every two target users is low, and the difference is high. It can be understood that, when the type of the target task learned by the machine is a general task type, the data of the user with low similarity and high diversity needs to be selected to ensure that the coverage of the sample data for machine learning is as wide as possible and diversified as possible, so that the finally learned model has good generalization capability. Therefore, in the present embodiment, the similarity between the target users may be set to be less than the similarity threshold value 0.5.
At 208, the electronic device performs machine learning of the target task based on the target user's data.
For example, after selecting the data of the target user, the electronic device may perform machine learning of the target task according to the data of the target user. That is, the electronic device may use the data of the target user as sample data for performing machine learning of the target task.
In another embodiment, the similarity threshold may be a preset fixed value. For example, the electronic device may preset a similarity threshold value to be 0.5 or 0.6, and the like, which is not limited by the examples herein.
Referring to fig. 4 to 5, fig. 4 to 5 are schematic views of a data processing method according to an embodiment of the present disclosure.
For example, as shown in fig. 4, the mobile terminal transmits the collected user data to the server. The server may obtain data of a plurality of users, and determine the user data as preset user sample data.
And then, the server can construct the user characterization vectors of the corresponding users according to the data of the users contained in the preset user sample data. For example, the user characterization vector may include a first component for representing the user's degree of autonomy, a second component for representing the user's degree of health, and a third component for representing the user's social role label. The user's degree of autonomy is divided into three levels of high, medium and low. Different levels may be represented quantitatively with different values. For example, a high degree of self-discipline can be represented by a numerical value of 3, a medium degree of self-discipline can be represented by a numerical value of 2, and a low degree of self-discipline can be represented by a numerical value of 1. The health degree of the user is divided into three grades of health, general health and unhealthy health. Different levels may be represented quantitatively with different values. For example, health may be represented by a value of 3, and generally by a value of 2, while unhealthy may be represented by a value of 1. Social character tags may include, for example, job arrivals, travel arrivals, game arrivals, and the like. Different social role labels can be represented quantitatively with different numerical values. For example, a work player may be represented by a value of 3, a travel player may be represented by a value of 2, a game player may be represented by a value of 1, and so on.
For example, for user a, the degree of autonomy is quantified as high, represented by the value 3; its health level is quantified as healthy, represented by the value 3; its social role label is determined to be a tourist, represented by the value 2. Then the user characterization vector for user a can be represented as (3,3, 2).
After user characterization vectors of all users included in the sample are constructed, the electronic device can obtain distances between different users in the sample according to the user characterization vectors of all the users, so that distance information between the users corresponding to preset user sample data is obtained. The inter-user distance information may be a numerical value indicating the magnitude of the distance, and the magnitude of the numerical value may be used to indicate the degree of similarity between two users. For example, the numeric value used to indicate the magnitude of the distance may be in the range of [0,1 ]. A larger value indicates a higher degree of similarity between two users, and a smaller value indicates a lower degree of similarity between two users.
In one embodiment, after obtaining the inter-user distance information, the electronic device may form a distance matrix from the user to the user accordingly. For example, taking user A, B, C, D as an example, user A's user characterizes a vector of (x)1,y1,z1) The user characterization vector of user B is (x)2,y2,z2) The user characterization vector of user C is (x)3,y3,z3) The user characterization vector of user D is (x)4,y4,z4). The electronic device is based on the vector (x)1,y1,z1) And (x)2,y2,z2) Using the euclidean distance formula, the distance between the user a and the user B is calculated to be 0.6, which may indicate that the similarity between the users a and B is 0.6. Similarly, for example, the electronic device calculates that the distance between the users a and C is 0.5, the distance between the users a and D is 0.3, the distance between the users B and C is 0.1, the distance between the users B and D is 0.4, and the distance between the users C and D is 0.9. Then, the electronic device may construct a user-to-user distance matrix as shown in table 1 below. It is understood that the distance matrix between the users is constructed by taking the user A, B, C, D as an example in the present embodiment. When the specific application is implemented, presetThe user sample data may include data of multiple users, and the constructed distance matrix between users is larger than that in table 1.
TABLE 1 distance matrix between users
| A | B | C | D | |
| A | / | / | / | / |
| B | 0.6 | / | / | / |
| C | 0.5 | 0.1 | / | / |
| D | 0.3 | 0.4 | 0.9 | / |
After obtaining the distance matrix between the users included in the preset user sample data, when the server performs machine learning, the server may first determine the type of the target task of machine learning. For example, the target task is to learn the behavior habit of a user who likes playing a game, that is, the server determines that the type of the target task learned by the machine is a specialized task type.
And then, the server acquires a distance matrix between the users contained in the preset user sample data constructed before, and determines a distance threshold according to the type of the target task. For example, the distance threshold is 0.5. After the distance threshold is determined, the server may select data of the target user from preset user sample data according to the distance matrix and the distance threshold. For example, the server may determine a base user from the sample and then determine users that are greater than or equal to a distance threshold from the base user as target users. For example, the server determines the user a as the base user, and then the server may determine the users having a distance greater than or equal to 0.5 from the user a as the associated users, and acquire these associated users and base user determined as the target users, and acquire data of the target users. For example, if the base user is a, then users B and C would be determined to be associated users, and the server may then determine user A, B, C as the target user and obtain user data for user A, B, C as sample data for machine learning.
The server may then perform machine learning of the target task based on the data of the user A, B, C.
After machine learning, the server may obtain a learning trained algorithm model. The server may transmit the algorithm model to the mobile terminal so that the mobile terminal may predict behavioral habits of a user who likes to play the game according to the algorithm model.
Referring to fig. 6, fig. 6 is a schematic structural diagram of a data processing apparatus according to an embodiment of the present disclosure. The data processing apparatus 300 may include: a first determining module 301, an obtaining module 302, a second determining module 303, a selecting module 304, and a learning module 305.
A first determining module 301, configured to determine a target task type for machine learning.
An obtaining module 302, configured to obtain inter-user similarity information corresponding to preset user sample data, where the inter-user similarity information is used to indicate similarities between different users.
A second determining module 303, configured to determine a similarity threshold according to the target task type.
And the selecting module 304 is configured to select data of the target user from preset user sample data according to the inter-user similarity information and the similarity threshold.
A learning module 305, configured to perform machine learning of the target task according to the data of the target user.
In one embodiment, the first determining module 301 may be further configured to:
acquiring data of a plurality of users and determining the data as preset user sample data;
according to the data of each user, constructing a user characterization vector of each corresponding user;
and obtaining the distance between different users according to the user depicting vectors of the users to obtain the distance information between the users, and determining the distance information between the users as the similarity information between the users.
In one embodiment, the first determining module 301 may be configured to:
according to the data of each user, a user characterization vector of each corresponding user is constructed, wherein the user characterization vector at least comprises a first component, a second component and a third component, the first component is used for representing the degree of autonomy of the user, the second component is used for representing the degree of health of the user, and the third component is used for representing the social role label of the user.
In one embodiment, the first determining module 301 may be configured to:
acquiring data of a plurality of users, wherein the data of the users at least comprises user behavior data, electronic equipment operation data and environment data;
determining a first component and a second component according to the user behavior data of each user and the electronic equipment operation data;
and determining a third component according to the user behavior data, the electronic equipment operation data and the environment data of each user.
In one embodiment, the first determining module 301 may be configured to:
and obtaining the distance between different users by using an Euclidean distance formula according to the user characterization vectors of the users to obtain the distance information between the users.
In one embodiment, the machine-learned task types include at least a first type and a second type.
The selecting module 304 may be configured to:
if the type of the target task learned by the machine is the first type, selecting data of a target user from preset user sample data according to the similarity information between the users and the similarity threshold, wherein the similarity between every two target users is greater than or equal to the similarity threshold;
and if the type of the target task learned by the machine is the second type, selecting data of a target user from preset user sample data according to the similarity information between the users and the similarity threshold, wherein the similarity between every two target users is smaller than the similarity threshold.
The present embodiment provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed on a computer, the computer is caused to execute the flow in the data processing method provided in this embodiment.
The embodiment of the present application further provides an electronic device, which includes a memory and a processor, where the processor is configured to execute the flow in the data processing method provided in this embodiment by calling the computer program stored in the memory.
For example, the electronic device may be a mobile terminal such as a tablet computer or a smart phone, or may be a server. Referring to fig. 7, fig. 7 is a schematic structural diagram of an electronic device according to an embodiment of the present disclosure.
The electronic device 400 may include components such as a memory 401, a processor 402, and the like. Those skilled in the art will appreciate that the electronic device configuration shown in fig. 7 does not constitute a limitation of the electronic device and may include more or fewer components than shown, or some components may be combined, or a different arrangement of components.
The memory 401 may be used to store applications and data. The memory 401 stores applications containing executable code. The application programs may constitute various functional modules. The processor 402 executes various functional applications and data processing by running an application program stored in the memory 401.
The processor 402 is a control center of the electronic device, connects various parts of the entire electronic device using various interfaces and lines, and performs various functions of the electronic device and processes data by running or executing an application program stored in the memory 401 and calling data stored in the memory 401, thereby integrally monitoring the electronic device.
In this embodiment, the processor 402 in the electronic device loads the executable code corresponding to the processes of one or more application programs into the memory 401 according to the following instructions, and the processor 402 runs the application programs stored in the memory 401, so as to execute:
determining a target task type of machine learning;
acquiring similarity information between users corresponding to preset user sample data, wherein the similarity information between the users is used for representing the similarity between different users;
determining a similarity threshold according to the target task type;
selecting data of a target user from preset user sample data according to the inter-user similarity information and the similarity threshold;
and performing machine learning of the target task according to the data of the target user.
Referring to fig. 8, an electronic device 500 may include a sensor 501, a memory 502, a processor 503, a speaker 504, a microphone 505, a display 506, and the like.
The sensor 501 may include a gyro sensor (e.g., a three-axis gyro sensor), an acceleration sensor, and the like.
The memory 502 may be used to store applications and data. Memory 502 stores applications containing executable code. The application programs may constitute various functional modules. The processor 503 executes various functional applications and data processing by running an application program stored in the memory 502.
The processor 503 is a control center of the electronic device, connects various parts of the whole electronic device by using various interfaces and lines, and performs various functions of the electronic device and processes data by running or executing an application program stored in the memory 502 and calling the data stored in the memory 502, thereby performing overall monitoring of the electronic device.
The speaker 504 may be used to play sound signals. The microphone 505 may be used to pick up sound signals in the environment. The display 506 may be used to display teletext information.
In this embodiment, the processor 503 in the electronic device loads the executable code corresponding to the processes of one or more application programs into the memory 502 according to the following instructions, and the processor 503 runs the application programs stored in the memory 502, so as to execute:
determining a target task type of machine learning;
acquiring similarity information between users corresponding to preset user sample data, wherein the similarity information between the users is used for representing the similarity between different users;
determining a similarity threshold according to the target task type;
selecting data of a target user from preset user sample data according to the inter-user similarity information and the similarity threshold;
and performing machine learning of the target task according to the data of the target user.
In one embodiment, before the determining the target task type for machine learning, the processor 503 may further perform: acquiring data of a plurality of users and determining the data as preset user sample data; according to the data of each user, constructing a user characterization vector of each corresponding user; and obtaining the distance between different users according to the user depicting vectors of the users to obtain the distance information between the users, and determining the distance information between the users as the similarity information between the users.
In one embodiment, when the processor 503 executes the process of constructing the user characterization vector of each corresponding user according to the data of each user, it may execute: according to the data of each user, a user characterization vector of each corresponding user is constructed, wherein the user characterization vector at least comprises a first component, a second component and a third component, the first component is used for representing the degree of autonomy of the user, the second component is used for representing the degree of health of the user, and the third component is used for representing the social role label of the user.
In one embodiment, the processor 503, when executing the acquiring of the data of the plurality of users, may execute: the method comprises the steps of obtaining data of a plurality of users, wherein the data of the users at least comprise user behavior data, electronic equipment operation data and environment data.
Then, when the processor 503 executes the user characterization vector for each corresponding user according to the data of each user, it may execute: determining a first component and a second component according to the user behavior data of each user and the electronic equipment operation data; and determining a third component according to the user behavior data, the electronic equipment operation data and the environment data of each user.
In an embodiment, when the processor 503 executes the above-mentioned obtaining the distance between different users according to the user characterization vector of each user, and obtains the distance information between users, it may execute: and obtaining the distance between different users by using an Euclidean distance formula according to the user characterization vectors of the users to obtain the distance information between the users.
In one embodiment, the machine-learned task types include at least a first type and a second type. Then, when the processor 503 executes the process of selecting the data of the target user from the preset user sample data according to the inter-user similarity information and the similarity threshold, the process may execute: if the type of the target task learned by the machine is the first type, selecting data of a target user from preset user sample data according to the similarity information between the users and the similarity threshold, wherein the similarity between every two target users is greater than or equal to the similarity threshold; and if the type of the target task learned by the machine is the second type, selecting data of a target user from preset user sample data according to the similarity information between the users and the similarity threshold, wherein the similarity between every two target users is smaller than the similarity threshold.
In the above embodiments, the descriptions of the embodiments have respective emphasis, and parts that are not described in detail in a certain embodiment may refer to the above detailed description of the data processing method, and are not described herein again.
The data processing apparatus provided in the embodiment of the present application and the data processing method in the above embodiment belong to the same concept, and any method provided in the embodiment of the data processing method may be run on the data processing apparatus, and a specific implementation process thereof is described in the embodiment of the data processing method in detail, and is not described herein again.
It should be noted that, for the data processing method described in the embodiment of the present application, it can be understood by those skilled in the art that all or part of the process of implementing the data processing method described in the embodiment of the present application can be completed by controlling the relevant hardware through a computer program, where the computer program can be stored in a computer-readable storage medium, such as a memory, and executed by at least one processor, and during the execution, the process of the embodiment of the data processing method can be included. The storage medium may be a magnetic disk, an optical disk, a Read Only Memory (ROM), a Random Access Memory (RAM), or the like.
In the data processing apparatus according to the embodiment of the present application, each functional module may be integrated into one processing chip, or each module may exist alone physically, or two or more modules are integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode. The integrated module, if implemented in the form of a software functional module and sold or used as a stand-alone product, may also be stored in a computer readable storage medium, such as a read-only memory, a magnetic or optical disk, or the like.
The foregoing detailed description has provided a data processing method, an apparatus, a storage medium, and an electronic device according to embodiments of the present application, and specific examples are applied herein to explain the principles and implementations of the present application, and the descriptions of the foregoing embodiments are only used to help understand the method and the core ideas of the present application; meanwhile, for those skilled in the art, according to the idea of the present application, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present application.
Claims (10)
1. A data processing method, comprising:
determining a target task type of machine learning;
acquiring similarity information between users corresponding to preset user sample data, wherein the similarity information between the users is used for representing the similarity between different users;
determining a similarity threshold according to the target task type;
selecting data of a target user from preset user sample data according to the inter-user similarity information and the similarity threshold;
and performing machine learning of the target task according to the data of the target user.
2. The data processing method of claim 1, further comprising, prior to the determining a target task type for machine learning:
acquiring data of a plurality of users and determining the data as preset user sample data;
according to the data of each user, constructing a user characterization vector of each corresponding user;
and obtaining the distance between different users according to the user depicting vectors of the users to obtain the distance information between the users, and determining the distance information between the users as the similarity information between the users.
3. The data processing method according to claim 2, wherein the constructing a user characterization vector for each corresponding user according to the data of each user comprises:
according to the data of each user, a user characterization vector of each corresponding user is constructed, wherein the user characterization vector at least comprises a first component, a second component and a third component, the first component is used for representing the degree of autonomy of the user, the second component is used for representing the degree of health of the user, and the third component is used for representing the social role label of the user.
4. The data processing method of claim 3, wherein the obtaining data of a plurality of users comprises: acquiring data of a plurality of users, wherein the data of the users at least comprises user behavior data, electronic equipment operation data and environment data;
the constructing of the user characterization vector of each corresponding user according to the data of each user includes:
determining a first component and a second component according to the user behavior data of each user and the electronic equipment operation data;
and determining a third component according to the user behavior data, the electronic equipment operation data and the environment data of each user.
5. The data processing method according to claim 2, wherein the obtaining of the distance between different users according to the user characterization vectors of the users to obtain the inter-user distance information comprises:
and obtaining the distance between different users by using an Euclidean distance formula according to the user characterization vectors of the users to obtain the distance information between the users.
6. The data processing method of claim 1, wherein the task types of machine learning include at least a first type and a second type;
the selecting data of the target user from preset user sample data according to the inter-user similarity information and the similarity threshold value comprises:
if the type of the target task learned by the machine is the first type, selecting data of a target user from preset user sample data according to the similarity information between the users and the similarity threshold, wherein the similarity between every two target users is greater than or equal to the similarity threshold;
and if the type of the target task learned by the machine is the second type, selecting data of a target user from preset user sample data according to the similarity information between the users and the similarity threshold, wherein the similarity between every two target users is smaller than the similarity threshold.
7. A data processing apparatus, comprising:
the first determination module is used for determining a target task type of machine learning;
the device comprises an acquisition module, a processing module and a display module, wherein the acquisition module is used for acquiring similarity information between users corresponding to preset user sample data, and the similarity information between the users is used for representing the similarity between different users;
the second determining module is used for determining a similarity threshold according to the target task type;
the selecting module is used for selecting the data of the target user from preset user sample data according to the similarity information among the users and the similarity threshold;
and the learning module is used for performing machine learning of the target task according to the data of the target user.
8. The data processing apparatus of claim 7, wherein the first determining module is further configured to:
acquiring data of a plurality of users and determining the data as preset user sample data;
according to the data of each user, constructing a user characterization vector of each corresponding user;
and obtaining the distance between different users according to the user depicting vectors of the users to obtain the distance information between the users, and determining the distance information between the users as the similarity information between the users.
9. A storage medium having stored thereon a computer program, characterized in that the computer program, when executed on a computer, causes the computer to execute the method according to any of claims 1 to 6.
10. An electronic device comprising a memory, a processor, wherein the processor is configured to perform the method of any of claims 1 to 6 by invoking a computer program stored in the memory.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910282467.2A CN111797878A (en) | 2019-04-09 | 2019-04-09 | Data processing method, data processing device, storage medium and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910282467.2A CN111797878A (en) | 2019-04-09 | 2019-04-09 | Data processing method, data processing device, storage medium and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111797878A true CN111797878A (en) | 2020-10-20 |
Family
ID=72805336
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910282467.2A Pending CN111797878A (en) | 2019-04-09 | 2019-04-09 | Data processing method, data processing device, storage medium and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111797878A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112232890A (en) * | 2020-11-11 | 2021-01-15 | 腾讯科技（深圳）有限公司 | Data processing method, device, equipment and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107609461A (en) * | 2017-07-19 | 2018-01-19 | 阿里巴巴集团控股有限公司 | The training method of model, the determination method, apparatus of data similarity and equipment |
| WO2018018610A1 (en) * | 2016-07-29 | 2018-02-01 | 深圳越界创新科技有限公司 | Method and device for computing degree of match, and user equipment |
| CN107730271A (en) * | 2017-09-20 | 2018-02-23 | 北京奇艺世纪科技有限公司 | Similar users based on virtual interacting object determine method, apparatus and electronic equipment |
-
2019
- 2019-04-09 CN CN201910282467.2A patent/CN111797878A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018018610A1 (en) * | 2016-07-29 | 2018-02-01 | 深圳越界创新科技有限公司 | Method and device for computing degree of match, and user equipment |
| CN107609461A (en) * | 2017-07-19 | 2018-01-19 | 阿里巴巴集团控股有限公司 | The training method of model, the determination method, apparatus of data similarity and equipment |
| CN107730271A (en) * | 2017-09-20 | 2018-02-23 | 北京奇艺世纪科技有限公司 | Similar users based on virtual interacting object determine method, apparatus and electronic equipment |
Non-Patent Citations (1)
| Title |
|---|
| 缪峰;贾华丁;熊于宁;: "基于服务相似度的移动用户近似邻居选取方法", 计算机工程, no. 05 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112232890A (en) * | 2020-11-11 | 2021-01-15 | 腾讯科技（深圳）有限公司 | Data processing method, device, equipment and storage medium |
| CN112232890B (en) * | 2020-11-11 | 2023-07-04 | 腾讯科技（深圳）有限公司 | Data processing method, device, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11908176B2 (en) | Data recognition model construction apparatus and method for constructing data recognition model thereof, and data recognition apparatus and method for recognizing data thereof | |
| CN111797858A (en) | Model training method, behavior prediction method, device, storage medium and equipment | |
| CN111798018A (en) | Behavior prediction method, behavior prediction device, storage medium and electronic equipment | |
| CN111797854B (en) | Scene model building method and device, storage medium and electronic equipment | |
| CN111798260A (en) | User behavior prediction model construction method and device, storage medium and electronic equipment | |
| CN111797861A (en) | Information processing method, information processing apparatus, storage medium, and electronic device | |
| CN111797288A (en) | Data screening method and device, storage medium and electronic equipment | |
| CN111798259A (en) | Application recommendation method and device, storage medium and electronic equipment | |
| CN111798978A (en) | User health assessment method and device, storage medium and electronic equipment | |
| CN111797870A (en) | Optimization method and device of algorithm model, storage medium and electronic equipment | |
| CN111797878A (en) | Data processing method, data processing device, storage medium and electronic equipment | |
| CN111797874B (en) | Behavior prediction method and device, storage medium and electronic equipment | |
| CN111797856A (en) | Modeling method, modeling device, storage medium and electronic equipment | |
| CN111797867A (en) | System resource optimization method and device, storage medium and electronic equipment | |
| CN111797986A (en) | Data processing method, data processing device, storage medium and electronic equipment | |
| CN111814812A (en) | Modeling method, modeling device, storage medium, electronic device and scene recognition method | |
| CN111797862A (en) | Task processing method and device, storage medium and electronic equipment | |
| CN111797849A (en) | User activity identification method and device, storage medium and electronic equipment | |
| CN111796663B (en) | Scene recognition model updating method and device, storage medium and electronic equipment | |
| CN111797289A (en) | Model processing method and device, storage medium and electronic equipment | |
| CN111797656B (en) | Face key point detection method and device, storage medium and electronic equipment | |
| CN111796916A (en) | Data distribution method, device, storage medium and server | |
| CN111797299A (en) | Model training method, webpage classification method, device, storage medium and equipment | |
| CN111797290A (en) | Data processing method, data processing device, storage medium and electronic equipment | |
| CN111796992B (en) | Behavior preference determination method and apparatus, storage medium and electronic device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |