Method and device for predicting user information of electric power company, electronic equipment and storage medium
Technical Field
The application relates to the technical field of electronic commerce, in particular to an electronic commerce user information prediction method, an electronic commerce user information prediction device, electronic equipment and a storage medium.
Background
In the current information age, online shopping has become popular, and with the increasing popularity of online shopping, the number of e-commerce platforms is increasing. At present, part of e-commerce platforms can predict user information of e-commerce users through big data, and further carry out personalized recommendation on the e-commerce users, so that consumption of the e-commerce users is improved and stimulated.
Most of the existing e-commerce platforms predict the historical consumption information of e-commerce users and mainly use a machine learning model for prediction, however, in the method, under the complex environment of large number of users and massive data, the user information is difficult to predict more accurately, the robustness is poor, and the recommendation conversion rate of the e-commerce platform is not ideal.
Disclosure of Invention
The embodiment of the application aims to provide an e-commerce user information prediction method, an e-commerce user information prediction device, an electronic device and a storage medium, which can predict user information more accurately and have better robustness, so that the recommendation conversion rate of an e-commerce platform becomes ideal.
In a first aspect, an embodiment of the present application provides a method for predicting e-commerce user information, including:
acquiring click data of contents clicked by an E-commerce user in preset time and attribute data of the clicked contents;
forming a click time sequence data set of the E-commerce user according to the click data and the attribute data;
preprocessing the data of the click time sequence data set to obtain corresponding aggregation coding features, target coding features and word vector coding features;
inputting the aggregation coding features, the target coding features and the word vector coding features to a preset deep learning fusion model to obtain a corresponding prediction result;
and predicting to obtain the user information of the electric power business user according to the prediction result.
In the implementation process, the e-commerce user information prediction method of the embodiment of the application forms the click time sequence data set of the e-commerce user through click data of contents clicked by the e-commerce user and attribute data of the click contents within a preset time, performs data preprocessing on the click time sequence data set to obtain corresponding aggregation coding features, target coding features and word vector coding features to perform user information prediction on the e-commerce user, not only considers historical consumption information of the e-commerce user, but also considers click browsing information of the e-commerce user, can contribute to prediction on user information of the e-commerce user, and also performs prediction through a preset deep learning fusion model, and the preset deep learning fusion model is suitable for operation under a complex environment with huge user number and massive data, and better predicts user information of the e-commerce user, and then user information can be predicted more accurately, and the robustness is also better, thereby the recommendation conversion rate of the e-commerce platform becomes ideal.
Further, the forming a click timing sequence data set of the e-commerce user according to the click data and the attribute data includes:
correspondingly linking the attribute data with the click data;
and aggregating the linked data by the E-commerce users, and sequencing the data by time to form a click time sequence data set of the E-commerce users.
In the implementation process, the method can better obtain the click time sequence data set of the e-commerce user, so that the prediction of the user information of the e-commerce user is facilitated.
Further, the preprocessing the data of the click timing sequence data set to obtain corresponding aggregation coding features, target coding features and word vector coding features includes:
carrying out data cleaning and noise reduction on the click time sequence data set to obtain a target click time sequence data set;
and performing data characteristic processing on the target click time sequence data set to obtain corresponding aggregation coding characteristics, target coding characteristics and word vector coding characteristics.
In the implementation process, the method carries out data cleaning and noise reduction on the click time sequence data set, can effectively reduce data interference, and can more accurately obtain corresponding aggregation coding features, target coding features and word vector coding features, thereby more accurately predicting user information.
Further, the preset deep learning fusion model comprises a preset Transform model, a preset LightGBM model and a preset Bi-LSTM + Attention model;
the step of inputting the aggregate coding feature, the target coding feature and the word vector coding feature to a preset deep learning fusion model to obtain a corresponding prediction result includes:
and correspondingly inputting the aggregation coding features, the target coding features and the word vector coding features into a preset Transform model, a preset LightGBM model and a preset Bi-LSTM + Attention model to obtain a corresponding prediction result.
In the implementation process, the method predicts the corresponding prediction result more accurately and further more accurately by one-to-one correspondence between the preset Transform model and the aggregation coding feature, between the preset LightGBM model and the target coding feature and between the preset Bi-LSTM + Attention model and the word vector coding feature.
Further, the predicting user information of the e-commerce user according to the prediction result includes:
and predicting to obtain the user information of the electric commercial user according to the prediction result and a preset weighting coefficient.
In the implementation process, the method combines the preset weighting coefficient, and can accurately predict the user information of the electric commercial user.
Further, the preprocessing the data of the click timing sequence data set to obtain corresponding word vector coding features includes:
and preprocessing the data of the click time sequence data set through a preset word2vec model to obtain corresponding word vector coding characteristics.
In the implementation process, the method carries out data preprocessing on the click time sequence data set through a preset word2vec model to obtain corresponding word vector coding features, can learn the interrelation and the internal rule among click data, and well obtain the corresponding word vector coding features, so that the prediction of user information of electric business users is facilitated.
In a second aspect, an embodiment of the present application provides an e-commerce user information prediction apparatus, including:
the system comprises an acquisition module, a display module and a display module, wherein the acquisition module is used for acquiring click data of contents clicked by an E-commerce user within preset time and attribute data of the click contents;
the generation module is used for forming a click time sequence data set of the E-commerce user according to the click data and the attribute data;
the data processing module is used for preprocessing the data of the click time sequence data set to obtain corresponding aggregation coding features, target coding features and word vector coding features;
the model prediction module is used for inputting the aggregation coding features, the target coding features and the word vector coding features to a preset deep learning fusion model to obtain a corresponding prediction result;
and the information prediction module is used for predicting the user information of the electric power business user according to the prediction result.
In the implementation process, the e-commerce user information prediction device of the embodiment of the application forms the click time sequence data set of the e-commerce user through click data of contents clicked by the e-commerce user and attribute data of the click contents within a preset time, performs data preprocessing on the click time sequence data set to obtain corresponding aggregation coding features, target coding features and word vector coding features to perform user information prediction on the e-commerce user, not only considers historical consumption information of the e-commerce user, but also considers click browsing information of the e-commerce user, can contribute to prediction on user information of the e-commerce user, and performs prediction through a preset deep learning fusion model, wherein the preset deep learning fusion model is suitable for operation under a complex environment with huge user number and massive data, and better predicts user information of the e-commerce user, and then user information can be predicted more accurately, and the robustness is also better, thereby the recommendation conversion rate of the e-commerce platform becomes ideal.
Further, the preset deep learning fusion model comprises a preset Transform model, a preset LightGBM model and a preset Bi-LSTM + Attention model;
the model prediction module is specifically configured to:
and correspondingly inputting the aggregation coding features, the target coding features and the word vector coding features into a preset Transform model, a preset LightGBM model and a preset Bi-LSTM + Attention model to obtain a corresponding prediction result.
In the implementation process, the device predicts the corresponding prediction result more accurately and further more accurately by one-to-one correspondence between the preset Transform model and the aggregation coding feature, between the preset LightGBM model and the target coding feature and between the preset Bi-LSTM + Attention model and the word vector coding feature.
In a third aspect, an embodiment of the present application provides an electronic device, which includes a memory and a processor, where the memory is used to store a computer program, and the processor runs the computer program to make the electronic device execute the foregoing method for predicting the e-commerce user information.
In a fourth aspect, the present application provides a computer-readable storage medium, which stores a computer program, and when the computer program is executed by a processor, the computer program implements the above-mentioned method for predicting the e-commerce user information.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are required to be used in the embodiments of the present application will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and that those skilled in the art can also obtain other related drawings based on the drawings without inventive efforts.
Fig. 1 is a schematic flowchart of a method for predicting e-commerce user information according to an embodiment of the present disclosure;
fig. 2 is a schematic flowchart of step S120 according to a first embodiment of the present application;
fig. 3 is a schematic flowchart of step S130 according to a first embodiment of the present application;
fig. 4 is a schematic diagram of a model structure of a Transform model according to an embodiment of the present application;
FIG. 5 is a schematic diagram of a Bi-LSTM + Attention model according to an embodiment of the present disclosure;
fig. 6 is a block diagram illustrating a configuration of an electric power consumer information prediction apparatus according to a second embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be described below with reference to the drawings in the embodiments of the present application.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures. Meanwhile, in the description of the present application, the terms "first", "second", and the like are used only for distinguishing the description, and are not to be construed as indicating or implying relative importance.
Most of the existing e-commerce platforms predict the historical consumption information of e-commerce users and mainly use a machine learning model for prediction, however, in the method, under the complex environment of large number of users and massive data, the user information is difficult to predict more accurately, the robustness is poor, and the recommendation conversion rate of the e-commerce platform is not ideal.
In view of the above problems in the prior art, the present application provides a method and an apparatus for predicting e-commerce user information, an electronic device, and a storage medium, which can predict user information more accurately and have better robustness, so that the recommended conversion rate of an e-commerce platform becomes ideal.
Example one
Referring to fig. 1, fig. 1 is a schematic flowchart of an e-commerce user information prediction method provided in an embodiment of the present application. The following method for predicting the information of the e-commerce user in the embodiment of the application can be applied to the server.
The method for predicting the information of the e-commerce user comprises the following steps:
step S110, acquiring click data of contents clicked by the E-commerce user in a preset time and attribute data of the clicked contents.
In the present embodiment, the predetermined time may be the last month, the last three months, or the last half year, or the like.
The electric utility customers are usually a plurality of electric utility customers, and the number of the electric utility customers is large, of course, the electric utility customers may be a single electric utility customer, and in the present embodiment, the electric utility customers are mainly used as a plurality of electric utility customers for the relevant illustration.
The contents clicked by the E-commerce user can be advertisements clicked by the E-commerce user and/or commodities clicked by the E-commerce user.
The click data of the contents clicked by the e-commerce user may include a click time of the contents clicked by the e-commerce user, a number of clicks of the contents clicked by the e-commerce user, and the like.
The attribute data of the clicked-through content may include an industry category, a product category, a manufacturer category, an advertiser category, and the like to which the clicked-through content belongs.
And step S120, forming a click time sequence data set of the E-commerce user according to the click data and the attribute data.
In this embodiment, the click time series data set of the e-commerce user, that is, the click time series data set of the e-commerce user, is sorted by the click time of the e-commerce user.
And the click time sequence data set of each e-commerce user is formed in a one-to-one correspondence mode according to the click data and the attribute data of each e-commerce user.
Step S130, preprocessing the data of the click timing sequence data set to obtain corresponding aggregate coding features, target coding features and word vector coding features.
In this embodiment, data preprocessing is performed on the click timing sequence data set of each e-commerce user to obtain an aggregation coding feature, a target coding feature and a word vector coding feature corresponding to each e-commerce user.
The aggregate encoding features may include, among other things, a maximum value, a minimum value, a standard deviation, etc. in the click timing data set of the e-commerce user.
The target coding feature is a structural feature that is supervised based on target variables (user information to be predicted).
The word vector coding feature may be obtained by coding an ID of each clicked content as a word, and an ID list of clicked contents within a predetermined time by a single tv provider user as a sentence.
Step S140, inputting the aggregate coding feature, the target coding feature and the word vector coding feature to a preset deep learning fusion model to obtain a corresponding prediction result.
In this embodiment, the aggregation coding feature, the target coding feature and the word vector coding feature of each e-commerce user are input to a preset deep learning fusion model, so as to obtain a prediction result corresponding to each e-commerce user.
The preset deep learning fusion model comprises a plurality of preset deep learning models, and the plurality of preset deep learning models are pre-trained deep learning models.
Optionally, the preset deep learning fusion model may include three preset deep learning models, where the three preset deep learning models correspond to the aggregate coding feature, the target coding feature, and the word vector coding feature one to one, that is, the aggregate coding feature is input to the first preset deep learning model, the target coding feature is input to the second preset deep learning model, and the word vector coding feature is input to the third preset deep learning model. Under the condition that the preset deep learning fusion model can comprise three preset deep learning models, three obtained prediction results corresponding to each e-commerce user correspond to the aggregation coding feature, the target coding feature and the word vector coding feature of each e-commerce user one by one.
The prediction result corresponds to the user information of the e-commerce user to be predicted, for example, the user information of the e-commerce user to be predicted is the age of the e-commerce user, and then the prediction result is the age.
In this embodiment, the deep learning model can be relatively suitable for operation in a complex environment with a large number of users and massive data, so that user information of e-commerce users can be better predicted, the deep learning fusion model integrates a plurality of deep learning models, and under the condition that the prediction accuracy of a single deep learning model is higher, the prediction accuracy of the deep learning fusion model is higher, and the robustness is better.
And S150, predicting to obtain the user information of the E-commerce user according to the prediction result.
In this embodiment, the user information of each electric utility user is obtained by prediction according to the prediction result corresponding to each electric utility user.
It can be understood that the user information of each e-commerce user is calculated according to the corresponding prediction result of each e-commerce user.
In this embodiment, the electric utility customer information prediction method according to the embodiment of the present application may be used to predict the age, sex, or hobby of the electric utility customer.
The E-commerce user information prediction method of the embodiment of the application forms the click time sequence data set of the E-commerce user through click data of contents clicked by the E-commerce user and attribute data of the click contents within preset time, carries out data preprocessing on the click time sequence data set to obtain corresponding aggregation coding features, target coding features and word vector coding features to carry out user information prediction on the E-commerce user, not only considers historical consumption information of the E-commerce user, but also considers click browsing information of the E-commerce user, can be conductive to prediction on user information of the E-commerce user, and carries out prediction through a preset deep learning fusion model which can be suitable for operation under complex environments with huge user quantity and massive data, better predicts the user information of the E-commerce user, and can further predict the user information more accurately, and the robustness is also better, so that the recommended conversion rate of the e-commerce platform becomes ideal.
Referring to fig. 2, fig. 2 is a schematic flowchart of step S120 provided in the embodiment of the present application.
As an optional implementation manner, the method for predicting the e-commerce user information in the embodiment of the present application, in step S120, forming a click timing data set of the e-commerce user according to the click data and the attribute data, may include the following steps:
step S121, correspondingly linking the attribute data and the click data;
and S122, aggregating the linked data by the E-commerce users, and sequencing the data by time to form a click time sequence data set of the E-commerce users.
The attribute data of each E-commerce user is linked with the click data, and the attribute data is correspondingly linked with the click data.
In the process, the click time sequence data set of the e-commerce user can be well obtained by the method, so that the prediction of the user information of the e-commerce user is facilitated.
In order to obtain corresponding aggregate coding features, target coding features, and word vector coding features more accurately, a possible implementation manner is provided in the embodiment of the present application, referring to fig. 3, fig. 3 is a schematic flowchart of step S130 provided in the embodiment of the present application, a method for predicting e-commerce user information in the embodiment of the present application, in which step S130, data is preprocessed on the click timing data set to obtain corresponding aggregate coding features, target coding features, and word vector coding features, and the method may include the following steps:
step S131, carrying out data cleaning and noise reduction on the click time sequence data set to obtain a target click time sequence data set;
step S132, performing data characteristic processing on the target click time sequence data set to obtain corresponding aggregation coding characteristics, target coding characteristics and word vector coding characteristics.
In the process, the method carries out data cleaning and noise reduction on the click time sequence data set, can effectively reduce data interference, and can more accurately obtain corresponding aggregation coding features, target coding features and word vector coding features, thereby more accurately predicting user information.
Optionally, when the click time sequence data set is subjected to data preprocessing to obtain corresponding word vector coding features, the click time sequence data set may be subjected to data preprocessing through a preset word2vec model to obtain corresponding word vector coding features.
In the process, the method carries out data preprocessing on the click time sequence data set through a preset word2vec model to obtain the corresponding word vector coding features, can learn the interrelation and the internal rule among click data, and well obtain the corresponding word vector coding features, so that the prediction of user information of electric business users is facilitated.
In this embodiment, the predetermined deep learning fusion model may include a predetermined Transform model, a predetermined LightGBM model, and a predetermined Bi-LSTM + Attention model.
In the method for predicting e-commerce user information of the embodiment of the application, in step S140, when the aggregation coding feature, the target coding feature and the word vector coding feature are input to a preset deep learning fusion model to obtain a corresponding prediction result, the aggregation coding feature, the target coding feature and the word vector coding feature are correspondingly input to a preset Transform model, a preset LightGBM model and a preset Bi-LSTM + Attention model to obtain a corresponding prediction result.
It is understood that the aggregate coding features are input into a preset Transform model, the target coding features are input into a preset LightGBM model, and the word vector coding features are input into a preset Bi-LSTM + Attention model.
Wherein, the schematic diagram of the model structure of the Transform model can be seen in fig. 4, and the schematic diagram of the model structure of the Bi-LSTM + Attention model can be seen in fig. 5.
The Transform model consists of the left N coding models and the right N decoding models, N usually being taken to be 6. Encoders are stacked together by 6 identical layers, each layer has two branch layers, the first branch layer is a Multi-Head Attention mechanism, and the second branch layer is a simple fully-connected feedforward network Feed Forward. A residual connection was added on the outside of both branches, followed by a layer nomalization operation. Decoders also stacks six identical layers, but each layer adds a third sub-layer in addition to those two sub-layers in the encoder, again using residual and layer normalization as shown in FIG. 4. The prediction result corresponding to the aggregation coding characteristics can be obtained more accurately through a Transform model.
The LightGBM model uses a decision tree algorithm based on histogram, so that the efficiency can be higher and the memory can be saved. Two new technologies and an improvement are introduced on the basis of the traditional GBDT, wherein (1) the Gradient-based One-Side Sampling (GOSS) technology can remove a large part of data with small Gradient, only the rest data is used for estimating information gain, and the influence of a low-Gradient long tail part is avoided; (2) the Exclusive Feature Bundling (EFB) technology is used for binding mutually Exclusive features to reduce the number of the features; (3) in the conventional GBDT algorithm, the most time-consuming step is to find the optimal division point, the conventional method is a Pre-Sorted method, which enumerates all possible feature points on the ordered feature values, and the conventional Pre-Sorted method is replaced by the histogram algorithm in the LightGBM model. The prediction result corresponding to the target coding feature can be obtained more accurately through the LightGBM model.
The Bi-LSTM + Attention model adds an Attention layer to the Bi-LSTM model, the output vector of the last time sequence is used as a feature vector in the Bi-LSTM model, then softmax classification is carried out, Attention can calculate the weight of each time sequence, then the vectors of all time sequences are weighted and used as the feature vector, then softmax classification is carried out, and the result can be better promoted by adding the Attention layer. The prediction result of the corresponding word vector coding characteristics can be more accurately obtained through the Bi-LSTM + Attention model.
In the process, the method predicts the corresponding prediction result more accurately and further predicts the user information more accurately through the one-to-one correspondence of the preset Transform model and the aggregation coding feature, the preset LightGBM model and the target coding feature, and the preset Bi-LSTM + Attention model and the word vector coding feature.
It should be noted that in other embodiments, the Transform model may also adopt other deep learning models, for example, RNN deep learning model or LSTM model, etc.; the LightGBM model may also employ other deep learning models, such as the XGBOOST model, etc.; the Bi-LSTM + Attention model may also employ other deep learning models, such as an RNN deep learning model or an LSTM model, etc.
In this embodiment, in the method for predicting the electric utility customer information according to the embodiment of the present application, in step S150, when the customer information of the electric utility customer is predicted according to the prediction result, the customer information of the electric utility customer can be predicted according to the prediction result and a preset weighting coefficient.
It is understood that the user information of the e-commerce user is calculated by means of weighted summation.
In the process, the method combines the preset weighting coefficient, and can accurately predict the user information of the electric commercial user.
Example two
In order to implement the corresponding method of the above-mentioned embodiments to achieve the corresponding functions and technical effects, an e-commerce user information prediction apparatus is provided below.
Referring to fig. 6, fig. 6 is a block diagram illustrating a structure of an electric utility customer information prediction apparatus according to an embodiment of the present disclosure.
The utility model provides an electricity merchant user information prediction device includes:
the obtaining module 210 is configured to obtain click data of content clicked by an e-commerce user within a predetermined time and attribute data of the clicked content;
a generating module 220, configured to form a click timing sequence data set of the e-commerce user according to the click data and the attribute data;
the data processing module 230 is configured to perform data preprocessing on the click timing sequence data set to obtain corresponding aggregate coding features, target coding features, and word vector coding features;
the model prediction module 240 is configured to input the aggregate coding feature, the target coding feature and the word vector coding feature to a preset deep learning fusion model to obtain a corresponding prediction result;
and the information prediction module 250 is used for predicting the user information of the e-commerce user according to the prediction result.
The utility model discloses an electricity merchant user information prediction device, click data and the attribute data of click content that the electricity merchant user clicked within the predetermined time form the click time sequence data set of electricity merchant user, carry out the preliminary treatment of data to click time sequence data set, carry out the user information prediction of electricity merchant user with corresponding aggregate coding feature, target coding feature and word vector coding feature that obtain, it not only considers the historical consumption information of electricity merchant user, also considers the click browse information of electricity merchant user, can help the prediction of electricity merchant user's user information, and, still predict through the predetermined deep learning fusion model, the predetermined deep learning fusion model can be more suitable for the operation under the complex environment of huge user quantity, massive data, the prediction of electricity merchant user's user information better, and then can predict user information more accurately, and the robustness is also better, so that the recommended conversion rate of the e-commerce platform becomes ideal.
As an optional implementation manner, the generating module 220 may specifically be configured to:
correspondingly linking the attribute data with click data;
and aggregating the linked data by the E-commerce users, and sequencing the data by time to form a click time sequence data set of the E-commerce users.
As an optional implementation manner, the data processing module 230 may specifically be configured to:
carrying out data cleaning and noise reduction on the click time sequence data set to obtain a target click time sequence data set;
and performing data characteristic processing on the target click time sequence data set to obtain corresponding aggregation coding characteristics, target coding characteristics and word vector coding characteristics.
As an optional implementation manner, when the data processing module 230 performs data preprocessing on the click time sequence data set to obtain a corresponding word vector coding feature, the data processing module may perform data preprocessing on the click time sequence data set through a preset word2vec model to obtain a corresponding word vector coding feature.
As an optional implementation manner, the preset deep learning fusion model includes a preset Transform model, a preset LightGBM model, and a preset Bi-LSTM + Attention model;
model prediction module 240 may be specifically configured to:
and correspondingly inputting the aggregation coding characteristics, the target coding characteristics and the word vector coding characteristics into a preset Transform model, a preset LightGBM model and a preset Bi-LSTM + Attention model to obtain a corresponding prediction result.
As an alternative implementation, the information prediction module 250 may specifically be configured to:
and predicting to obtain the user information of the electric commercial user according to the prediction result and a preset weighting coefficient.
The electric utility customer information prediction device can implement the electric utility customer information prediction method of the first embodiment. The alternatives in the first embodiment are also applicable to the present embodiment, and are not described in detail here.
The rest of the embodiments of the present application may refer to the contents of the first embodiment, and in this embodiment, details are not repeated.
EXAMPLE III
An embodiment of the present application provides an electronic device, which includes a memory and a processor, where the memory is used to store a computer program, and the processor runs the computer program to make the electronic device execute the foregoing method for predicting the e-commerce user information.
Alternatively, the electronic device may be a server.
In addition, an embodiment of the present application further provides a computer-readable storage medium, which stores a computer program, and the computer program, when executed by a processor, implements the foregoing method for predicting the user information of the e-commerce.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus and method can be implemented in other ways. The apparatus embodiments described above are merely illustrative, and for example, the flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
In addition, functional modules in the embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.
The functions, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application or portions thereof that substantially contribute to the prior art may be embodied in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
The above description is only an example of the present application and is not intended to limit the scope of the present application, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present application shall be included in the protection scope of the present application. It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures.
The above description is only for the specific embodiments of the present application, but the scope of the present application is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present application, and shall be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.