CN113760521A - Virtual resource allocation method and device - Google Patents
Virtual resource allocation method and device Download PDFInfo
- Publication number
- CN113760521A CN113760521A CN202011003976.6A CN202011003976A CN113760521A CN 113760521 A CN113760521 A CN 113760521A CN 202011003976 A CN202011003976 A CN 202011003976A CN 113760521 A CN113760521 A CN 113760521A
- Authority
- CN
- China
- Prior art keywords
- user
- virtual
- allocated
- resource
- virtual resources
- 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
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5061—Partitioning or combining of resources
- G06F9/5077—Logical partitioning of resources; Management or configuration of virtualized resources
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
Abstract
The invention discloses a virtual resource allocation method and device, and relates to the technical field of internet. One embodiment of the method comprises: obtaining historical behavior data of a first user, wherein the historical behavior data comprises: a characteristic of the first user with respect to the virtual resource; obtaining behavior preference of a first user about the virtual resources according to historical behavior data by using a first classification model, wherein the first classification model is obtained by training according to the historical behavior data of a plurality of second users; according to the behavior preference and the attribute of the virtual resource to be distributed, predicting the probability of the first user using the virtual resource to be distributed by using a prediction model, wherein the prediction model is obtained by training according to historical behavior data of a plurality of second users using the virtual resource and the attribute of the virtual resource used by the second users; and when the probability is greater than a preset threshold value, allocating the virtual resources to be allocated to the first user. The embodiment improves the precision of virtual resource allocation and the utilization rate of the virtual resources.
Description
Technical Field
The present invention relates to the field of internet technologies, and in particular, to a method and an apparatus for allocating virtual resources.
Background
In the e-commerce field, with the wide application of mobile payment, an e-commerce platform usually pushes a message to a user through a pop-up window or other modes on a website or app to remind the user to obtain virtual resources provided by the platform.
In the process of implementing the invention, the inventor finds that at least the following problems exist in the prior art:
at present, a virtual resource is provided for a user by an e-commerce platform in a random mode, and the user is not distinguished, so that the allocation of the virtual resource is not accurate enough, the utilization rate of the virtual resource provided by the platform is low, and the waste of the virtual resource is caused.
Disclosure of Invention
In view of this, embodiments of the present invention provide a method and an apparatus for allocating virtual resources, which can obtain behavior preferences of a first user according to historical behavior data of the first user, predict a probability that the first user uses the virtual resources to be allocated according to the behavior preferences of the first user regarding the virtual resources and attributes of the virtual resources to be allocated, and allocate the virtual resources to be allocated to a first user with a high usage probability, thereby improving accuracy of allocating the virtual resources and a usage rate of the virtual resources.
To achieve the above object, according to an aspect of the embodiments of the present invention, a method for allocating virtual resources is provided.
The method for allocating the virtual resources comprises the following steps:
obtaining historical behavior data of a first user, wherein the historical behavior data comprises: a characteristic of the first user with respect to the virtual resource;
obtaining behavior preference of the first user about the virtual resources according to historical behavior data by using a first classification model; the first classification model is obtained by training according to historical behavior data of a plurality of second users;
predicting the probability of the first user using the virtual resources to be distributed by using a prediction model according to the behavior preference and the attributes of the virtual resources to be distributed, wherein the prediction model is obtained by training according to historical behavior data of a plurality of second users using the virtual resources and the attributes of the virtual resources used by the second users;
and when the probability is greater than a preset threshold value, allocating the virtual resources to be allocated to the first user.
Alternatively,
the attributes of the virtual resources to be allocated include: provider information of the virtual resources to be allocated;
obtaining behavior preference of the first user about the resource provider according to historical behavior data by using a first classification model;
and according to the behavior preference and the provider information of the virtual resources to be allocated, predicting the probability of the first user using the virtual resources to be allocated provided by the resource provider by using a prediction model.
Alternatively,
the attributes of the virtual resource to be allocated further include: the value of the virtual resource to be allocated;
and predicting the probability of the first user using the virtual resource to be allocated provided by the resource provider by using a prediction model according to the behavior preference of the first user on the resource provider and the value of the virtual resource to be allocated.
Alternatively,
acquiring historical transaction data of a resource provider, wherein the historical transaction data comprises: the corresponding resource provider provides the characteristics of the virtual resource;
respectively determining attribute preference of the resource provider about the virtual resources according to historical transaction data by utilizing a second classification model;
obtaining behavior preference of the first user about the resource provider according to historical behavior data by using a first classification model;
and predicting the probability of the first user using the virtual resource to be allocated provided by the resource provider by using a prediction model according to the attribute preference, the behavior preference of the first user about the resource provider and the attribute of the virtual resource to be allocated.
Alternatively,
the method further comprises the following steps:
determining a plurality of original characteristics of the historical behavior data of the first user, and inputting the plurality of original characteristics into a characteristic selector to obtain the characteristics of the first user about the virtual resources.
Alternatively,
the prediction model is obtained by training according to a random forest algorithm and any three of a GBDT algorithm, a Catboost algorithm, a LightGBM algorithm, an XGboost algorithm and an LR algorithm.
Alternatively,
respectively constructing three initial models by using any three of a GBDT algorithm, a Catboost algorithm, a LightGBM algorithm, an XGboost algorithm and an LR algorithm;
respectively inputting the historical behavior data of the second user into the three initial models to train the three initial models;
and taking the output of the three trained initial models and the behavior preference of a second user on the virtual resources output by the first classification model as the input of a random forest algorithm to train the prediction model.
Alternatively,
the first classification model and the second classification model are obtained based on clustering algorithm training.
Alternatively,
the feature selector is derived based on the feature _ selection library.
To achieve the above object, according to another aspect of the embodiments of the present invention, an apparatus for allocating virtual resources is provided.
An apparatus for allocating virtual resources according to an embodiment of the present invention includes: the device comprises a data acquisition module, a classification module, a prediction module and an allocation module; wherein the content of the first and second substances,
the data acquisition module is used for acquiring historical behavior data of a first user, wherein the historical behavior data comprises: a characteristic of the first user with respect to the virtual resource;
the classification module is used for obtaining behavior preference of the first user about the virtual resources according to the historical behavior data acquired by the data acquisition module by utilizing the first classification model; the first classification model is obtained by training according to historical behavior data of a plurality of second users;
the prediction module is used for predicting the probability of the first user using the virtual resources to be distributed by using a prediction model according to the behavior preference and the attributes of the virtual resources to be distributed, which are obtained by the classification module, wherein the prediction model is obtained by training according to historical behavior data of a plurality of second users using the virtual resources and the attributes of the virtual resources used by the second users;
and the distribution module is used for distributing the virtual resources to be distributed to the first user when the probability predicted by the prediction module is greater than a preset threshold value.
To achieve the above object, according to still another aspect of the embodiments of the present invention, there is provided a virtual resource allocation server.
The virtual resource allocation server of the embodiment of the invention comprises: one or more processors; the storage device is used for storing one or more programs, and when the one or more programs are executed by one or more processors, the one or more processors implement the allocation method of the virtual resources according to the embodiment of the invention.
To achieve the above object, according to still another aspect of embodiments of the present invention, there is provided a computer-readable storage medium.
A computer-readable storage medium of an embodiment of the present invention stores thereon a computer program, and when the computer program is executed by a processor, the computer program implements a virtual resource allocation method of an embodiment of the present invention.
One embodiment of the above invention has the following advantages or benefits: after historical behavior data of a first user are obtained, the historical behavior data containing relevant characteristics of virtual resources are firstly input into a first classification model to obtain behavior preference of the first user about the virtual resources, then the behavior preference and attributes of the virtual resources to be distributed are input into a prediction model to predict the probability of the first user using the virtual resources to be distributed, and if the obtained probability is larger than a first threshold value, the virtual resources to be distributed are distributed to the first user. As can be seen from the above description, according to the embodiment of the present invention, the first user can be divided into the corresponding user types according to the historical behavior data of the first user, the behavior preference of the first user is obtained according to the behavior preference of the corresponding user type, the probability that the first user uses the virtual resource to be allocated is predicted according to the behavior preference of the first user on the virtual resource and the attribute of the virtual resource to be allocated, and the virtual resource to be allocated is allocated to the first user with a high usage probability, so that the accuracy of allocating the virtual resource and the usage rate of the virtual resource are improved.
Further effects of the above-mentioned non-conventional alternatives will be described below in connection with the embodiments.
Drawings
The drawings are included to provide a better understanding of the invention and are not to be construed as unduly limiting the invention. Wherein:
fig. 1 is a schematic diagram of main steps of a method for allocating virtual resources according to an embodiment of the present invention;
FIG. 2 is a diagram illustrating the main steps of another virtual resource allocation method according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of the main steps of a training method of a predictive model according to an embodiment of the invention;
FIG. 4 is a diagram illustrating the main steps of another virtual resource allocation method according to an embodiment of the present invention;
FIG. 5 is a diagram illustrating the main steps of another virtual resource allocation method according to an embodiment of the present invention;
fig. 6 is a schematic diagram of main modules of an apparatus for allocating virtual resources according to an embodiment of the present invention;
FIG. 7 is an exemplary system architecture diagram in which embodiments of the present invention may be employed;
fig. 8 is a schematic structural diagram of a computer system suitable for implementing a terminal device or a server according to an embodiment of the present invention.
Detailed Description
Exemplary embodiments of the present invention are described below with reference to the accompanying drawings, in which various details of embodiments of the invention are included to assist understanding, and which are to be considered as merely exemplary. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the invention. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.
It should be noted that the embodiments of the present invention and the technical features of the embodiments may be combined with each other without conflict.
Fig. 1 is a schematic diagram illustrating main steps of a method for allocating virtual resources according to an embodiment of the present invention.
As shown in fig. 1, a method for allocating virtual resources according to an embodiment of the present invention mainly includes the following steps:
step S101: obtaining historical behavior data of a first user, wherein the historical behavior data comprises: a characteristic of the first user with respect to the virtual resource.
In the embodiment of the present invention, when the historical behavior data of the first user includes both the characteristics of the first user about the virtual resource and the characteristics of the first user unrelated to the virtual resource, the characteristics of the first user about the virtual resource may be screened from the historical behavior data of the first user, and then the subsequent steps are performed. For example, the historical behavior data of a certain first user includes: the number of times of logging in by the first user, the commodities browsed by the first user within the last 30 days, the number of times of consumption by the first user, the average amount of money consumed by the first user each time, the number of times of getting the virtual resource by the first user, the number of times of using the virtual resource by the first user, the amount of money of the virtual resource used by the first user, the average time interval from getting the virtual resource by the first user to using the virtual resource, the minimum consumption amount of money after the virtual resource is used by the first user, and the like. From the above historical behavioral data, it can be seen that: the number of times that the first user logs in, the commodities browsed by the first user in the last 30 days, the number of times that the first user consumes and the average amount of money consumed by each user by the first user belong to characteristics irrelevant to the virtual resources, and the characteristics of the number of times that the first user draws the virtual resources, the number of times that the first user uses the virtual resources, the amount of money used by the first user, the average time interval from drawing the virtual resources by the first user to using the virtual resources and the minimum amount of money consumed after the first user uses the virtual resources are relevant to the virtual resources, at this time, a characteristic selector can be used to filter the characteristics irrelevant to the virtual resources, and subsequent steps are carried out after the characteristics relevant to the virtual resources are screened.
Therefore, as shown in fig. 2, an embodiment of the present invention further provides another virtual resource allocation method, which mainly includes the following steps S201 to S206:
step S201: acquiring historical behavior data of a first user;
step S202: determining a plurality of original characteristics of historical behavior data of a first user, wherein the original characteristics comprise characteristics of the first user about virtual resources and characteristics of the first user unrelated to the virtual resources;
step S203: inputting a plurality of original characteristics into a characteristic selector to obtain the characteristics of the first user about the virtual resources;
step S204: obtaining behavior preference of a first user about virtual resources according to characteristics of the first user about the virtual resources by using a first classification model, wherein the first classification model is obtained by training according to historical behavior data of a plurality of second users;
step S205: predicting the probability of the first user using the virtual resources to be distributed by using a prediction model according to the behavior preference and the attributes of the virtual resources to be distributed, wherein the prediction model is obtained by training according to historical behavior data of a plurality of second users using the virtual resources and the attributes of the virtual resources used by the second users;
step S206: and when the probability is larger than a preset threshold value, allocating the virtual resource to the first user.
In this embodiment of the present invention, the feature selector may be obtained based on a feature _ selection library in Python, and is used to remove irrelevant features and redundant features in the historical behavior data of the first user according to the needs of the model in the subsequent step: irrelevant features, i.e. features that are not highly relevant to the model, e.g. features that are irrelevant to the virtual resources of the first user, belong to this category, should be eliminated; redundant features, i.e., features that do not diverge, e.g., features with variance close to 0, indicate that the historical behavior data contains the features, and the features contain little information and can be eliminated. The feature selector can reduce the number of features, reduce the data dimensionality and improve the accuracy of the output result of the subsequent model.
Step S102: obtaining behavior preference of the first user about the virtual resources according to historical behavior data by using a first classification model; the first classification model is trained according to historical behavior data of a plurality of second users.
In the embodiment of the present invention, since first users with similar historical behavior data may have similar behavior preferences for virtual resources with different attributes, the first user may be categorized by using the first classification model to obtain the behavior preferences of the first user with respect to the virtual resources. The first classification model may be configured to compare historical behavior data of the first user with parameters corresponding to user categories existing in the first classification model, classify the first user into an appropriate user category according to a comparison result, and obtain behavior preference of the first user about virtual resources according to behavior preference of the user of the type about the virtual resources. For example, historical behavior data of a first user shows that, in the last month, the first user has taken five times of virtual resources and used twice of virtual resources, and the first classification model compares the data with parameters corresponding to existing user categories to obtain that the first user belongs to the user category of "occasionally using virtual resources", so that the behavior preference of the first user for virtual resources is "occasionally used" according to the behavior preference of the user category, and the behavior preference of the first user for "occasionally using virtual resources is obtained.
In the embodiment of the invention, the first classification model is obtained by training a Clustering algorithm according to historical behavior data of a plurality of second users, wherein the Clustering algorithm can be a DBSCAN (Dens-Based Spatial Clustering of Applications with Noise Based on density) algorithm or a K-means algorithm. Wherein the second user may be a user having similar historical behavioral data as the first user, e.g., the first user and the second user are users using the same e-commerce platform. In a preferred embodiment of the present invention, the first classification model is based on DBSCAN training. The training process of the first classification model specifically includes: inputting a plurality of predefined original features into a feature selector to obtain sample features, wherein the sample features comprise: a feature associated with a virtual resource; acquiring historical behavior data of a plurality of second users according to the sample characteristics; inputting historical behavior data of a plurality of second users into a first classification model, classifying the plurality of second users into different user categories according to the historical behavior data, a preset neighborhood distance threshold epsilon and a sample number threshold MinPts in a neighborhood by the first classification model, and outputting the number of the user categories obtained by the classification, wherein each user category may have different behavior preferences for the same virtual resource. For example, the output after the first classification model is trained is "users in X class", which means that a plurality of second users of the input are divided into different user classes in X class.
Step S103: and predicting the probability of the first user using the virtual resources to be distributed by using a prediction model according to the behavior preference and the attributes of the virtual resources to be distributed, wherein the prediction model is obtained by training according to historical behavior data of a plurality of second users using the virtual resources and the attributes of the virtual resources used by the second users.
Step S104: and when the probability is greater than a preset threshold value, allocating the virtual resources to be allocated to the first user.
In the embodiment of the invention, the prediction model is obtained by training according to any three of a random forest algorithm, a GBDT algorithm, a Catboost algorithm, a LightGBM algorithm, an XGboost algorithm and an LR algorithm. As shown in fig. 3, the training method of the prediction model mainly includes the following steps S301 to S306:
step S301: and respectively constructing three initial models as the first layers of the prediction models by using any three of the GBDT algorithm, the Catboost algorithm, the LightGBM algorithm, the XGboost algorithm and the LR algorithm.
In a preferred embodiment of the present invention, the GBDT algorithm, the Catboost algorithm, and the LightGBM algorithm are selected to construct three initial models, respectively.
Step S302: and randomly dividing the historical behavior data of the second user into a training set and a verification set.
For example, the training set may include 80% of the historical behavior data and the validation set may include an additional 20% of the historical behavior data.
Step S303: and respectively inputting the training set into the three initial models to train the three initial models to obtain importance scores of the features included in the training set, and taking the features which are more than the median of the importance scores and are included in the training set as first output of a first layer of the prediction model.
Step S304: and respectively inputting the verification set into the three initial models to respectively carry out parameter adjustment and optimization on the three initial models, obtaining importance scores of the features included in the verification set, and taking the features which are more than the median of the importance scores and are included in the verification set as second output of the first layer of the prediction model.
Step S305: and taking a random forest algorithm as a second layer of the prediction model, inputting the first output and the second output of the first layer of the trained prediction model and behavior preference of a second user output by the first classification model on the virtual resources into the random forest algorithm, and training the second layer of the prediction model.
Step S306: and randomly selecting partial data from the historical behavior data of the second user as a test set, and performing parameter adjustment optimization on the second layer of the prediction model.
In a preferred embodiment of the invention, the prediction model is trained according to a random forest algorithm, a GBDT algorithm, a Catboost algorithm and a LightGBM algorithm, and the operation rate and the result accuracy of the prediction model trained by the four algorithms are higher. The GBDT algorithm generates a weak classifier through multiple iterations, each iteration generates a weak classifier, a loss function descends along the gradient direction, each classifier is trained on the basis of the residual error of the last classifier, and finally the weak classifiers obtained through each training are weighted and summed to obtain a final total classifier; the Catboost algorithm can automatically process the class characteristics in a special mode, namely, firstly, statistics is carried out on the class characteristics, the frequency of the certain class characteristics is calculated, then, the super parameters are added, and a new numerical characteristic is generated so as to avoid manual processing of the class characteristics; the LightGBM algorithm is a decision tree algorithm based on a histogram algorithm, uses a leaf-based growth algorithm with depth limitation, adds a limitation of maximum depth on the algorithm to ensure high efficiency and prevent overfitting, optimizes the support of class characteristics, can directly input the class characteristics, has good efficiency and expandability when facing high-dimensional big data, and has processing speed three times that of a common algorithm; the random forest algorithm improves the establishment of the decision tree, and further enhances the generalization capability of the model by randomly selecting a part of sample characteristics on the nodes and then selecting an optimal characteristic from the sample characteristics to divide left and right subtrees of the decision tree.
In a preferred embodiment of the present invention, a Blending method may also be adopted to perform model fusion on the above four algorithms, so as to obtain an optimal prediction model and further improve the overall performance of the prediction model.
In the embodiment of the present invention, the output of the first classification model, that is, the behavior preference of the first user with respect to the virtual resource and the attribute of the virtual resource to be allocated are input into the prediction model, so as to obtain the probability that the first user uses the virtual resource to be allocated, where the attribute of the virtual resource to be allocated includes information related to the virtual resource to be allocated. The information included in the attributes of the virtual resources to be allocated may be different, and may include, for example, provider information thereof, and may also include a value thereof itself.
The following describes a method for allocating virtual resources according to an embodiment of the present invention, by taking an example that attributes of virtual resources to be allocated include provider information of the virtual resources to be allocated.
As shown in fig. 4, an embodiment of the present invention provides another virtual resource allocation method, where the method mainly includes the following steps S401 to S404:
step S401: obtaining historical behavior data of a first user, wherein the historical behavior data comprises: a characteristic of the first user with respect to the virtual resource.
Step S402: and utilizing the first classification model to obtain the behavior preference of the first user about the resource provider according to the historical behavior data.
In the embodiment of the present invention, since first users with similar historical behavior data may have similar behavior preferences for different resource providers, the first user may be categorized using the first classification model to obtain the behavior preferences of the first user with respect to the resource providers. For example, when the virtual resource is a virtual resource on an e-commerce platform, the resource provider may be an operator of the e-commerce platform or a seller who is resident on the e-commerce platform, and the virtual resource is an electronic ticket that can participate in an event, or a coupon redeemable for a discount amount, a service, or a product.
For example, if a first user uses virtual resources related to articles of manufacture provided by a resource provider twice within a month according to historical behavior data of the first user, the first classification model may classify the first user into a user category of "browsing the articles of manufacture category resource provider" to indicate that the first user has a higher behavior preference for the resource provider providing the articles of manufacture of use.
Step S403: and according to the behavior preference and the provider information of the virtual resources to be allocated, predicting the probability of the first user using the virtual resources to be allocated provided by the resource provider by using a prediction model.
In the embodiment of the present invention, according to the provider information of the virtual resource to be allocated included in the attribute of the virtual resource to be allocated, the prediction model may compare the behavior preference of the first user with respect to the resource provider with the provider information of the virtual resource to be allocated, thereby predicting the probability that the first user uses the virtual resource to be allocated provided by the resource provider. For example, if the first user has a high behavioral preference for the resource provider providing the articles of the articles of the articles of the articles of the articles of the literature, the first user of the article of articles of the article of articles of the article of the first user of the article of the first user of the article of the first user of the article are providing the article of the first user of the article of the resource of the article of the first user of the article; and if the first user has a higher behavior preference for the resource provider providing the articles of articles for making articles.
In this embodiment of the present invention, the attribute of the virtual resource to be allocated may further include: the value of the virtual resource to be allocated. The predictive model may predict a probability that the first user uses the virtual resource to be allocated provided by the resource provider based on behavioral preferences of the first user with respect to the resource provider and a value of the virtual resource to be allocated. For example, if the first user has a high behavioral preference for a resource provider that provides items with a mean price of 30 dollars, and the value of the virtual resource to be allocated is full 50 minus 6, the probability that the first user uses the virtual resource to be allocated provided by the resource provider is high; and if the first user has a higher behavioral preference for a resource provider that provides an item with a mean price of 30 dollars, but the value of the virtual resource to be allocated is full 1000 minus 100, the probability that the first user uses the virtual resource to be allocated provided by the resource provider is lower.
Step S404: and when the probability is greater than a preset threshold value, allocating the virtual resources to be allocated to the first user.
In addition, in a preferred embodiment of the present invention, historical transaction data of the resource provider may also be obtained, and according to the historical behavior data of the first user and the historical transaction data of the resource provider, a probability that the first user uses the virtual resource to be allocated provided by the resource provider is predicted, so as to further improve accuracy of a prediction result, and further improve accuracy of allocating the virtual resource to be allocated provided by the resource provider to the target first user. As shown in fig. 5, an embodiment of the present invention provides another virtual resource allocation method, which mainly includes the following steps S501 to S505:
step S501: acquiring historical behavior data of a first user and historical transaction data of a resource provider, wherein the historical behavior data comprises: a characteristic of the first user with respect to the virtual resource; the historical transaction data includes: the corresponding resource provider provides the characteristics of the virtual resource.
In the embodiment of the present invention, when the historical behavior data of the first user includes both the characteristics of the first user about the virtual resource and the characteristics of the first user unrelated to the virtual resource, the characteristics selector may be used to screen out the characteristics of the first user about the virtual resource from the historical behavior data of the first user, and then the subsequent steps are performed. Similarly, when the historical transaction data of the resource provider includes both the characteristics of the virtual resource provided by the resource provider and the characteristics of the virtual resource provided by the resource provider, the characteristics of the virtual resource provided by the resource provider can be screened from the historical transaction data of the resource provider by using the characteristic selector, and then the subsequent steps are performed.
In an embodiment of the present invention, the characteristics of the virtual resource provided by the resource provider may include: the value of the virtual resources provided by the resource provider, the number of times the virtual resources provided by the resource provider are provided, the number of times the virtual resources provided by the resource provider are picked up, the number of times the virtual resources provided by the resource provider are used, the pick-up-usage rate of the virtual resources provided by the resource provider, and the like.
Step S502: and respectively determining attribute preference of the resource provider about the virtual resources according to historical transaction data by utilizing a second classification model.
In the embodiment of the invention, the second classification model is obtained by training a Clustering algorithm according to historical transaction data of a plurality of second users, wherein the Clustering algorithm can be DBSCAN (Dens-Based Spatial Clustering of Applications with Noise Based on density) or K-means algorithm. In a preferred embodiment of the present invention, the second classification model is also based on DBSCAN training. The training process of the second classification model is similar to the training process of the first classification model, and is not described herein again.
In the embodiment of the present invention, since resource providers with similar historical transaction data may have similar attribute preferences for virtual resources with different attributes, the resource providers may be categorized by using the second classification model to obtain the attribute preferences of the resource providers about the virtual resources. The second classification model may compare the historical transaction data of the resource provider with the parameters corresponding to the user categories existing in the second classification model, classify the resource provider into a proper user category according to the comparison result, and further obtain the attribute preference of the resource provider about the virtual resources according to the attribute preference of the user to the virtual resources. For example, if the historical transaction data of a resource provider shows that the resource provider has issued virtual resources related to articles for use for writing twice within one month, the second classification model may compare the data with the parameters corresponding to the existing user categories to obtain that the resource provider belongs to the user category of "documents for use for writing frequently issued virtual resources", and thus, according to the attribute preference of the user category on the virtual resources, the attribute preference of the resource provider on "documents for use for writing frequently issued" is obtained.
Step S503: and utilizing the first classification model to obtain the behavior preference of the first user about the resource provider according to the historical behavior data.
In the embodiment of the present invention, since first users with similar historical behavior data may have similar behavior preferences for different resource providers, the first user may be categorized using the first classification model to obtain the behavior preferences of the first user with respect to the resource providers. For example, if a first user uses virtual resources related to articles of manufacture provided by a resource provider twice within a month according to historical behavior data of the first user, the first classification model may classify the first user into a user category of "resource providers like browsing articles of manufacture" to indicate that the first user has a higher behavior preference for the resource provider providing the articles of manufacture of goods.
Step S504: and according to the attribute preference, the behavior preference and the attribute of the virtual resource to be allocated, predicting the probability of the first user using the virtual resource to be allocated provided by the resource provider by using a prediction model.
In the embodiment of the present invention, since it is known through step S501 that the virtual resource to be allocated is provided by the resource provider, the provider of the virtual resource to be allocated may not be included in the attribute of the virtual resource to be allocated.
In the embodiment of the invention, the attribute preference of the resource provider on the virtual resource, the behavior preference of the first user on the resource provider and the attribute of the virtual resource to be allocated are input into the prediction model, and the prediction model can predict the probability of the first user using the virtual resource to be allocated provided by the resource provider according to the corresponding relation of the three. For example, if the resource provider has an attribute preference of "frequently issuing articles for; and if the resource provider has attribute preference of "frequently issuing articles for literary use" virtual resources, but the first user has lower preference for behavior of the resource provider providing articles for literary use, the probability that the first user uses the virtual resources to be allocated, which are provided by the resource provider and have value of full 50 minus 6, is lower.
Step S505: and when the probability is greater than a preset threshold value, allocating the virtual resources to be allocated to the first user.
In the embodiment of the present invention, when the probability predicted by the prediction model is greater than the preset threshold, it indicates that the first user is likely to use the virtual resource to be allocated, so that the virtual resource to be allocated provided by the resource provider is allocated to the first user, so that the virtual resource can be allocated to the first user with a higher usage probability, and the usage rate of the virtual resource is further improved.
According to the virtual resource allocation method provided by the embodiment of the invention, after the historical behavior data of the first user is obtained, the historical behavior data containing the relevant characteristics of the virtual resources are firstly input into the first classification model to obtain the behavior preference of the first user about the virtual resources, then the behavior preference and the attributes of the virtual resources to be allocated are input into the prediction model to predict the probability of the first user using the virtual resources to be allocated, and if the obtained probability is greater than the first threshold, the virtual resources to be allocated are allocated to the first user. As can be seen from the above description, according to the embodiment of the present invention, the first user can be divided into the corresponding user types according to the historical behavior data of the first user, the behavior preference of the first user is obtained according to the behavior preference of the corresponding user type, the probability that the first user uses the virtual resource to be allocated is predicted according to the behavior preference of the first user on the virtual resource and the attribute of the virtual resource to be allocated, and the virtual resource to be allocated is allocated to the first user with a high usage probability, so that the accuracy of allocating the virtual resource and the usage rate of the virtual resource are improved.
Fig. 6 is a schematic diagram of main blocks of an apparatus for allocating virtual resources according to an embodiment of the present invention.
As shown in fig. 6, an apparatus for allocating virtual resources according to an embodiment of the present invention includes: a data acquisition module 601, a classification module 602, a prediction module 603 and an allocation module 604; wherein the content of the first and second substances,
a data obtaining module 601, configured to obtain historical behavior data of the first user, where the historical behavior data includes: a characteristic of the first user with respect to the virtual resource;
the classification module 602 is configured to obtain, by using the first classification model, behavior preference of the first user with respect to the virtual resource according to the historical behavior data obtained by the data obtaining module 601; the first classification model is obtained by training according to historical behavior data of a plurality of second users;
a predicting module 603, configured to predict, according to the behavior preference and the attribute of the virtual resource to be allocated obtained by the classifying module 602, a probability that the first user uses the virtual resource to be allocated by using a prediction model, where the prediction model is obtained by training according to historical behavior data of a plurality of second users using the virtual resource and the attribute of the virtual resource used by the second users;
an allocating module 604, configured to allocate the virtual resource to be allocated to the first user when the probability predicted by the predicting module 603 is greater than a preset threshold.
In the embodiment of the present invention, the attributes of the virtual resource to be allocated include: provider information of the virtual resources to be allocated; the classification module 602 may be further configured to obtain, by using the first classification model, behavior preferences of the first user with respect to the resource provider based on the historical behavior data; the prediction module 603 may be further configured to predict, according to the behavior preference and the provider information of the virtual resource to be allocated, a probability that the first user uses the virtual resource to be allocated provided by the resource provider by using a prediction model.
In this embodiment of the present invention, the attribute of the virtual resource to be allocated further includes: the value of the virtual resource to be allocated; the prediction module 603 may be further configured to predict, by using a prediction model, a probability that the first user uses the virtual resource to be allocated provided by the resource provider according to the behavior preference of the first user with respect to the resource provider and the value of the virtual resource to be allocated.
In this embodiment of the present invention, the data obtaining module 601 may further be configured to obtain historical transaction data of the resource provider, where the historical transaction data includes: the corresponding resource provider provides the characteristics of the virtual resource; the classification module 602 may be further configured to determine, by using the second classification model, attribute preferences of the resource provider regarding the virtual resources according to the historical transaction data, and obtain, by using the first classification model, behavior preferences of the first user regarding the resource provider according to the historical behavior data; the prediction module 603 may be further configured to predict, by using a prediction model, a probability that the first user uses the virtual resource to be allocated provided by the resource provider according to the attribute preference, the behavior preference of the first user with respect to the resource provider, and the attribute of the virtual resource to be allocated.
In this embodiment of the present invention, the data obtaining module 601 may further be configured to: determining a plurality of original characteristics of the historical behavior data of the first user, and inputting the plurality of original characteristics into a characteristic selector to obtain the characteristics of the first user about the virtual resources.
In the embodiment of the invention, the prediction model is obtained by training according to any three of a random forest algorithm, a GBDT algorithm, a Catboost algorithm, a LightGBM algorithm, an XGboost algorithm and an LR algorithm.
In this embodiment of the present invention, the prediction module 603 may further be configured to: respectively constructing three initial models by using any three of a GBDT algorithm, a Catboost algorithm, a LightGBM algorithm, an XGboost algorithm and an LR algorithm; respectively inputting the historical behavior data of the second user into the three initial models to train the three initial models; and taking the output of the three trained initial models and the behavior preference of a second user on the virtual resources output by the first classification model as the input of a random forest algorithm to train the prediction model.
In the embodiment of the invention, the first classification model and the second classification model are obtained based on clustering algorithm training.
In the embodiment of the present invention, the feature selector is obtained based on the feature _ selection library.
According to the virtual resource allocation device in the embodiment of the invention, after the historical behavior data of the first user is acquired, the historical behavior data containing the relevant characteristics of the virtual resources are firstly input into the first classification model to obtain the behavior preference of the first user about the virtual resources, then the behavior preference and the attributes of the virtual resources to be allocated are input into the prediction model to predict the probability of the first user using the virtual resources to be allocated, and if the obtained probability is greater than the first threshold, the virtual resources to be allocated are allocated to the first user. As can be seen from the above description, according to the embodiment of the present invention, the first user can be divided into the corresponding user types according to the historical behavior data of the first user, the behavior preference of the first user is obtained according to the behavior preference of the corresponding user type, the probability that the first user uses the virtual resource to be allocated is predicted according to the behavior preference of the first user on the virtual resource and the attribute of the virtual resource to be allocated, and the virtual resource to be allocated is allocated to the first user with a high usage probability, so that the accuracy of allocating the virtual resource and the usage rate of the virtual resource are improved.
Fig. 7 shows an exemplary system architecture 700 of an allocation method of virtual resources or an allocation apparatus of virtual resources to which an embodiment of the present invention can be applied.
As shown in fig. 7, the system architecture 700 may include terminal devices 701, 702, 703, a network 704, and a server 705. The network 704 serves to provide a medium for communication links between the terminal devices 701, 702, 703 and the server 705. Network 704 may include various connection types, such as wired, wireless communication links, or fiber optic cables, to name a few.
A user may use the terminal devices 701, 702, 703 to interact with a server 705 over a network 704, to receive or send messages or the like. Various communication client applications, such as shopping applications, web browser applications, search applications, instant messaging tools, mailbox clients, social platform software, and the like, may be installed on the terminal devices 701, 702, and 703.
The terminal devices 701, 702, 703 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop portable computers, desktop computers, and the like.
The server 705 may be a server that provides various services, such as a background management server that supports shopping websites browsed by users using the terminal devices 701, 702, and 703. The background management server may analyze and perform other processing on the received data such as the product information query request, and feed back a processing result (e.g., target push information and product information) to the terminal device.
It should be noted that the method for allocating virtual resources provided in the embodiment of the present invention is generally executed by the server 705, and accordingly, the virtual resource allocating apparatus is generally disposed in the server 705.
It should be understood that the number of terminal devices, networks, and servers in fig. 7 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation.
Referring now to FIG. 8, shown is a block diagram of a computer system 800 suitable for use with a terminal device implementing an embodiment of the present invention. The terminal device shown in fig. 8 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiments of the present invention.
As shown in fig. 8, the computer system 800 includes a Central Processing Unit (CPU)801 that can perform various appropriate actions and processes in accordance with a program stored in a Read Only Memory (ROM)802 or a program loaded from a storage section 808 into a Random Access Memory (RAM) 803. In the RAM 803, various programs and data necessary for the operation of the system 800 are also stored. The CPU 801, ROM 802, and RAM 803 are connected to each other via a bus 804. An input/output (I/O) interface 805 is also connected to bus 804.
The following components are connected to the I/O interface 805: an input portion 806 including a keyboard, a mouse, and the like; an output section 807 including a signal such as a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), and the like, and a speaker; a storage portion 808 including a hard disk and the like; and a communication section 809 including a network interface card such as a LAN card, a modem, or the like. The communication section 809 performs communication processing via a network such as the internet. A drive 810 is also connected to the I/O interface 805 as necessary. A removable medium 811 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like is mounted on the drive 810 as necessary, so that a computer program read out therefrom is mounted on the storage section 808 as necessary.
In particular, according to the embodiments of the present disclosure, the processes described above with reference to the flowcharts may be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising program code for performing the method illustrated in the flow chart. In such an embodiment, the computer program can be downloaded and installed from a network through the communication section 809 and/or installed from the removable medium 811. The computer program executes the above-described functions defined in the system of the present invention when executed by the Central Processing Unit (CPU) 801.
It should be noted that the computer readable medium shown in the present invention can be a computer readable signal medium or a computer readable storage medium or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the computer readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present invention, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In the present invention, however, a computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, fiber optic cable, RF, etc., or any suitable combination of the foregoing.
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present invention. 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 or flowchart illustration, and combinations of blocks in the block diagrams 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.
The modules described in the embodiments of the present invention may be implemented by software or hardware. The described modules may also be provided in a processor, which may be described as: a processor includes a data acquisition module, a classification module, a prediction module, and an assignment module. The names of these modules do not constitute a limitation to the module itself in some cases, for example, the data acquisition module may also be described as a "module that acquires historical behavior data of the first user".
As another aspect, the present invention also provides a computer-readable medium that may be contained in the apparatus described in the above embodiments; or may be separate and not incorporated into the device. The computer readable medium carries one or more programs which, when executed by a device, cause the device to comprise: obtaining historical behavior data of a first user, wherein the historical behavior data comprises: a characteristic of the first user with respect to the virtual resource; obtaining behavior preference of the first user about the virtual resources according to historical behavior data by using a first classification model; the first classification model is obtained by training according to historical behavior data of a plurality of second users; predicting the probability of the first user using the virtual resources to be distributed by using a prediction model according to the behavior preference and the attributes of the virtual resources to be distributed, wherein the prediction model is obtained by training according to historical behavior data of a plurality of second users using the virtual resources and the attributes of the virtual resources used by the second users; and when the probability is greater than a preset threshold value, allocating the virtual resources to be allocated to the first user.
According to the technical scheme of the embodiment of the invention, after the historical behavior data of the first user is obtained, the historical behavior data containing the relevant characteristics of the virtual resources are firstly input into a first classification model to obtain the behavior preference of the first user about the virtual resources, then the behavior preference and the attributes of the virtual resources to be allocated are input into a prediction model to predict the probability of the first user using the virtual resources to be allocated, and if the obtained probability is greater than a first threshold value, the virtual resources to be allocated are allocated to the first user. As can be seen from the above description, according to the embodiment of the present invention, the first user can be divided into the corresponding user types according to the historical behavior data of the first user, the behavior preference of the first user is obtained according to the behavior preference of the corresponding user type, the probability that the first user uses the virtual resource to be allocated is predicted according to the behavior preference of the first user on the virtual resource and the attribute of the virtual resource to be allocated, and the virtual resource to be allocated is allocated to the first user with a high usage probability, so that the accuracy of allocating the virtual resource and the usage rate of the virtual resource are improved.
The above-described embodiments should not be construed as limiting the scope of the invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions can occur, depending on design requirements and other factors. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (11)
1. A method for allocating virtual resources, comprising:
obtaining historical behavior data of a first user, wherein the historical behavior data comprises: a characteristic of the first user with respect to the virtual resource;
obtaining behavior preference of the first user about the virtual resources according to the historical behavior data by utilizing a first classification model; the first classification model is obtained by training according to historical behavior data of a plurality of second users;
predicting the probability of the first user using the virtual resources to be distributed by using a prediction model according to the behavior preference and the attributes of the virtual resources to be distributed, wherein the prediction model is obtained by training according to historical behavior data of a plurality of second users using the virtual resources and the attributes of the virtual resources used by the second users;
and when the probability is greater than a preset threshold value, allocating the virtual resources to be allocated to the first user.
2. The method of claim 1, wherein the attributes of the virtual resource to be allocated comprise: provider information of the virtual resources to be allocated;
obtaining behavior preference of the first user about a resource provider according to the historical behavior data by using a first classification model;
and according to the behavior preference and the provider information of the virtual resources to be allocated, predicting the probability of the first user using the virtual resources to be allocated provided by the resource provider by using a prediction model.
3. The method of claim 2, wherein the attributes of the virtual resource to be allocated further comprise: a value of the virtual resource to be allocated;
and predicting the probability of the first user using the virtual resource to be allocated provided by the resource provider by utilizing a prediction model according to the behavior preference of the first user about the resource provider and the value of the virtual resource to be allocated.
4. The method of claim 1,
obtaining historical transaction data of a resource provider, wherein the historical transaction data comprises: the corresponding resource provider provides the characteristics of the virtual resource;
respectively determining attribute preference of the resource provider about the virtual resources according to the historical transaction data by utilizing a second classification model;
obtaining behavior preference of the first user about a resource provider according to the historical behavior data by using a first classification model;
and predicting the probability of the first user using the virtual resource to be allocated provided by the resource provider by utilizing a prediction model according to the attribute preference, the behavior preference of the first user about the resource provider and the attribute of the virtual resource to be allocated.
5. The method of claim 1, further comprising:
determining a plurality of original characteristics of the historical behavior data of the first user, and inputting the plurality of original characteristics into a characteristic selector to obtain the characteristics of the first user about the virtual resource.
6. The method of claim 2, wherein the predictive model is trained from a random forest algorithm and any three of a GBDT algorithm, a Catboost algorithm, a LightGBM algorithm, an XGBoost algorithm, and an LR algorithm.
7. The method of claim 6,
respectively constructing three initial models by using any three of a GBDT algorithm, a Catboost algorithm, a LightGBM algorithm, an XGboost algorithm and an LR algorithm;
inputting historical behavior data of the second user into the three initial models respectively so as to train the three initial models;
and taking the output of the three initial models after training and the behavior preference of the second user on the virtual resources output by the first classification model as the input of the random forest algorithm so as to train the prediction model.
8. The method of any of claims 1 to 4, comprising:
the first classification model and the second classification model are obtained based on clustering algorithm training;
and/or the presence of a gas in the gas,
the feature selector is derived based on the feature _ selection library.
9. An apparatus for allocating virtual resources, comprising: the device comprises a data acquisition module, a classification module, a prediction module and an allocation module; wherein the content of the first and second substances,
the data acquisition module is configured to acquire historical behavior data of a first user, where the historical behavior data includes: a characteristic of the first user with respect to the virtual resource;
the classification module is used for obtaining behavior preference of the first user about the virtual resources according to the historical behavior data acquired by the data acquisition module by using a first classification model; the first classification model is obtained by training according to historical behavior data of a plurality of second users;
the prediction module is used for predicting the probability of the first user using the virtual resources to be allocated by using a prediction model according to the behavior preference and the attributes of the virtual resources to be allocated, which are obtained by the classification module, wherein the prediction model is obtained by training according to historical behavior data of a plurality of second users using the virtual resources and the attributes of the virtual resources used by the second users;
the allocation module is configured to allocate the virtual resource to be allocated to the first user when the probability predicted by the prediction module is greater than a preset threshold.
10. A virtual resource allocation server, comprising:
one or more processors;
a storage device for storing one or more programs,
when executed by the one or more processors, cause the one or more processors to implement the method of any one of claims 1-8.
11. A computer-readable medium, on which a computer program is stored, which, when being executed by a processor, carries out the method according to any one of claims 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011003976.6A CN113760521A (en) | 2020-09-22 | 2020-09-22 | Virtual resource allocation method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011003976.6A CN113760521A (en) | 2020-09-22 | 2020-09-22 | Virtual resource allocation method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113760521A true CN113760521A (en) | 2021-12-07 |
Family
ID=78785839
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011003976.6A Pending CN113760521A (en) | 2020-09-22 | 2020-09-22 | Virtual resource allocation method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113760521A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115526438A (en) * | 2022-11-28 | 2022-12-27 | 中国西安卫星测控中心 | Virtual resource pool expansion generation method based on ensemble learning model |
| CN116681454A (en) * | 2023-05-25 | 2023-09-01 | 北京阿帕科蓝科技有限公司 | Virtual resource proportioning strategy generation method and device, computer equipment and storage medium |
-
2020
- 2020-09-22 CN CN202011003976.6A patent/CN113760521A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115526438A (en) * | 2022-11-28 | 2022-12-27 | 中国西安卫星测控中心 | Virtual resource pool expansion generation method based on ensemble learning model |
| CN115526438B (en) * | 2022-11-28 | 2023-04-07 | 中国西安卫星测控中心 | Virtual resource pool expansion generation method based on ensemble learning model |
| CN116681454A (en) * | 2023-05-25 | 2023-09-01 | 北京阿帕科蓝科技有限公司 | Virtual resource proportioning strategy generation method and device, computer equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10504120B2 (en) | Determining a temporary transaction limit | |
| US10937060B2 (en) | Intelligent location based notification | |
| CN101013427B (en) | Method and system for managing data | |
| CN112148987B (en) | Message pushing method based on target object activity and related equipment | |
| US11281992B2 (en) | Predicting geofence performance for optimized location based services | |
| CN104077723B (en) | A kind of social networks commending system and method | |
| US10692045B2 (en) | Intelligent attention management for unified messaging | |
| WO2014107441A2 (en) | Social media impact assessment | |
| US10956470B2 (en) | Facet-based query refinement based on multiple query interpretations | |
| US20160232548A1 (en) | Adaptive pricing analytics | |
| US20130325863A1 (en) | Data Clustering for Multi-Layer Social Link Analysis | |
| CN112925859A (en) | Data storage method and device | |
| CN113760521A (en) | Virtual resource allocation method and device | |
| US10304081B1 (en) | Yielding content recommendations based on serving by probabilistic grade proportions | |
| CN110245684A (en) | Data processing method, electronic equipment and medium | |
| US20130325866A1 (en) | Community Profiling for Social Media | |
| US20190147543A1 (en) | Composite account structure | |
| CN112287208B (en) | User portrait generation method, device, electronic equipment and storage medium | |
| CN111598360A (en) | Service policy determination method and device and electronic equipment | |
| US11294917B2 (en) | Data attribution using frequent pattern analysis | |
| CN109426998A (en) | Information-pushing method and device | |
| US10354313B2 (en) | Emphasizing communication based on past interaction related to promoted items | |
| CN114925275A (en) | Product recommendation method and device, computer equipment and storage medium | |
| US20220036006A1 (en) | Feature vector generation for probabalistic matching | |
| CN112348658A (en) | Resource allocation method and device and electronic equipment |
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 |