CN111626767B - Resource data issuing method, device and equipment - Google Patents
Resource data issuing method, device and equipment Download PDFInfo
- Publication number
- CN111626767B CN111626767B CN202010357056.8A CN202010357056A CN111626767B CN 111626767 B CN111626767 B CN 111626767B CN 202010357056 A CN202010357056 A CN 202010357056A CN 111626767 B CN111626767 B CN 111626767B
- Authority
- CN
- China
- Prior art keywords
- user
- entity object
- resource data
- entity
- data
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0207—Discounts or incentives, e.g. coupons or rebates
- G06Q30/0222—During e-commerce, i.e. online transactions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0207—Discounts or incentives, e.g. coupons or rebates
- G06Q30/0224—Discounts or incentives, e.g. coupons or rebates based on user history
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Abstract
The application discloses a method, a device and equipment for issuing resource data, which relate to the technical field of data processing and can consider interaction behaviors among a user, an entity and the user and the entity in the process of issuing the resource data, so that the accurate issuing of the resource data is realized and the return rate of the resource data is improved. The method comprises the following steps: acquiring characteristic parameters related to a user and an entity object; forming a plurality of dimension characteristics by the characteristic parameters related to the user and the entity object, and inputting the dimension characteristics into a pre-constructed conversion rate model to obtain a prediction function expression; according to the prediction function expression, determining the resource data provided by the corresponding entity object when the conversion rate of the resource data used by the user is highest as first resource data which is required to be issued to the user by the entity object; and issuing the first resource data to a user.
Description
Technical Field
The present application relates to the field of data processing technologies, and in particular, to a method, an apparatus, and a device for issuing resource data.
Background
In the mobile internet era, with the rapid development of electronic commerce technology, a network platform has become an important tool for people to consume daily. In order to achieve better interaction with the user, merchants often use some interaction parameters to lead the user to obtain certain resource data in advance, so that the user can use the obtained resource data to realize the order transaction of the object in the network platform, and thereby enjoy the order preference.
In the related art, in the process of issuing resource data, a merchant uses some operation rules to determine the value of the resource data, for example, a constant value is performed for the resource data based on the frequency of a user login platform as a rule, a user logged in for less than 2 times a month is a low-frequency user, a user logged in for more than 10 times a month is a high-frequency user, the low-frequency user sets high-value resource data, and the high-frequency user sets low-value resource data.
In carrying out the present invention, the inventors have found that the related art has at least the following problems:
in order to better participate in order transaction when the network platform distributes resource data, the interactivity between merchants and users is improved. Because the resource data issued by different users are set by utilizing the manually specified rules, the issuing mode of the resource data is too single, if the users do not pay attention to the product, even if the numerical value provided in the resource data is higher, the users cannot be promoted to purchase the product, and the users focusing on the object have certain unfairness and cannot accurately distribute the resource data, so that the return rate of the resource data is lower, and the interactivity between merchants and the users is reduced.
Disclosure of Invention
In view of this, the application provides a method, a device and equipment for issuing resource data, and mainly aims to solve the problem that in the prior art, the issuing mode of the resource data is too single and the resource data cannot be accurately issued.
According to an aspect of the present application, there is provided a resource data issuing method including:
acquiring characteristic parameters related to a user and an entity object;
the characteristic parameters related to the user and the entity object form a plurality of dimension characteristics and are input into a pre-constructed conversion rate model to obtain a prediction function expression, wherein the conversion rate model outputs the prediction function expression based on the input plurality of dimension characteristics, and the prediction function expression is used for describing the relation between the conversion rate of the user using the resource data and the change of the resource data provided by the entity object;
according to the prediction function expression, determining the resource data provided by the corresponding entity object when the conversion rate of the resource data used by the user is highest as first resource data which is required to be issued to the user by the entity object;
and issuing the first resource data to a user.
Further, the feature parameters related to the user and the entity object include a user feature parameter, an entity object feature parameter, and a feature parameter of the user interacting with the entity object, and the obtaining the feature parameter related to the user and the entity object specifically includes:
Analyzing user behavior data by using a data model trained by user tags, and obtaining user characteristic parameters;
analyzing a browsing sequence of a user on the entity object within a first preset time by utilizing a mapping model trained by a node sequence of the entity object, and obtaining characteristic parameters of the entity object;
and acquiring characteristic parameters of interaction between the user and the entity object by counting access data of the user to the entity object in a second preset time after the entity object provides the resource data.
Further, the analyzing, by using the mapping model trained by the node sequence of the entity object, the browsing sequence of the user on the entity object within the first preset time, and obtaining the feature parameter of the entity object specifically includes:
mapping the browsing sequence of the entity object by the user within a first preset time into a distributed expression of the entity object in space by using a mapping model trained by the node sequence of the entity object;
and calculating the connection information between the entity objects according to the distributed expression of the entity objects in space, and acquiring the characteristic parameters of the entity objects.
Further, the obtaining the characteristic parameters of the interaction between the user and the entity object by counting the interaction data of the user on the entity object in the second preset time after the entity object provides the resource data specifically includes:
Collecting behavior data related to user operation in the entity object data by burying the data of the entity object after the entity object provides the resource data;
based on the behavior data related to user operation in the entity object data, statistics is carried out on the interaction data of the user on the entity object in a second preset time, and feature parameters of interaction between the user and the entity object are obtained.
Further, a plurality of decision trees are built in the conversion rate model, each decision tree is used for predicting the relation between the conversion rate of the resource data used by the user and the change of the resource data provided by the entity object from different dimension characteristics, the characteristic parameters related to the user and the entity object form a plurality of dimension characteristics and are input into a pre-built conversion rate model to obtain a prediction function expression, and the method specifically comprises the following steps:
utilizing a pre-constructed conversion rate model to form a plurality of dimension characteristics of characteristic parameters related to the user and the entity object, and distributing the dimension characteristics to each leaf node in the decision tree, wherein each leaf node corresponds to a parameter with characteristic mapping;
and predicting the decision tree of the optimal structure according to the parameters of the feature mapping of each leaf node in the decision tree and the leaf node, and taking the decision tree of the optimal structure as a prediction function expression.
Further, the method predicts the decision tree of the optimal structure according to the parameters of the feature mapping between each leaf node and the leaf node in the decision tree, and takes the decision tree of the optimal structure as a prediction function expression, and specifically comprises the following steps:
sequencing all leaf nodes in the decision tree from large to small according to the gradient data, and listing segmentation points for segmenting the decision tree;
scoring the segmentation points for segmenting the decision tree according to the parameters of the feature mapping corresponding to each leaf node in the decision tree;
and taking the decision tree formed by the highest scoring point as the decision tree of the optimal structure to obtain the predictive function expression.
Further, after determining that the entity object should issue the target resource data to the user according to the prediction function expression of the resource data provided to the entity object by the user, the method further includes:
respectively aggregating users containing the same attribute characteristics and aggregating the entity objects containing the same attribute characteristics by utilizing the characteristic parameters related to the users and the entity objects to obtain a target user group and a target entity object group;
determining second resource data which is to be issued to a target user group by the target entity object group according to the first resource data which is to be issued to the user by the entity object;
The issuing the target resource data to the user specifically comprises the following steps:
and issuing the second resource data to the target user group.
According to another aspect of the present application, there is provided a resource data issuing apparatus including:
the acquisition unit is used for acquiring characteristic parameters related to the user and the entity object;
the prediction unit is used for inputting the characteristic parameters related to the user and the entity object into a pre-constructed conversion rate model to obtain a prediction function expression, wherein the conversion rate model outputs the prediction function expression based on the input multiple dimensional characteristics, and the prediction function expression is used for describing the relation between the conversion rate of the resource data used by the user and the change of the resource data provided by the entity object;
the first determining unit is used for determining the resource data provided by the corresponding entity object when the conversion rate of the resource data used by the user is highest as first resource data which should be issued to the user by the entity object according to the prediction function expression;
and the issuing unit is used for issuing the first resource data to the user.
Further, the feature parameters related to the user and the entity object comprise a user feature parameter, an entity object feature parameter and a feature parameter of interaction between the user and the entity object, and the acquisition unit comprises
The first acquisition module is used for analyzing the user behavior data by utilizing a data model trained by the user tag and acquiring the user characteristic parameters;
the second acquisition module is used for analyzing the browsing sequence of the user on the entity object in the first preset time by utilizing the mapping model trained by the node sequence of the entity object to acquire the characteristic parameters of the entity object;
and the third acquisition module is used for acquiring the characteristic parameters of interaction between the user and the entity object by counting the access data of the user to the entity object in a second preset time after the entity object provides the resource data.
Further, the second acquisition module includes:
the mapping sub-module is used for mapping the browsing sequence of the entity object by the user within a first preset time into the spatial distributed expression of the entity object by using a mapping model trained by the node sequence of the entity object;
and the calculating sub-module is used for calculating the connection information between the entity objects according to the distributed expression of the entity objects in space and obtaining the characteristic parameters of the entity objects.
Further, the third acquisition module includes:
the collecting sub-module is used for collecting behavior data related to user operation in the entity object data by burying the data of the entity object after the entity object provides the resource data;
And the statistics sub-module is used for counting the interaction data of the user on the entity object in the second preset time based on the behavior data related to the user operation in the entity object data, and acquiring the characteristic parameters of the interaction between the user and the entity object.
Further, a plurality of decision trees are built in the conversion rate model, each decision tree is used for predicting the relation between the conversion rate of the user using the resource data and the change of the resource data provided by the entity object from different dimension characteristics, and the prediction unit comprises:
the distribution module is used for forming a plurality of dimension characteristics by utilizing a pre-constructed conversion rate model and the characteristic parameters related to the user and the entity object, and distributing the dimension characteristics to each leaf node in the decision tree, wherein each leaf node corresponds to the parameter with characteristic mapping;
and the prediction module is used for predicting the decision tree of the optimal structure according to the parameters of the feature mapping of each leaf node and the leaf node in the decision tree, and taking the decision tree of the optimal structure as a prediction function expression.
Further, the prediction module includes:
the sorting sub-module is used for sorting all leaf nodes in the decision tree from large to small according to the gradient data, and listing segmentation points for segmenting the decision tree;
The scoring module is used for scoring the segmentation points for segmenting the decision tree according to the parameters of the feature mapping corresponding to each leaf node in the decision tree;
and the prediction sub-module is used for taking a decision tree formed by the highest scoring point as a decision tree of an optimal structure to obtain a prediction function expression.
Further, the apparatus further comprises:
the aggregation unit is used for respectively aggregating the users with the same attribute characteristics and aggregating the entity objects with the same attribute characteristics by utilizing the characteristic parameters related to the users and the entity objects after determining the target resource data which is required to be issued to the users by the entity objects according to the prediction function expression of the resource data provided by the users to the entity objects, so as to obtain a target user group and a target entity object group;
the second determining unit is used for determining second resource data which is required to be issued to the target user group by the target entity object group according to the first resource data which is required to be issued to the user by the entity object;
the issuing unit is specifically further configured to issue the second resource data to the target user group.
According to still another aspect of the present application, there is provided a storage medium having stored thereon a computer program which, when executed by a processor, implements the above-described resource data issuing method.
According to still another aspect of the present application, there is provided a resource data issuing entity apparatus including a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, the processor implementing the above resource data issuing method when executing the program.
By means of the technical scheme, compared with the existing resource data distribution mode in which the numerical value of the resource data is determined by manual experience, the resource data distribution method, device and equipment provided by the application take the characteristic parameters related to the user and the entity object into consideration, form a plurality of dimension characteristics and input the plurality of dimension characteristics to a pre-constructed conversion rate model, output a prediction function expression of the user for providing the resource data for the entity object based on the plurality of dimension characteristics by using the conversion rate model, wherein the prediction function expression is used for describing the relation between the conversion rate of the user using the resource data and the change of the resource data provided by the entity object, further provide the resource data corresponding to the entity object when the conversion rate of the user using the resource data is maximum as the resource data correspondingly issued by the user, enable the user to participate in the interaction of the resource data more willingly, realize accurate distribution of the resource data, and improve the return rate of the resource data.
The foregoing description is only an overview of the present application, and is intended to be implemented in accordance with the teachings of the present application in order that the same may be more clearly understood and to make the same and other objects, features and advantages of the present application more readily apparent.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain the application and do not constitute a limitation on the application. In the drawings:
fig. 1 is a schematic flow chart of a method for issuing resource data according to an embodiment of the present application;
FIG. 2 is a schematic flow chart of another method for issuing resource data according to an embodiment of the present application;
fig. 3 is a schematic structural diagram of a resource data distributing device according to an embodiment of the present application;
fig. 4 is a schematic structural diagram of another resource data distributing device according to an embodiment of the present application.
Detailed Description
The application will be described in detail hereinafter with reference to the drawings in conjunction with embodiments. It should be noted that, without conflict, the embodiments of the present application and features of the embodiments may be combined with each other.
At present, by the existing resource data issuing mode, high-frequency users can be mostly subsidized with the numerical value of some resource data, and low-frequency users can be mostly subsidized with the numerical value of the resource data, however, the essence of resource data issuing is to promote users to consume the numerical value of the resource data, and because the numerical value pricing of the resource data issued by different users is defined by manual experience, the numerical value pricing of the resource data is unreasonable, the resource data cannot be accurately distributed theoretically, and the return rate of the resource data is lower.
In order to solve the problem, this embodiment provides a method for issuing resource data, as shown in fig. 1, including:
101. and acquiring characteristic parameters related to the user and the entity object.
The feature parameters related to the user and the entity object may include a user feature parameter, an entity object feature parameter, and a parameter of interaction between the user and the entity object, where the entity object is an online merchant.
In the embodiment of the invention, the feature data related to the user and the entity object can be specifically obtained by collecting the user behavior data set in the application log, and the user behavior data set contains various behaviors of the user in the process of accessing the entity object, such as login, browsing, ordering and the like, so that the user behavior data set can be utilized to extract the user feature parameters, the entity object feature parameters and the parameters of interaction between the user and the entity object.
The user characteristic parameters are characteristic information describing the user in each dimension, including but not limited to sensitivity of the user to values in the resource data, user portraits, conditions of the user accessing the app, and the like, specifically, a tagged user model may be constructed according to information such as user characteristics, service scene user behaviors, and the like, further, the characteristic information of the user in each dimension is predicted according to the user model data, the user portraits may include gender, age, occupation, consumption condition of the resource providing platform, age stage of the user, and the like, specifically, the user logging into account information of the resource providing platform may be determined, the sensitivity degree of the user to values in the resource data may include use condition of the user for the resource data to be provided to the account, specifically, the conditions of the user accessing the app may include probability of the user logging into the app, probability of the user performing operation on the product, and the like, specifically, and the user may be determined by collecting behavior logs of the user in the app.
The above-mentioned feature parameters of the physical objects are feature information describing the similarity between the physical objects, and may be represented by vectors, specifically, the similarity between the physical objects may be predicted according to a browsing sequence of the user to the physical objects within a preset time, for example, the obtaining a browsing sequence of the user to the physical objects within 30 minutes is: the method comprises the steps of A, B and C entity objects, converting the correlation of different entity objects into points on space for description, calculating the vector of the entity object A and the vector of the entity object B, wherein the closer the spatial distance between the two vectors is, the higher the similarity between the entity object A and the entity object B is.
The interaction parameters between the user and the entity object are used for describing the access condition of the user to the entity object, and specifically, the numerical value in the resource data can be provided by the entity object through statistics, and the behavior characteristics of the product provided by the user to the entity object in the preset time, such as access, re-purchase behavior, purchase behavior and the like, can be provided.
It can be understood that when the entity object issues the resource data by using the resource delivery platform, only dimensional characteristics such as frequency of using the resource data by the user may be considered, but dimensional characteristics related to the entity object are not considered, for example, the use of the resource data by the user may increase access probability of the user, and whether the user uses the resource data has referenceability, and the access condition of the user to the entity object may indicate the attention degree of the user to the entity object, and also whether the user uses the resource data has referenceability, so that by acquiring characteristic parameters related to the user and the entity object is equivalent to preparation work for predicting the conversion rate of using the resource data by the user subsequently, the conversion rate of using the resource data obtained by prediction is used as a basis for numerical pricing for issuing the resource data.
102. And forming a plurality of dimension characteristics by the characteristic parameters related to the user and the entity object, and inputting the dimension characteristics into a pre-constructed conversion rate model to obtain a prediction function expression.
In the embodiment of the invention, the pre-constructed conversion rate model outputs the prediction function expression based on a plurality of input dimensional characteristics, the prediction function expression is used for describing the relation between the conversion rate of the user using the resource data and the change of the resource data provided by the entity object, and the obtained plurality of dimensional characteristics are different for different users and entity objects, so that the prediction function expression suitable for the users and the entity object is also different.
It may be understood that the conversion rate model may output a prediction function expression, so as to predict a relationship between a conversion rate of using resource data by a user and a change of providing resource data by an entity object, and specifically, the conversion rate model may be constructed by using a feature parameter related to the user and the entity object as a training sample based on whether the user uses a tag of providing resource data by the entity object or not, and continuously iterating the input training sample through a preset learning model.
In the embodiment of the present invention, the selection of the preset learning model may be a supervision model with learning effects, including, but not limited to, a boost model, an xgboost model, and the like, which is not limited herein.
103. And according to the prediction function expression, determining the resource data provided by the corresponding entity object when the conversion rate of the resource data used by the user is highest as the first resource data which the entity object should issue to the user.
In the embodiment of the invention, the conversion rate of the resource data used by the user can be predicted according to the numerical range change of the resource data provided by the entity object by the prediction function expression, so that the higher the conversion rate of the resource data used by the user, the higher the probability of purchasing a product by the user using the resource data is, and the higher the resource data provided by the non-entity object is, the higher the conversion rate of the resource data used by the user can be, and the resource data provided by the corresponding entity object when the conversion rate of the resource data used by the user is the highest can be selected as the first resource data which is required to be issued to the user by the entity object.
104. And issuing the first resource data to a user.
The first resource data corresponds to a numerical subsidy issued to the user by the entity object through the resource issuing platform, so that the user can use the first resource data to avoid corresponding numerical values when consuming, and the first resource data issued to the user can include, but is not limited to, forms of red packets, coupons, full coupons and the like.
It can be understood that, the entity object does not directly issue the resource data to the user, in order to accurately issue the resource data suitable for interaction to the user, the resource issuing platform is utilized to calculate the resource data suitable for pushing to the user, so the execution subject of the embodiment of the application can be a resource data issuing device, which can be generally configured at the resource issuing platform side, and the resource data of interest can be rapidly determined for the user through the resource data issuing platform according to the resource data determined by the interaction characteristics of the user and the entity object, so that the user can more effectively use the resource data, and the interactivity between the entity object and the user can be further improved.
Compared with the existing resource data distribution mode in which the numerical value of the resource data is determined by manual experience, the resource data distribution method provided by the embodiment of the application takes the characteristic parameters related to the user and the entity object into consideration, the characteristic parameters related to the user and the entity object form a plurality of dimension characteristics and input the plurality of dimension characteristics into the pre-constructed conversion rate model, the conversion rate model is utilized to output a prediction function expression of the resource data provided by the user for the entity object based on the plurality of input dimension characteristics, the prediction function expression is used for describing the relation between the conversion rate of the resource data used by the user and the change of the resource data provided by the entity object, and further the resource data provided by the entity object corresponding to the user when the conversion rate of the resource data used by the user is the maximum is used as the resource data correspondingly distributed by the user, so that the user obtains the resource data with the optimal numerical value, the user has willingness to participate in the interaction of the resource data, the accurate distribution of the resource data is realized, and the return rate of the resource data is improved.
Further, as a refinement and extension of the foregoing embodiment, in order to fully describe a specific implementation procedure of the embodiment, this embodiment provides another resource data issuing method, as shown in fig. 2, where the method includes:
201. and acquiring characteristic parameters related to the user and the entity object.
In the embodiment of the invention, the feature parameters related to the user and the entity object can comprise the feature parameters of the user, the feature parameters of the entity object and the parameters of the interaction between the user and the entity object.
Aiming at the user characteristic parameters, the user behavior data can be analyzed by using a data model trained by the user tag to obtain the user characteristic parameters, the user behavior data can be mapped to a plurality of subspaces by using the data model trained by the user tag, the characteristic weight of the user behavior data on the user tag is captured, and the user characteristic parameters are predicted.
For the feature parameters of the entity object, the mapping model trained by the node sequence of the entity object can be utilized to analyze the browsing sequence of the entity object by the user within a first preset time to obtain the feature parameters of the entity object, and the mapping model trained by the node sequence of the entity object can be utilized to map the browsing sequence of the entity object by the user within the first preset time to be the distributed expression of the entity object in space, and further, the feature parameters of the entity object are obtained according to the connection information among the entity objects calculated by the distributed expression of the entity object in space. The browsing sequence of the entity object by the user in the first preset time can generate a node sequence, namely, the node sequence of the entity object browsed by the user, the mapping model can map the node sequence into points on the space, the vector formed by each point represents one entity object, and the closer the space distance of the vector formed by the entity object is, the higher the similarity between the entity objects is.
For the parameters of the interaction between the user and the entity object, the characteristic parameters of the interaction between the user and the entity object can be obtained by counting the access data of the user to the entity object in a second preset time after the entity object provides the resource data. Specifically, after the entity object provides the resource data, the entity object is subjected to data embedding, behavior data related to user operation in the entity object data is collected, and further based on the behavior data related to user operation in the entity object data, interactive data of the entity object by a user in a second preset time is counted, so that characteristic parameters of interaction between the user and the entity object are obtained.
202. And forming a plurality of dimension characteristics by utilizing the pre-constructed conversion rate model and the characteristic parameters related to the user and the entity object, and distributing the dimension characteristics to each leaf node in the decision tree.
Specifically, in the process of constructing the conversion rate model, the preset learning model can be used for continuously adding decision trees, feature splitting is continuously performed to grow decision trees, one decision tree is added each time, a new prediction function expression is learned, the residual error of the last prediction is fitted, when training is completed, a plurality of decision trees are formed by a user, each decision tree is used for predicting the relation between conversion rate of using resource data by the user and change of the resource data provided by the entity object from different dimensional features, so that the characteristic parameters related to the user and the entity object are input into the pre-constructed conversion rate model, the plurality of dimensional features are required to be distributed to each leaf node in the decision tree, and each leaf node corresponds to the feature mapping parameter.
203. And predicting the decision tree of the optimal structure according to the parameters of the feature mapping of each leaf node in the decision tree and the leaf node, and taking the decision tree of the optimal structure as a prediction function expression.
In the embodiment of the invention, the leaf nodes in the decision tree can be sequenced from large to small according to the gradient data, and the segmentation points for segmenting the decision tree are listed; scoring segmentation points for segmenting the decision tree according to parameters of feature mapping corresponding to each leaf node in the decision tree; and taking the decision tree formed by the highest scoring point as the decision tree of the optimal structure to obtain the predictive function expression. The prediction function expression can predict the ordering probability of the user in the numerical range of the resource data given by the entity object, for example, the numerical range of the resource data provided by the entity object is 3-7 yuan, and then the ordering probability of the user using the resource data can be predicted in the numerical range.
Specifically, because the learning algorithms adopted by the constructed conversion rate models are different, the prediction function expressions output based on the conversion rate models may be different, for example, the prediction function expressions may be smooth curves with conversion rate increased along with the increase of the numerical value of the resource data provided by the entity object, saw-tooth curves with conversion rate not monotonous along with the increase of the numerical value of the resource data provided by the entity object, curves with conversion rate increased and then decreased along with the increase of the numerical value of the resource data provided by the entity object, and the like, and the prediction function expressions formed by the user along with the change of the resource data provided by the entity object are obtained based on a plurality of dimension features formed by the characteristic parameters related to the input user and the entity object.
204. And according to the prediction function expression, determining the resource data provided by the corresponding entity object when the conversion rate of the resource data used by the user is highest as the first resource data which the entity object should issue to the user.
In order to save the resource cost of the entity object, in general, the entity object will set an upper limit value for the numerical value of the resource data in advance, if the numerical value of the first resource data which is determined to be issued to the user by the entity object exceeds the upper limit value set by the entity object according to the predictive function expression of the resource data provided by the user to the entity object, it is indicated that the entity object cannot provide full-scale resource data for the user, and it is necessary to provide differential resource data for the user by means of the resource issuing platform, so that the full-scale first resource data is formed by combining the entity object and the resource data provided by the resource issuing platform; if the value of the first resource data which is determined to be issued to the user by the entity object does not exceed the upper limit value set by the entity object according to the predictive function expression for providing the resource data to the entity object by the user, the entity object can issue full-scale first resource data to the user without providing any resource data by means of a resource issuing platform. If the full-scale resource data can not be provided for the user by means of the resource providing platform, the numerical value of the resource data provided by the corresponding entity object when the conversion rate of the resource data used by the user is highest can be found out as the first resource data which should be provided for the user by the entity object in the numerical range of the resource data expected to be provided for the entity object. For example, the numerical range of the resource data provided by the entity object is 3-7 yuan, and the numerical value of the resource data provided by the corresponding entity object when the conversion rate of the resource data is highest by the user is 15 yuan by utilizing the predictive function expression output by the conversion rate model, and the numerical value of the redundant resource data can not be provided by the resource issuing platform for 8 yuan, so that the numerical range of the resource data provided by the entity object is provided, the numerical value of the resource data provided by the corresponding entity object when the conversion rate of the resource data is highest by the user is 6 yuan, and the resource data is used as the first resource data which should be issued to the user by the entity object.
205. And respectively aggregating the users containing the same attribute characteristics and the entity objects containing the same attribute characteristics by utilizing the characteristic parameters related to the users and the entity objects to obtain a target user group and a target entity object group.
It will be appreciated that the first resource data that the entity object determined according to the expression of the prediction function should issue to the user, there is a personalized resource data value for each entity object for each user, and a large amount of data support is required in actually performing the resource data issue process. In order to save the resource occupation, similar users and similar entity objects can be aggregated, so that a target user group containing the same attribute features and a target entity object group containing the same attribute features are obtained.
For example, users A1-A5 may contain the same attribute features, users A1-A5 may belong to a similar user population, and entity objects B1-B5 may contain the same attribute features, and entity objects B1-B5 may belong to a similar entity object population.
206. And determining second resource data which is required to be issued to the target user group by the target entity object group according to the first resource data which is required to be issued to the user by the entity object.
Because each user can determine a resource data value corresponding to each entity object, the resource data value which should be issued to the target user group for the target entity object group can be used as one sample by each entity object in the target entity object group, and the average value of a plurality of samples is calculated to be used as the second resource data which should be issued to the target user group by the target entity object group.
For example, if the users A1-A5 are target user groups and the users B1-B5 belong to target entity object groups, 25 samples are mapped to the entity objects B1-B5 for the users A1-A5, and an average value of the 25 samples is obtained as second resource data to be issued to the target entity object groups.
207. And issuing the second resource data to the target user group.
It will be appreciated that in the process of issuing the second resource data to the user, in order to increase the conversion rate of the resource data used by the user, the issuing form of the resource data may be set, for example, the number of the resource data issued for the user with lower frequency of using the resource data is greater than the number of the resource data issued for the user with higher frequency of using the resource data, so as to promote the user with lower frequency of using the resource data to use.
Further, as a specific implementation of the methods of fig. 1 and fig. 2, an embodiment of the present application provides a device for issuing resource data, as shown in fig. 3, where the device includes: an acquisition unit 31, a prediction unit 32, a determination unit 33, and a distribution unit 34.
An obtaining unit 31, configured to obtain feature parameters related to the user and the entity object;
the prediction unit 32 may be configured to input the feature parameters related to the user and the entity object to a pre-constructed conversion rate model to obtain a prediction function expression, where the conversion rate model outputs the prediction function expression based on the input multiple dimensional features, and the prediction function expression is used to describe a relationship between the conversion rate of the user using the resource data and the change of the resource data provided by the entity object;
the first determining unit 33 may be configured to determine, according to the prediction function expression, resource data provided by the entity object corresponding to the highest conversion rate of the resource data used by the user as first resource data that the entity object should issue to the user;
the issuing unit 34 may be configured to issue the first resource data to a user.
Compared with the existing resource data distribution mode in which the numerical value of the resource data is determined by manual experience, the resource data distribution device provided by the embodiment of the application considers the characteristic parameters related to the user and the entity object, forms a plurality of dimension characteristics by the characteristic parameters related to the user and the entity object, inputs the dimension characteristics into the pre-constructed conversion rate model, utilizes the conversion rate model to output a prediction function expression of the resource data provided by the user for the entity object based on the input dimension characteristics, and the prediction function expression is used for describing the relation between the conversion rate of the resource data used by the user and the change of the resource data provided by the entity object, so that the corresponding entity object provides the resource data when the conversion rate of the resource data used by the user is the maximum as the resource data correspondingly distributed by the user, the user obtains the resource data with the optimal numerical value, and the user has a higher willingness to participate in the interaction of the resource data, thereby realizing accurate distribution of the resource data and improving the return rate of the resource data.
In a specific application scenario, as shown in fig. 4, the feature parameters related to the user and the entity object include a user feature parameter, an entity object feature parameter, and a feature parameter of the user interacting with the entity object, and the obtaining unit 31 includes:
the first obtaining module 311 may be configured to analyze user behavior data by using a data model trained by user tags, and obtain user feature parameters;
the second obtaining module 312 may be configured to analyze a browsing sequence of the user on the entity object within a first preset time by using a mapping model trained by a node sequence of the entity object, and obtain feature parameters of the entity object;
the third obtaining module 313 may be configured to obtain, after the entity object provides the resource data, a characteristic parameter of interaction between the user and the entity object by counting access data of the user to the entity object in a second preset time.
In a specific application scenario, as shown in fig. 4, the second obtaining module 312 includes:
mapping submodule 3121, configured to map a browsing sequence of the entity object by the user within a first preset time to a spatially distributed expression of the entity object by using a mapping model trained by a node sequence of the entity object;
The calculation submodule 3122 may be configured to calculate connection information between the entity objects according to the spatial distributed expression of the entity objects, and obtain feature parameters of the entity objects.
In a specific application scenario, as shown in fig. 4, the third obtaining module 313 includes:
the collecting submodule 3131 is used for collecting behavior data related to user operation in the entity object data by burying data into the entity object after the entity object provides resource data;
the statistics submodule 3132 may be configured to, based on the behavior data related to the user operation in the entity object data, count interaction data of the user on the entity object in a second preset time, and obtain a feature parameter of interaction between the user and the entity object.
In a specific application scenario, as shown in fig. 4, a plurality of decision trees are built in the conversion rate model, each decision tree is used for predicting a relationship between conversion rate of user usage resource data and change of entity object providing resource data from different dimension characteristics, and the prediction unit 32 includes:
the allocation module 321 is configured to allocate the feature parameters related to the user and the entity object to each leaf node in the decision tree by using a pre-constructed conversion rate model to form multiple dimension features, where each leaf node corresponds to a parameter with feature mapping;
The prediction module 322 may be configured to predict a decision tree with an optimal structure according to parameters mapped by each leaf node in the decision tree and the feature corresponding to the leaf node, and use the decision tree with the optimal structure as a prediction function expression.
In a specific application scenario, as shown in fig. 4, the prediction module 322 includes:
the sorting submodule 3221 can be used for sorting all leaf nodes in the decision tree from large to small according to gradient data, and listing segmentation points for segmenting the decision tree;
the scoring submodule 3222 is configured to score, for parameters of feature mapping corresponding to each leaf node in the decision tree, the segmentation points for segmenting the decision tree;
the prediction submodule 3223 may be configured to use a decision tree formed by the highest scoring point as a decision tree of an optimal structure to obtain a prediction function expression.
In a specific application scenario, as shown in fig. 4, the apparatus further includes:
the aggregation unit 35 may be configured to, after determining, according to the prediction function expression of the resource data provided by the user to the entity object, that the entity object should issue target resource data to the user, respectively aggregate users including the same attribute features and aggregate entity objects including the same attribute features by using feature parameters related to the user and the entity object, so as to obtain a target user group and a target entity object group;
A second determining unit 36, configured to determine, according to the first resource data that the entity object should be issued to the user, second resource data that the target entity object group should be issued to the target user group;
the issuing unit 34 may be further configured to issue the second resource data to the target user group.
It should be noted that, for other corresponding descriptions of each functional unit related to the resource data issuing apparatus provided in this embodiment, reference may be made to corresponding descriptions in fig. 1 and fig. 2, and details are not repeated here.
Based on the above-described method shown in fig. 1 and 2, correspondingly, the embodiment of the present application further provides a storage medium, on which a computer program is stored, which when executed by a processor, implements the above-described resource data distribution method shown in fig. 1 and 2.
Based on such understanding, the technical solution of the present application may be embodied in the form of a software product, which may be stored in a non-volatile storage medium (may be a CD-ROM, a U-disk, a mobile hard disk, etc.), and includes several instructions for causing a computer device (may be a personal computer, a server, or a network device, etc.) to execute the method described in the respective implementation scenario of the present application.
Based on the methods shown in fig. 1 and fig. 2 and the virtual device embodiments shown in fig. 3 and fig. 4, in order to achieve the above objects, the embodiments of the present application further provide an entity device for issuing resource data, which may specifically be a computer, a smart phone, a tablet computer, a smart watch, a server, or a network device, where the entity device includes a storage medium and a processor; a storage medium storing a computer program; a processor for executing the computer program to implement the above-described resource data issuing method as shown in fig. 1 and 2.
Optionally, the physical device may further include a user interface, a network interface, a camera, radio Frequency (RF) circuitry, sensors, audio circuitry, WI-FI modules, and the like. The user interface may include a Display screen (Display), an input unit such as a Keyboard (Keyboard), etc., and the optional user interface may also include a USB interface, a card reader interface, etc. The network interface may optionally include a standard wired interface, a wireless interface (e.g., WI-FI interface), etc.
It will be appreciated by those skilled in the art that the entity device structure for issuing resource data provided in this embodiment is not limited to the entity device, and may include more or fewer components, or may combine some components, or may be a different arrangement of components.
The storage medium may also include an operating system, a network communication module. The operating system is a program that manages the physical device hardware and software resources of the store search information processing described above, supporting the execution of information processing programs and other software and/or programs. The network communication module is used for realizing communication among all components in the storage medium and communication with other hardware and software in the information processing entity equipment.
From the above description of the embodiments, it will be apparent to those skilled in the art that the present application may be implemented by means of software plus necessary general hardware platforms, or may be implemented by hardware. By applying the technical scheme of the application, compared with the prior art, the method and the device have the advantages that the characteristic parameters related to the user and the entity object are considered, the relation between the conversion rate of the user using the resource data and the change of the resource data provided by the entity object is predicted, and then the entity object corresponding to the maximum conversion rate of the user using the resource data is provided with the resource data as the resource data correspondingly issued by the user, so that the user obtains the resource data with the optimal value, the accurate issuing of the resource data is realized, and the return rate of the resource data is improved.
Those skilled in the art will appreciate that the drawing is merely a schematic illustration of a preferred implementation scenario and that the modules or flows in the drawing are not necessarily required to practice the application. Those skilled in the art will appreciate that modules in an apparatus in an implementation scenario may be distributed in an apparatus in an implementation scenario according to an implementation scenario description, or that corresponding changes may be located in one or more apparatuses different from the implementation scenario. The modules of the implementation scenario may be combined into one module, or may be further split into a plurality of sub-modules.
The above-mentioned inventive sequence numbers are merely for description and do not represent advantages or disadvantages of the implementation scenario. The foregoing disclosure is merely illustrative of some embodiments of the application, and the application is not limited thereto, as modifications may be made by those skilled in the art without departing from the scope of the application.
Claims (14)
1. A method for issuing resource data, comprising:
the method comprises the steps that feature parameters related to a user and an entity object are obtained, the entity object is an online merchant, the feature parameters related to the user and the entity object comprise user feature parameters, entity object feature parameters and parameters for interaction between the user and the entity object, the user feature parameters are feature information describing the user in each dimension, the entity object feature parameters are feature information describing similarity between the entity objects, the interaction parameters between the user and the entity object are used for describing access conditions of the user to the entity object, and particularly, user behavior data are analyzed by utilizing a data model trained by user tags to obtain the user feature parameters; analyzing a browsing sequence of a user on the entity object within a first preset time by utilizing a mapping model trained by a node sequence of the entity object, and obtaining characteristic parameters of the entity object; the method comprises the steps that after resource data are provided by an entity object, interactive data of the entity object are counted by a user in a second preset time, and characteristic parameters of interaction between the user and the entity object are obtained;
The characteristic parameters related to the user and the entity object form a plurality of dimension characteristics and are input into a pre-constructed conversion rate model to obtain a prediction function expression, wherein the conversion rate model outputs the prediction function expression based on the input plurality of dimension characteristics, and the prediction function expression is used for describing the relation between the conversion rate of the user using the resource data and the change of the resource data provided by the entity object;
according to the prediction function expression, determining the resource data provided by the corresponding entity object when the conversion rate of the resource data used by the user is highest as first resource data which is required to be issued to the user by the entity object;
and issuing the first resource data to a user.
2. The method of claim 1, wherein the analyzing the browsing sequence of the user on the physical object in the first preset time by using the mapping model trained by the node sequence of the physical object, and obtaining the feature parameters of the physical object specifically comprises:
mapping the browsing sequence of the entity object by the user within a first preset time into a distributed expression of the entity object in space by using a mapping model trained by the node sequence of the entity object;
And calculating the connection information between the entity objects according to the distributed expression of the entity objects in space, and acquiring the characteristic parameters of the entity objects.
3. The method of claim 1, wherein the obtaining the feature parameters of the user interaction with the physical object by counting the interaction data of the user with the physical object in the second preset time after the physical object provides the resource data specifically comprises:
collecting behavior data related to user operation in the entity object data by burying the data of the entity object after the entity object provides the resource data;
based on the behavior data related to user operation in the entity object data, statistics is carried out on the interaction data of the user on the entity object in a second preset time, and feature parameters of interaction between the user and the entity object are obtained.
4. A method according to any one of claims 1-3, wherein a plurality of decision trees are built in the conversion rate model, each decision tree is used for predicting a relationship between conversion rate of user using resource data along with resource data change provided by an entity object from different dimension characteristics, the characteristic parameters related to the user and the entity object are formed into a plurality of dimension characteristics and are input into a pre-built conversion rate model, and a prediction function expression is obtained, and specifically includes:
Utilizing a pre-constructed conversion rate model to form a plurality of dimension characteristics of characteristic parameters related to the user and the entity object, and distributing the dimension characteristics to each leaf node in the decision tree, wherein each leaf node corresponds to a parameter with characteristic mapping;
and predicting the decision tree of the optimal structure according to the parameters of the feature mapping of each leaf node in the decision tree and the leaf node, and taking the decision tree of the optimal structure as a prediction function expression.
5. The method according to claim 4, wherein predicting the decision tree of the optimal structure according to the parameters of the feature mapping corresponding to each leaf node in the decision tree and the leaf node, and taking the decision tree of the optimal structure as the prediction function expression specifically includes:
sequencing all leaf nodes in the decision tree from large to small according to the gradient data, and listing segmentation points for segmenting the decision tree;
scoring the segmentation points for segmenting the decision tree according to the parameters of the feature mapping corresponding to each leaf node in the decision tree;
and taking the decision tree formed by the highest scoring point as the decision tree of the optimal structure to obtain the predictive function expression.
6. The method according to claim 1, wherein after determining, according to the expression of the prediction function, resource data provided by the corresponding entity object when the conversion rate of the resource data used by the user is the highest as the first resource data that the entity object should issue to the user, the method further comprises:
respectively aggregating users containing the same attribute characteristics and aggregating the entity objects containing the same attribute characteristics by utilizing the characteristic parameters related to the users and the entity objects to obtain a target user group and a target entity object group;
determining second resource data which is to be issued to a target user group by the target entity object group according to the first resource data which is to be issued to the user by the entity object;
and issuing the second resource data to the target user group.
7. A resource data issuing apparatus, comprising:
the acquisition unit is used for acquiring characteristic parameters related to the user and the entity object;
the prediction unit is used for forming a plurality of dimension characteristics by the characteristic parameters related to the user and the entity object, inputting the dimension characteristics into a pre-constructed conversion rate model to obtain a prediction function expression, wherein the conversion rate model outputs the prediction function expression based on the input dimension characteristics, and the prediction function expression is used for describing the relation between the conversion rate of the resource data used by the user and the change of the resource data provided by the entity object;
The first determining unit is used for determining the resource data provided by the corresponding entity object when the conversion rate of the resource data used by the user is highest as first resource data which should be issued to the user by the entity object according to the prediction function expression;
a release unit, configured to release the first resource data to a user;
the entity object is an online merchant, the feature parameters related to the user and the entity object include a user feature parameter, an entity object feature parameter, and a feature parameter for the user to interact with the entity object, the user feature parameter is feature information describing the user in each dimension, the entity object feature parameter is feature information describing the similarity between the entity objects, and the interaction parameter between the user and the entity object is used for describing the access condition of the user to the entity object, and the obtaining unit includes:
the first acquisition module is used for analyzing the user behavior data by utilizing a data model trained by the user tag and acquiring the user characteristic parameters;
the second acquisition module is used for analyzing the browsing sequence of the user on the entity object in the first preset time by utilizing the mapping model trained by the node sequence of the entity object to acquire the characteristic parameters of the entity object;
And the third acquisition module is used for acquiring characteristic parameters of interaction between the user and the entity object by counting interaction data of the user on the entity object in a second preset time after the entity object provides the resource data.
8. The apparatus of claim 7, wherein the second acquisition module comprises:
the mapping sub-module is used for mapping the browsing sequence of the entity object by the user within a first preset time into the spatial distributed expression of the entity object by using a mapping model trained by the node sequence of the entity object;
and the calculating sub-module is used for calculating the connection information between the entity objects according to the distributed expression of the entity objects in space and obtaining the characteristic parameters of the entity objects.
9. The apparatus of claim 7, wherein the third acquisition module comprises:
the collecting sub-module is used for collecting behavior data related to user operation in the entity object data by burying the data of the entity object after the entity object provides the resource data;
and the statistics sub-module is used for counting the interaction data of the user on the entity object in the second preset time based on the behavior data related to the user operation in the entity object data, and acquiring the characteristic parameters of the interaction between the user and the entity object.
10. The apparatus according to any one of claims 7-9, wherein a plurality of decision trees are built in the conversion model, each decision tree being configured to predict a relationship between conversion of user usage resource data from different dimension characteristics as a function of entity object provision resource data, the prediction unit comprising:
the distribution module is used for forming a plurality of dimension characteristics by utilizing a pre-constructed conversion rate model and the characteristic parameters related to the user and the entity object, and distributing the dimension characteristics to each leaf node in the decision tree, wherein each leaf node corresponds to the parameter with characteristic mapping;
and the prediction module is used for predicting the decision tree of the optimal structure according to the parameters of the feature mapping of each leaf node and the leaf node in the decision tree, and taking the decision tree of the optimal structure as a prediction function expression.
11. The apparatus of claim 10, wherein the prediction module comprises:
the sorting sub-module is used for sorting all leaf nodes in the decision tree from large to small according to the gradient data, and listing segmentation points for segmenting the decision tree;
the scoring module is used for scoring the segmentation points for segmenting the decision tree according to the parameters of the feature mapping corresponding to each leaf node in the decision tree;
And the prediction sub-module is used for taking a decision tree formed by the highest scoring point as a decision tree of an optimal structure to obtain a prediction function expression.
12. The apparatus of claim 7, wherein the apparatus further comprises:
the aggregation unit is used for respectively aggregating the users containing the same attribute characteristics and aggregating the entity objects containing the same attribute characteristics by utilizing the characteristic parameters related to the users and the entity objects after determining the resource data provided by the corresponding entity objects when the conversion rate of the resource data used by the users is highest as the first resource data which is required to be issued to the users by the entity objects according to the prediction function expression, so as to obtain a target user group and a target entity object group;
the second determining unit is used for determining second resource data which is required to be issued to the target user group by the target entity object group according to the first resource data which is required to be issued to the user by the entity object;
the issuing unit is specifically further configured to issue the second resource data to the target user group.
13. A storage medium having stored thereon a computer program, wherein the program when executed by a processor implements the resource data issuing method according to any one of claims 1 to 6.
14. A resource data distribution apparatus comprising a storage medium, a processor and a computer program stored on the storage medium and executable on the processor, characterized in that the processor implements the resource data distribution method according to any one of claims 1 to 6 when executing the program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010357056.8A CN111626767B (en) | 2020-04-29 | 2020-04-29 | Resource data issuing method, device and equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010357056.8A CN111626767B (en) | 2020-04-29 | 2020-04-29 | Resource data issuing method, device and equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111626767A CN111626767A (en) | 2020-09-04 |
| CN111626767B true CN111626767B (en) | 2023-09-08 |
Family
ID=72273220
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010357056.8A Active CN111626767B (en) | 2020-04-29 | 2020-04-29 | Resource data issuing method, device and equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111626767B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112215509A (en) * | 2020-10-21 | 2021-01-12 | 拉扎斯网络科技（上海）有限公司 | Resource parameter determination method, device and equipment |
| CN112422711B (en) * | 2020-11-06 | 2021-10-08 | 北京五八信息技术有限公司 | Resource allocation method and device, electronic equipment and storage medium |
| CN112308635A (en) * | 2020-11-25 | 2021-02-02 | 拉扎斯网络科技（上海）有限公司 | Data processing method and device and resource providing method and device |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106203635A (en) * | 2016-06-29 | 2016-12-07 | 北京师范大学 | A kind of on-line study behavior puts into data collection and transmission and method |
| CN106878405A (en) * | 2017-01-25 | 2017-06-20 | 咪咕动漫有限公司 | A kind of method and device for adjusting push project |
| CN108665355A (en) * | 2018-05-18 | 2018-10-16 | 深圳壹账通智能科技有限公司 | Financial product recommends method, apparatus, equipment and computer storage media |
| CN109241451A (en) * | 2018-11-08 | 2019-01-18 | 北京点网聚科技有限公司 | A kind of content combined recommendation method, apparatus and readable storage medium storing program for executing |
| CN110111090A (en) * | 2019-03-26 | 2019-08-09 | 口口相传(北京)网络技术有限公司 | A kind of distribution method and device of electronics red packet |
| US10402723B1 (en) * | 2018-09-11 | 2019-09-03 | Cerebri AI Inc. | Multi-stage machine-learning models to control path-dependent processes |
| CN110968802A (en) * | 2019-12-04 | 2020-04-07 | 上海风秩科技有限公司 | User characteristic analysis method, analysis device and readable storage medium |
-
2020
- 2020-04-29 CN CN202010357056.8A patent/CN111626767B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106203635A (en) * | 2016-06-29 | 2016-12-07 | 北京师范大学 | A kind of on-line study behavior puts into data collection and transmission and method |
| CN106878405A (en) * | 2017-01-25 | 2017-06-20 | 咪咕动漫有限公司 | A kind of method and device for adjusting push project |
| CN108665355A (en) * | 2018-05-18 | 2018-10-16 | 深圳壹账通智能科技有限公司 | Financial product recommends method, apparatus, equipment and computer storage media |
| US10402723B1 (en) * | 2018-09-11 | 2019-09-03 | Cerebri AI Inc. | Multi-stage machine-learning models to control path-dependent processes |
| CN109241451A (en) * | 2018-11-08 | 2019-01-18 | 北京点网聚科技有限公司 | A kind of content combined recommendation method, apparatus and readable storage medium storing program for executing |
| CN110111090A (en) * | 2019-03-26 | 2019-08-09 | 口口相传(北京)网络技术有限公司 | A kind of distribution method and device of electronics red packet |
| CN110968802A (en) * | 2019-12-04 | 2020-04-07 | 上海风秩科技有限公司 | User characteristic analysis method, analysis device and readable storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 吴华意 ; 李锐 ; 周振 ; 蒋捷 ; 桂志鹏 ; .公共地图服务的群体用户访问行为时序特征模型及预测.武汉大学学报(信息科学版).2015,第40卷(第10期),第1279-1286页. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111626767A (en) | 2020-09-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109902708A (en) | A kind of recommended models training method and relevant apparatus | |
| CN111626767B (en) | Resource data issuing method, device and equipment | |
| CN109492772B (en) | Method and device for generating information | |
| CN109460514A (en) | Method and apparatus for pushed information | |
| CN107808314B (en) | User recommendation method and device | |
| CN107894998B (en) | Video recommendation method and device | |
| CN107944481A (en) | Method and apparatus for generating information | |
| CN108665148B (en) | Electronic resource quality evaluation method and device and storage medium | |
| CN110008397A (en) | A kind of recommended models training method and device | |
| CN116541610B (en) | Training method and device for recommendation model | |
| CN111611496A (en) | Product recommendation method and device | |
| CN112417294A (en) | Intelligent business recommendation method based on neural network mining model | |
| CN107886241A (en) | Resource analysis method, apparatus, medium and electronic equipment | |
| CN114417174B (en) | Content recommendation method, device, equipment and computer storage medium | |
| CN110555749B (en) | Credit behavior prediction method and device based on neural network | |
| CN112036987B (en) | Method and device for determining recommended commodity | |
| CN113569162A (en) | Data processing method, device, equipment and storage medium | |
| CN112449002A (en) | Method, device and equipment for pushing object to be pushed and storage medium | |
| WO2019242453A1 (en) | Information processing method and device, storage medium, and electronic device | |
| WO2022156589A1 (en) | Method and device for determining live broadcast click rate | |
| CN115238588A (en) | Graph data processing method, risk prediction model training method and device | |
| CN113792952A (en) | Method and apparatus for generating a model | |
| CN113762675A (en) | Information generation method, device, server, system and storage medium | |
| CN110992109A (en) | Real estate customer analysis method, device and storage medium based on association rule | |
| CN117217852B (en) | Behavior recognition-based purchase willingness prediction method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |