CN113269574A - Information distribution method, device, equipment and storage medium - Google Patents
Information distribution method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN113269574A CN113269574A CN202010092716.4A CN202010092716A CN113269574A CN 113269574 A CN113269574 A CN 113269574A CN 202010092716 A CN202010092716 A CN 202010092716A CN 113269574 A CN113269574 A CN 113269574A
- Authority
- CN
- China
- Prior art keywords
- value
- modification information
- information
- value modification
- target
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 57
- 238000009826 distribution Methods 0.000 title claims abstract description 44
- 238000003860 storage Methods 0.000 title claims abstract description 26
- 230000004048 modification Effects 0.000 claims abstract description 276
- 238000012986 modification Methods 0.000 claims abstract description 276
- 239000013598 vector Substances 0.000 claims abstract description 99
- 238000012417 linear regression Methods 0.000 claims abstract description 98
- 238000012549 training Methods 0.000 claims abstract description 37
- 238000012216 screening Methods 0.000 claims abstract description 32
- 238000004422 calculation algorithm Methods 0.000 claims description 11
- 230000009466 transformation Effects 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 5
- 230000001419 dependent effect Effects 0.000 claims description 4
- 230000000694 effects Effects 0.000 claims description 3
- 230000006399 behavior Effects 0.000 description 28
- 230000008569 process Effects 0.000 description 13
- 238000010606 normalization Methods 0.000 description 11
- 238000012545 processing Methods 0.000 description 10
- 238000004364 calculation method Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 6
- 230000003287 optical effect Effects 0.000 description 6
- 230000008901 benefit Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 230000000699 topical effect Effects 0.000 description 3
- 238000007418 data mining Methods 0.000 description 2
- 239000013307 optical fiber Substances 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 230000000644 propagated effect Effects 0.000 description 2
- 238000003491 array Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012821 model calculation Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000008707 rearrangement Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
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
- G06Q30/0213—Consumer transaction fees
-
- 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
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
Landscapes
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Marketing (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The embodiment of the invention discloses an information distribution method, an information distribution device, information distribution equipment and a storage medium. The method comprises the following steps: acquiring feature vectors corresponding to the value modification information to be screened respectively, wherein the feature vectors comprise sub-feature vectors corresponding to at least one target information feature dimension of the value modification information respectively, and the value modification information is used for modifying the value attribute value when the article is acquired; determining the total predicted sales value of each piece of value modification information to be screened based on each feature vector and a preset linear regression model, wherein the linear regression model is obtained by training a mixed regularization model of linear regression in advance; and screening target value modification information from the value modification information to be screened according to the total predicted sales value, and distributing the target value modification information to target users. Through the technical scheme, the value modification information is distributed in a personalized mode, and the exposure rate and the utilization rate of the value modification information are improved.
Description
Technical Field
Embodiments of the present invention relate to computer technologies, and in particular, to an information distribution method, apparatus, device, and storage medium.
Background
For e-commerce platforms, value modification information (e.g., coupons) may stimulate the conversion rate of orders placed for a particular good, increase the customer unit price of a user, direct the platform to place an order for a new user for the first time, recall the lost user back to the e-commerce platform again, etc. How to effectively issue coupons becomes one of the major problems facing e-commerce websites.
At present, the coupon circulation process in the e-commerce platform is as follows: coupon production-coupon display-coupon pickup-coupon use. Wherein, the coupon production is to produce various coupons through operation activity and cost accounting; the coupon display is performed through a page such as the coupon center 110 or the product detail page 120 shown in fig. 1; the coupon getting is that a user actively gets the coupon on a coupon center or a page such as a commodity detail page; coupon use is the selection of the appropriate coupon settlement when the user places an order.
In the process of implementing the invention, the inventor finds that at least the following problems exist in the prior art: because the user needs to actively find and receive the coupon in the existing coupon process, the coupon display page is fixed, so that the user is not easy to find the coupon, and the exposure rate and the receiving rate of the coupon are reduced; (2) various coupons are displayed in a mixed mode, for example, the coupons of items with low customer orders and the coupons of items with high customer orders are displayed in a mixed mode, so that the sensitivity of a user to the coupons is reduced, the coupons which the user is interested in are not easy to find, and the receiving rate and the using rate of the coupons are reduced.
Disclosure of Invention
The embodiment of the invention provides an information distribution method, an information distribution device, information distribution equipment and a storage medium, which are used for realizing personalized distribution of value modification information and improving the exposure rate and the utilization rate of the value modification information.
In a first aspect, an embodiment of the present invention provides an information distribution method, including:
acquiring feature vectors corresponding to the value modification information to be screened respectively, wherein the feature vectors comprise sub-feature vectors corresponding to at least one target information feature dimension of the value modification information respectively, and the value modification information is used for modifying the value attribute value of the article when the article is acquired;
determining the total predicted sales value of each piece of value modification information to be screened based on each feature vector and a preset linear regression model, wherein the linear regression model is obtained by training a linear regression mixed regularization model in advance;
and screening target value modification information from the value modification information to be screened according to the total predicted sales value, and distributing the target value modification information to target users.
In a second aspect, an embodiment of the present invention further provides an information distribution apparatus, where the apparatus includes:
the characteristic vector acquisition module is used for acquiring characteristic vectors corresponding to the value modification information to be screened respectively, wherein the characteristic vectors comprise sub characteristic vectors corresponding to at least one target information characteristic dimension of the value modification information respectively, and the value modification information is used for modifying a value attribute value when an article is acquired;
the total predicted sales value determining module is used for determining the total predicted sales value of each piece of value modification information to be screened based on each feature vector and a preset linear regression model, wherein the linear regression model is obtained by training a linear regression mixed regularization model in advance;
and the target value modification information distribution module is used for screening out target value modification information from the value modification information to be screened according to the total predicted sales value and distributing the target value modification information to target users.
In a third aspect, an embodiment of the present invention further provides an electronic device, where the electronic device includes:
one or more processors;
a storage device for storing one or more programs,
when the one or more programs are executed by the one or more processors, the one or more processors implement the information distribution method provided by any embodiment of the present invention.
In a fourth aspect, the embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the computer program implements the information distribution method provided in any embodiment of the present invention.
The embodiment of the invention obtains the linear regression model by training the linear regression mixed regularization model in advance, realizes more accurate determination of model parameters, constructs the linear regression model between the information characteristic dimensions of the total value of the forecast sales and the value modification information, and provides an accurate model basis for screening and distribution of the value modification information. The method comprises the steps that characteristic vectors corresponding to value modification information to be screened are obtained, wherein the characteristic vectors comprise sub characteristic vectors corresponding to at least one target information characteristic dimension of the value modification information, and the value modification information is used for modifying a value attribute value when an article is obtained; determining the total predicted sales value of each piece of value modification information to be screened based on each feature vector and a preset linear regression model, wherein the linear regression model is obtained by training a mixed regularization model of linear regression in advance; and screening target value modification information from the value modification information to be screened according to the total predicted sales value, and distributing the target value modification information to target users. The method and the device realize the calculation of the predicted total sales value of each piece of value modification information to be screened by using the preset linear regression model with high calculation accuracy and high speed, further screen out the target value modification information for the target user according to each predicted total sales value and distribute the target value modification information to the target user, avoid the process that the target user actively finds and obtains the target value modification information, and further improve the exposure rate and the utilization rate of the modified value information.
Drawings
FIG. 1 is a schematic illustration of a presentation page of value modification information (e.g., coupons) in the prior art;
fig. 2 is a flowchart of an information distribution method in a first embodiment of the present invention;
FIG. 3 is a radar chart of feature vectors of value modification information to be filtered, which are formed by sub-feature vectors of feature dimensions of target information in the first embodiment of the present invention;
FIG. 4 is a diagram illustrating a coupon circulation flow according to a first embodiment of the invention;
fig. 5 is a flowchart of an information distribution method in the second embodiment of the present invention;
fig. 6 is a schematic structural diagram of an information distribution apparatus in a third embodiment of the present invention;
fig. 7 is a schematic structural diagram of an electronic device in a fourth embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It should be further noted that, for the convenience of description, only some of the structures related to the present invention are shown in the drawings, not all of the structures.
Example one
The information distribution method provided by the embodiment can be suitable for issuing value modification information (such as coupons) of articles (such as commodities) in electronic commerce. The method may be performed by an information distribution apparatus, which may be implemented by software and/or hardware, and may be integrated in an electronic device, such as a laptop, a desktop, a server, or the like. Referring to fig. 2, the method of the present embodiment specifically includes the following steps:
s110, obtaining feature vectors corresponding to the value modification information to be screened respectively, wherein the feature vectors comprise sub feature vectors corresponding to at least one target information feature dimension of the value modification information respectively, and the value modification information is used for modifying the value attribute value when the article is obtained.
The value modification information is information for modifying the value attribute value (for example, the selling price of a commodity) when the item is acquired, and includes modification conditions (for example, an applicable item and modification timing) to be satisfied when the value attribute value is modified, a modification width of the value attribute value (that is, a value attribute value modification value), and the like. The value modification information to be screened refers to the value modification information participating in information screening, and may be all or part of the value modification information generated in the e-commerce platform (such as generated coupons). The feature vector is a digital representation of the value modification information to be screened, and is composed of sub-feature vectors of feature dimensions of each target information of the value modification information. The sub-feature vector refers to a digitized representation of a feature dimension of a certain target information, and may be a single numerical value or a vector composed of a plurality of numerical values.
The information feature dimension refers to a feature that can reflect characteristics of the value modification information, such as a feature that reflects inherent attributes of the value modification information, a feature that reflects exposure conditions in which the value modification information is visible and acquirable by a user, a feature that reflects usage conditions of the value modification information, and the like. The target information characteristic dimension is an information characteristic dimension selected from all characteristic dimensions of the value modification information, and is used for calculating the total value of sales of the items (i.e. the total predicted value of sales) generated by the value modification information to be screened, which is possibly used by the user for obtaining the items (such as purchased commodities), and further screening the value modification information. The total value of sales here may be, for example, the total amount of sales. Illustratively, the target information characteristic dimension is an information characteristic dimension which satisfies a preset threshold value in the contribution degree of the value modification information to being hit. In the embodiment of the present invention, the target value modification information filtered based on the target information feature dimension is to be pushed to a user, so that the user uses the target value modification information with a greater probability, so the target information feature dimension selected in the embodiment is an information feature dimension that the user pays more attention to when hitting the value modification information (for example, using a coupon), and may be determined by performing statistical analysis on each information feature dimension of the value modification information and hit data of the value modification information, for example, a ratio of a certain information feature dimension to all information feature dimensions when the value modification information is hit exceeds a corresponding preset threshold (for example, a preset certain ratio value), and the information feature dimension may be determined as a target information feature dimension.
Illustratively, the target information characteristic dimension is a threshold value attribute value that modifies a value attribute value at the time the item is acquired, a value attribute value modification value, a number of days of validity, an exposure amount within a set time period, a pickup amount within a set time period, or a consumption amount within a set time period. Taking the coupon as an example, the threshold value attribute value is the threshold amount of the coupon, and if the coupon is a full-minus coupon of which the value is 400 minus 50, the threshold value attribute value is 400 in the full-minus coupon. The value attribute value is modified to a coupon amount, such as 50 out of the coupon. The number of valid days is the number of available days for the coupon. The exposure amount refers to the number of times the coupon is viewed by the user. The pickup amount refers to the number of coupons picked up by the user. Consumption refers to the number of coupons that are used by the user to place a purchase order. Referring to fig. 3, in the present embodiment, each target information feature dimension is determined as a threshold value attribute value, a value attribute value modification value, the number of valid days, the exposure amount within a set time period (e.g., 7 days), a 7-day fetch amount, and a 7-day consumption amount. The threshold value attribute value, the value attribute value modification value and the valid days can reflect the inherent attribute of the value modification information, are irrelevant to the user behavior and are information characteristics concerned by the user in the process of using the coupon; the 7-day exposure and the 7-day pickup may reflect the exposure of the value modification information, which is related to the user behavior; the 7-day consumption may reflect usage of value modification information, which is related to user behavior. The exposure condition of the value modification information and the sale use condition of the goods are also the more concerned information characteristics in the process of using the coupons by the users. The advantage of this arrangement is that the relevant data of these target information feature dimensions can be obtained quickly, thereby further improving the information distribution efficiency.
Specifically, the total predicted sales value of each piece of value modification information to be screened is obtained by performing model calculation on the feature vector of the corresponding piece of value modification information to be screened through a mathematical model, so that the feature vector of each piece of value modification information to be screened needs to be obtained first. The feature vector may be obtained by performing statistics on related data generated after the value modification information to be screened is generated according to each target information feature dimension, for example, the feature vector may be obtained by performing statistics on all user historical behavior data on the e-commerce platform, or may be obtained by reading from a data mining result of the e-commerce platform.
And S120, determining the total predicted sales value of the value modification information to be screened based on the feature vectors and a preset linear regression model, wherein the linear regression model is obtained by training a linear regression mixed regularization model in advance.
The linear regression model is a preset linear weighting model and is used for performing weighted summation operation on each sub-feature vector in the feature vector. The model parameters of the linear regression model are obtained by training a linear regression model (i.e., a hybrid regularization model of linear regression) to which at least two regularization terms are attached in advance. The mixed regularization model of the linear regression can effectively prevent the over-fitting phenomenon of the linear regression due to the introduction of different types of regularization terms (such as an L1 regularization term and an L2 regularization term), and meanwhile, the correlation among model variables can be eliminated, so that the precision of the model parameters of the linear regression model is improved. In the training process of the linear regression mixed regularization model, the Martian effect in the recommendation process of the value modification information can be reduced by adjusting the total sales values corresponding to the cold value modification information and the hot value modification information respectively.
Specifically, if the contribution of a certain value modification information to the total value of sales of the article is larger, that is, the total sales amount generated by the user shopping using the value modification information is larger, it is indicated that the value modification information can better meet the requirement of the user when the user acquires the article, so the basis for screening the value modification to be screened in the embodiment of the present invention is the predicted total value of sales of the value modification information to be screened. In specific implementation, each sub-feature vector of the feature vector of each piece of value modification information to be screened is subjected to linear weighted summation operation by using a preset linear regression model, so that the total predicted sales value of the corresponding value modification information to be screened can be obtained.
Illustratively, the linear regression model is obtained by pre-training as follows:
A. and acquiring the feature vectors and the total sales value of the value modification information of at least two samples as each training sample.
The sample value modification information refers to value modification information involved in the model training process, and can be generated in the e-commerce platform and can be acquired and used by a user.
Specifically, the mixed regularization model of the linear regression calculates the total sales value of the value modification information through the feature vectors, so that each set of training samples should include the feature vectors and the total sales value of the sample value modification information. The acquisition mode and the processing mode of the feature vector of the sample value modification information and the acquisition mode and the processing mode (see the following description) of the feature vector of the value modification information to be screened, and the total sales value of the sample value modification information can be obtained through the user historical behavior data or the data mining result in the e-commerce platform.
It should be noted that, if the sample value modification information belongs to cold value modification information with less user behaviors, such as user browsing or use, the total sales value of the sample value modification information can be properly increased on the basis of the total sales value obtained through statistics, so as to improve the exposure rate of the cold value modification information. Similarly, if the sample value modification information belongs to the topical value modification information with more user behaviors, the total sales value of the sample value modification information can be properly reduced on the basis of the total sales value obtained by statistics, so as to reduce the exposure rate of the topical value modification information. The adjustment of the total value of sales of the topical value modification information is an optional operation that is not necessary.
B. Inputting each training sample into a linear regression mixed regularization model for model training, and determining the value of each model parameter in the linear regression mixed regularization model, wherein the dependent variable and the independent variable of the linear regression mixed regularization model are respectively the total predicted sales value and each target information characteristic dimension corresponding to the characteristic vector, the regular terms of the linear regression mixed regularization model are the L1 regular term and the L2 regular term, and each model parameter is the weighting weight of the corresponding target information characteristic dimension.
Specifically, the linear regression hybrid regularization model is used for performing weighted summation operation on each sub-feature vector in the feature vector to obtain a total predicted sales value, so a dependent variable in the linear regression hybrid regularization model is the total predicted sales value, an independent variable is each target information feature dimension corresponding to the feature vector, and a model parameter is a weighted weight of each variable. Since there is a strong correlation between the exposure amount in the set time period, the captured amount in the set time period, and the consumption amount in the set time period in each target information feature dimension, the present embodiment needs to introduce an L1 regular term and an L2 regular term to eliminate the correlation between the variables.
In the embodiment, the linear regression mixture regularization model is an elastic network regression model, and each target information feature dimension is, for example, a threshold value attribute value reset _ score, a value attribute value modification value discrete _ score, an effective day active _ score, an exposure amount exposure _ score in a set time period, a pickup amount click _ score in a set time period, or a consumption amount use _ score in a set time period, so that the elastic network regression model is in the form of the formula (1):
wherein y represents the total predicted sales value; x is the number ofiA sub-feature vector representing the ith target information feature dimension; w is aiA weighting weight representing the ith target information characteristic dimension; both alpha and beta are penalty coefficients, the larger the value of the coefficient is, the larger the penalty term is for a more complex model, the alpha and beta are usually set to be 0.5 and 0.6 respectively, and the value of the value depends on the precision requirement of the model;
the method is an L1 regular term with three characteristics of exposure, acquisition quantity and consumption, and can generally solve sparse solution;
is characterized by three characteristics of exposure, quantity of acquisition and consumptionThe L2 regular term, a smooth solution is usually obtained.
Inputting the feature vector and total sales value of each sample value modification information into formula (1), obtaining model equations of corresponding sample value modification information, solving the optimal solution of the model equations, obtaining each weighted weight, namely training to obtain each model parameter wiThe value of (a).
For the optimal solution, since the elastic network regression model is a nonlinear programming problem, gradient descent or other iterative algorithms can be used to solve the optimal values of the model parameters. Taking the gradient descent algorithm to find the optimal solution as an example, based on the elastic network regression model shown in formula (1), the loss function can be shown in formula (2):
based on the loss function of equation (2), the gradient directions of the features of the three irregular terms, i.e., the threshold value attribute value, the value attribute value modification value, and the effective days, are shown in equation (3), and the gradient directions of the features of the three regular terms, i.e., the exposure amount, the extraction amount, and the consumption amount, are shown in equation (4). Based on the gradient directions of equation (3) and equation (4), w may be updated as shown in equation (5)iGradient descent formula (updated model parameter value is w)i')。
Based on the gradient correlation formulas, learning rate l (l value and model precision requirement) of model trainingCorrelation, e.g. 0.01) is continuously performed with gradient-decreasing iterative operations to obtain model parameters wiSee table 1.
TABLE 1 values of model parameters of Linear regression models
| Target information feature dimension | Model parameter values before adjustment |
| Threshold value attribute value restict score | -0.17 |
| Value attribute value modification value discrete _ score | 0.47 |
| Active _ score of effective days | 0.04 |
| Exposure exposure _ score | 0.36 |
| Fetch amount click score | 0.22 |
| Usage amount use _ score | 0.10 |
C. And constructing a linear regression model by using the fitting term of the linear regression mixed regularization model and the value of each model parameter.
Concretely, each model obtained in the step B is participated inNumber wiSubstituting the values of (a) into a fitting term of a linear regression mixed regularization model, e.g. a linear fitting term of an elastic network regression model
A linear regression model may be constructed. The method has the advantages that the training sample set is constructed by utilizing the plurality of sample value modification information in the e-commerce platform and model training is carried out, the linear regression model with simple model form and high calculation precision can be obtained, the linear regression model can be suitable for calculating the total predicted sales value of all the value modification information to be screened in the e-commerce platform, and the determining efficiency and the determining precision of the total predicted sales value of the value modification information to be screened are further improved.
After determining the values of the model parameters in the linear regression hybrid regularization model, and before constructing the linear regression model by using the fitting terms of the linear regression hybrid regularization model and the values of the model parameters, the method further includes: if a negative value exists in the values of the model parameters, the negative value is adjusted to be a positive value smaller than any positive value in the values. When the embodiment of the invention utilizes the total predicted sales value to screen the value modification information to be screened, the information is sorted based on the relative size of the total predicted sales value, and the absolute value of the total predicted sales value is not concerned, so that the value (namely the weighting weight) of each model parameter only needs to ensure that the relative size relationship is correct. On the basis, in this embodiment, all the model parameters are set to be positive values, and if there is a negative value in the model parameters obtained in step B, the negative value needs to be adjusted to be a positive value smaller than any one of the positive values of the model parameter values. If a plurality of negative value exists, the magnitude relationship between the adjusted positive values corresponding to the negative values is required to be consistent with the magnitude relationship of each negative value. For example, for the values of the model parameters obtained in table 1, the negative value is adjusted to be half of the minimum positive value, and the values of the model parameters before and after adjustment are shown in table 2.
TABLE 2 adjustment results of model parameters of the Linear regression model
| Target information feature dimension | Model parameter values before adjustment | Adjusted model parameter values |
| Threshold value attribute value restict score | -0.17 | 0.02 |
| Value attribute value modification value discrete _ score | 0.47 | 0.47 |
| Active _ score of effective days | 0.04 | 0.04 |
| Exposure exposure _ score | 0.36 | 0.36 |
| Fetch amount click score | 0.22 | 0.22 |
| Usage amount use _ score | 0.10 | 0.10 |
According to the adjusted values of the model parameters in the table 2 and the linear fitting terms in the formula (1), a linear regression model as in the formula (6) can be constructed, and the linear regression model can be used for calculating the total value Score _ coupon of the predicted sales of the value modification information to be screened.
Score_coupon＝0.36*exposure_score+0.22*click_score+0.1*usage_score+0.02*restrict_score+0.47*discount_score+0.04*active_score (6)
S130, screening target value modification information from the value modification information to be screened according to the total predicted sales value, and distributing the target value modification information to target users.
The target user refers to a user for whom value modification information needs to be screened.
Specifically, the rule for screening according to the predicted total sales value of each piece of value modification information to be screened is related to the business requirements in practical application, and the business requirements may include the number, the order, whether to classify, and the like of the screening. And screening part of the value modification information to be screened which meets the service requirement from all the value modification information to be screened as target value modification information according to the service requirement and the total value of each predicted sale. And then directly putting the target value modification information into an account of the target user, and informing the target user.
By taking the coupon as an example, referring to fig. 4, a coupon display link and a coupon picking link in the original coupon circulation flow are replaced by a coupon releasing link corresponding to the information distribution method in the embodiment of the invention, so that the coupon circulation flow is reduced, the process that a platform automatically issues the coupon to a user is replaced by the process that the user actively discovers and picks the coupon, the screened and released coupon can better meet the user requirements, and the exposure rate and the utilization rate of the coupon are improved to a great extent.
According to the technical scheme, the linear regression model is obtained by training the linear regression mixed regularization model in advance, so that the model parameters are determined more accurately, the linear regression model between the total sales value and each information characteristic dimension of the value modification information is built, and an accurate model basis is provided for screening and distributing the value modification information. The method comprises the steps that characteristic vectors corresponding to value modification information to be screened are obtained, wherein the characteristic vectors comprise sub characteristic vectors corresponding to at least one target information characteristic dimension of the value modification information, and the value modification information is used for modifying a value attribute value when an article is obtained; determining the total predicted sales value of each piece of value modification information to be screened based on each feature vector and a preset linear regression model, wherein the linear regression model is obtained by training a mixed regularization model of linear regression in advance; and screening target value modification information from the value modification information to be screened according to the total predicted sales value, and distributing the target value modification information to target users. The method and the device realize the calculation of the predicted total sales value of each piece of value modification information to be screened by using the preset linear regression model with high calculation accuracy and high speed, further screen out the target value modification information for the target user according to each predicted total sales value and distribute the target value modification information to the target user, avoid the process that the target user actively finds and obtains the target value modification information, and further improve the exposure rate and the utilization rate of the modified value information.
Example two
In this embodiment, based on the first embodiment, further optimization is performed on "obtaining the value modification information to be screened". On the basis, the feature vectors respectively corresponding to the value modification information to be screened can be further optimized. On the basis, the target value modification information screened from the value modification information to be screened can be further optimized according to the total predicted sales value. Wherein explanations of the same or corresponding terms as those of the above embodiments are omitted. Referring to fig. 5, the information distribution method provided in this embodiment includes:
s210, screening out value modification information to be screened from the value modification information according to the article types in the historical behavior data of the target user.
Specifically, for users in the e-commerce platform, who have certain article type preferences, for example, some users prefer skin care products, some users prefer electronic products, and the like, the users may also prefer their preferred article types when obtaining the value modification information. Therefore, in order to make the screened target value modification information more fit with the requirements of the target user, the embodiment matches the article types corresponding to the value modification information according to the article types related in the historical behavior data of the target user, and determines the value modification information with each article type matching consistently as the value modification information to be screened.
S220, acquiring target user portrait data of a target user.
The user portrait data is data obtained by virtually digitizing an actual user, and describes attributes and behaviors of a class of users, such as sex, age, consumption level, mother and infant features, viewed item data, and purchased item data. The target user representation data is user representation data of the target user. The target user image data includes not only the historical behavior data of the target user but also the historical behavior data of other users belonging to the same user image as the target user.
Specifically, the sub-feature vectors of each target information feature dimension in the feature vectors are obtained through the historical behavior data of the users in the e-commerce platform, and the data volume of the historical behavior data of the target users is relatively small. In order to improve the effectiveness of the feature vector, the feature vector is obtained by using historical behavior data of a plurality of users in the target user portrait data in the embodiment. Therefore, it is necessary to determine target user image data based on user information of a target user, and for example, the target user image data may be constructed based on historical behavior data of the target user, or may be obtained by matching a plurality of constructed user image data based on user information.
And S230, determining a feature vector of each piece of value modification information to be screened according to historical behavior data in the target user portrait data.
Specifically, after the value modification information to be screened is determined, the feature vector of each article to be screened can be obtained from historical behavior data in the portrait data of the target user according to the feature dimension of the target information.
Illustratively, determining the feature vector of each value modification information to be screened according to the historical behavior data in the target user portrait data comprises:
D. and determining the characteristic value of each target information characteristic dimension of each piece of value modification information to be screened according to historical behavior data in the target user portrait data.
Specifically, the value (i.e., the characteristic value) of each target information characteristic dimension of each to-be-screened value modification information may be obtained through statistics from historical behavior data in the target user portrait data, or the characteristic value of each target information characteristic dimension of each to-be-screened value modification information may be obtained through reading from a data report generated through statistics of historical behavior data. Therefore, the characteristic value of each target information characteristic dimension of each piece of value modification information to be screened is obtained. Taking a coupon as an example, and taking the characteristic dimensions of each target information as a threshold value attribute value (threshold amount), a value attribute value modification value (benefit amount), the number of valid days, an exposure amount in a set time period (7-day exposure amount), a retrieval amount in a set time period (7-day retrieval amount), or a consumption amount in a set time period (7-day consumption amount), respectively, as an example, characteristic values as shown in table 3 can be obtained.
TABLE 3 characteristic values of characteristic dimensions of each target information of value modification information to be screened
E. And based on a preset nonlinear transformation algorithm, performing data standardization processing on each characteristic value of each piece of value modification information to be screened to generate a characteristic vector of each piece of value modification information to be screened.
Specifically, as can be seen from table 3, since each target information feature dimension has a different meaning, dimensions of feature values of each target information feature dimension are different, and numerical value intervals have differences, which may reduce the accuracy of the linear regression model. Based on this, in this embodiment, the obtained feature value of each target information feature dimension is subjected to normalization processing to eliminate the above various differences, thereby further improving the modeling accuracy of the linear regression model and the calculation accuracy of the total predicted sales value.
In particular, linear transformation algorithms may be used, e.g.
(wherein, Xnorm、X、XminAnd XmaxRespectively, a feature value after normalization, a feature value before normalization, a feature value minimum value before normalization, and a feature value maximum value before normalization in the same target information feature dimension), normalizing each feature value to [0, 1 ]]. It is also possible to use a predetermined non-linear transformation algorithm, e.g.
(wherein, XnormX and XmaxRespectively, a feature value after normalization, a feature value before normalization, and a feature value maximum value before normalization in the same target information feature dimension), normalizing each feature value to [0, 1 ]]。
The linear transformation algorithm or the nonlinear transformation algorithm adopted in the feature value standardization depends on the precision requirement of the service requirement, a specific linear regression model form, a training model form of a linear regression mixed regularization model adopted in the model training and the like. With respect to the requirement of higher precision of the elastic network regression model and the value modification information screening, the nonlinear transformation algorithm is selected in this embodiment to perform the normalization processing of the characteristic value.
In addition to the selection of the normalization algorithm, in this embodiment, additional processing is performed on the selection of the maximum value of the feature value, for example, the feature value at the 99-quantile position of all the feature value distributions in the same target information feature dimension is selected as the maximum value of the feature value, and then the normalization processing of each feature value can be performed, and the obtained result is each normalized feature value. And forming a feature vector of the corresponding value modification information to be screened by the normalized feature values of each value modification information to be screened.
In the model training phase, the total sales value in each training sample also needs to be subjected to the same standardization processing operation.
S240, determining the total predicted sales value of each piece of value modification information to be screened based on each feature vector and a preset linear regression model.
And S250, classifying the value modification information to be screened, and determining the article type to which the value modification information to be screened belongs.
Specifically, the types of articles involved in the e-commerce platform are various, and in order to avoid that the value modification information is recommended to fall into a small range of one or some article types, the value modification information is isolated from the article types in this embodiment. In specific implementation, all the value modification information to be screened is classified according to the article type corresponding to each value modification information to be screened. The classification of the article classification can be determined according to business requirements, such as a primary class, a secondary class, a tertiary class or a custom class.
S260, aiming at each article class, screening target value modification information from the value modification information to be screened belonging to the article class according to the predicted total sales value of the value modification information to be screened belonging to the article class.
Specifically, after all the value modification information to be screened is classified into categories, all the value modification information to be screened belonging to each item category is sorted according to the predicted total sales value, and the sorting result of the value modification information to be screened for distinguishing the item categories as exemplified in table 4 is obtained. Then, the value modification information can be screened based on the sorting result according to the screening rule (such as screening quantity) in the service requirement for screening the value modification information. If the item type determined in S250 exceeds the item type corresponding to the target user, then in this operation, screening is performed according to the screening rule in each piece of value modification information to be screened under the item type corresponding to the target user.
TABLE 4 ranking results of value modification information to be screened based on tertiary classification and predicted total sales value
And S270, distributing the target value modification information to the target users.
According to the technical scheme of the embodiment, the value modification information to be screened is screened out from the value modification information according to the article types in the historical behavior data of the target user. The method and the device have the advantages that all the value modification information to be screened is determined based on the historical behavior data of the target user, so that the screened target value modification information is more in line with the item type preference of the target user, the matching degree between the target object value modification information and the target user is improved, and the distribution accuracy and the individuation of the target value modification information are further improved. Acquiring target user portrait data of a target user; and determining a feature vector of each piece of value modification information to be screened according to historical behavior data in the target user portrait data. The effectiveness of the feature vector and the relevance between the feature vector and the target user are improved, the matching degree between the target object value modification information and the target user is further improved, and therefore the distribution individuality of the target value modification information is further improved. Classifying the value modification information to be screened to determine the article class to which the value modification information to be screened belongs; and screening target value modification information from the value modification information to be screened belonging to the article category according to the predicted total sales value of the value modification information to be screened belonging to the article category aiming at each article category. The information distribution based on the article types is realized, and the diversity of the distributed value modification information is further increased on the basis of matching the target object value modification information with the target user.
EXAMPLE III
The present embodiment provides an information distribution apparatus, and referring to fig. 6, the apparatus specifically includes:
the feature vector acquisition module 610 is configured to acquire feature vectors corresponding to the value modification information to be screened, where the feature vectors include sub-feature vectors corresponding to at least one target information feature dimension of the value modification information, and the value modification information is used to modify a value attribute value of the article when the article is acquired;
the total predicted sales value determining module 620 is configured to determine the total predicted sales value of each piece of value modification information to be screened based on each feature vector and a preset linear regression model, where the linear regression model is obtained by training a linear regression hybrid regularization model in advance;
and the target value modification information distribution module 630 is configured to screen out target value modification information from the value modification information to be screened according to each predicted total sales value, and distribute the target value modification information to the target users.
Optionally, the feature vector obtaining module 610 is specifically configured to:
and screening out the value modification information to be screened from the value modification information according to the article types in the historical behavior data of the target user.
Optionally, the feature vector obtaining module 610 is further specifically configured to:
acquiring target user portrait data of a target user;
and determining a feature vector of each piece of value modification information to be screened according to historical behavior data in the target user portrait data.
Further, the feature vector obtaining module 610 is further specifically configured to:
determining the characteristic value of each target information characteristic dimension of each piece of value modification information to be screened according to historical behavior data in the target user portrait data;
and based on a preset nonlinear transformation algorithm, performing data standardization processing on each characteristic value of each piece of value modification information to be screened to generate a characteristic vector of each piece of value modification information to be screened.
Optionally, the target value modification information distribution module 630 is specifically configured to:
classifying the value modification information to be screened, and determining the article class to which the value modification information to be screened belongs;
and screening target value modification information from the value modification information to be screened belonging to the article category according to the predicted total sales value of the value modification information to be screened belonging to the article category aiming at each article category.
Optionally, on the basis of the foregoing apparatus, the apparatus further includes a model training module, configured to obtain a linear regression model through pre-training in the following manner:
obtaining the characteristic vectors and the total sales value of at least two sample value modification information as each training sample;
inputting each training sample into a linear regression mixed regularization model for model training, and determining the value of each model parameter in the linear regression mixed regularization model, wherein the dependent variable and the independent variable of the linear regression mixed regularization model are respectively a total predicted sales value and each target information characteristic dimension corresponding to a characteristic vector, the regular terms of the linear regression mixed regularization model are an L1 regular term and an L2 regular term, and each model parameter is the weighting weight of the corresponding target information characteristic dimension;
and constructing a linear regression model by using the fitting term of the linear regression mixed regularization model and the value of each model parameter.
Further, the model training module is further configured to:
after the values of the model parameters in the linear regression mixed regularization model are determined, and before the linear regression model is constructed by using the fitting terms of the linear regression mixed regularization model and the values of the model parameters, if a negative value exists in the values of the model parameters, the negative value is adjusted to be a positive value smaller than any positive value in the values.
Optionally, the target information feature dimension is an information feature dimension whose contribution degree to the hit of the value modification information satisfies a preset threshold.
Further, the target information characteristic dimension is a threshold value attribute value for modifying the value attribute value when the article is acquired, a value attribute value modification value, the number of valid days, the amount of exposure in a set time period, the amount of acquisition in a set time period, or the amount of consumption in a set time period.
Through the information distribution device of the third embodiment of the invention, the total predicted sales value of each piece of value modification information to be screened is calculated by using the preset linear regression model with high calculation precision and high speed, the target value modification information is screened out for the target user according to each total predicted sales value and is distributed to the target user, and the process that the target user actively finds and acquires the target value modification information is avoided, so that the exposure rate and the utilization rate of the modification value information are improved.
The information distribution device provided by the embodiment of the invention can execute the information distribution method provided by any embodiment of the invention, and has corresponding functional modules and beneficial effects of the execution method.
It should be noted that, in the embodiment of the information distribution apparatus, each unit and each module included in the embodiment are only divided according to functional logic, but are not limited to the above division as long as the corresponding functions can be implemented; in addition, specific names of the functional units are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present invention.
Example four
Referring to fig. 7, the present embodiment provides an electronic device 700, which includes: one or more processors 720; the storage device 710 is configured to store one or more programs, and when the one or more programs are executed by the one or more processors 720, the one or more processors 720 implement the information distribution method provided in the embodiment of the present invention, including:
acquiring feature vectors corresponding to the value modification information to be screened respectively, wherein the feature vectors comprise sub-feature vectors corresponding to at least one target information feature dimension of the value modification information respectively, and the value modification information is used for modifying the value attribute value when the article is acquired;
determining the total predicted sales value of each piece of value modification information to be screened based on each feature vector and a preset linear regression model, wherein the linear regression model is obtained by training a mixed regularization model of linear regression in advance;
and screening target value modification information from the value modification information to be screened according to the total predicted sales value, and distributing the target value modification information to target users.
Of course, those skilled in the art will understand that the processor 720 can also implement the technical solution of the information distribution method provided by any embodiment of the present invention.
The electronic device 700 shown in fig. 7 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. 7, electronic device 700 is embodied in the form of a general purpose computing device. The components of the electronic device 700 may include, but are not limited to: one or more processors 720, a memory device 710, and a bus 750 that couples the various system components (including the memory device 710 and the processors 720).
The storage 710 may include computer system readable media in the form of volatile memory, such as Random Access Memory (RAM)711 and/or cache memory 712. The electronic device 700 may further include other removable/non-removable, volatile/nonvolatile computer system storage media. By way of example only, storage system 713 may be used to read from and write to non-removable, nonvolatile magnetic media (not shown in FIG. 7, commonly referred to as a "hard drive"). Although not shown in FIG. 7, a magnetic disk drive for reading from and writing to a removable, nonvolatile magnetic disk (e.g., a "floppy disk") and an optical disk drive for reading from or writing to a removable, nonvolatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In such cases, each drive may be connected to bus 750 by one or more data media interfaces. Storage 710 may include at least one program product having a set (e.g., at least one) of program modules that are configured to carry out the functions of embodiments of the invention.
A program/utility 714 having a set (at least one) of program modules 715 may be stored, for instance, in storage 710, such program modules 715 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each of which examples or some combination may comprise an implementation of a network environment. The program modules 715 generally perform the functions and/or methodologies of any of the embodiments described herein.
The electronic device 700 may also communicate with one or more external devices 760 (e.g., keyboard, pointing device, display 770, etc.), with one or more devices that enable a user to interact with the electronic device 700, and/or with any devices (e.g., network card, modem, etc.) that enable the electronic device 700 to communicate with one or more other computing devices. Such communication may occur via input/output (I/O) interface 730. Also, the electronic device 700 may communicate with one or more networks (e.g., a Local Area Network (LAN), a Wide Area Network (WAN), and/or a public network, such as the internet) via the network adapter 740. As shown in FIG. 7, the network adapter 740 communicates with the other modules of the electronic device 700 via the bus 750. It should be appreciated that although not shown in the figures, other hardware and/or software modules may be used in conjunction with the electronic device 700, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, among others.
EXAMPLE five
The present embodiments provide a storage medium containing computer-executable instructions which, when executed by a computer processor, are operable to perform a method of information distribution, the method comprising:
acquiring feature vectors corresponding to the value modification information to be screened respectively, wherein the feature vectors comprise sub-feature vectors corresponding to at least one target information feature dimension of the value modification information respectively, and the value modification information is used for modifying the value attribute value when the article is acquired;
determining the total predicted sales value of each piece of value modification information to be screened based on each feature vector and a preset linear regression model, wherein the linear regression model is obtained by training a mixed regularization model of linear regression in advance;
and screening target value modification information from the value modification information to be screened according to the total predicted sales value, and distributing the target value modification information to target users.
Of course, the storage medium provided by the embodiment of the present invention contains computer-executable instructions, and the computer-executable instructions are not limited to the above method operations, and may also perform related operations in the information distribution method provided by any embodiment of the present invention.
Computer storage media for embodiments of the invention may employ any combination of one or more computer-readable media. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. 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 (a non-exhaustive list) of the computer readable storage medium would include the following: 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 context of this document, 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.
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, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.
Computer program code for carrying out operations for aspects of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C + +, or the like, as well as conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.
Claims (12)
1. An information distribution method, comprising:
acquiring feature vectors corresponding to the value modification information to be screened respectively, wherein the feature vectors comprise sub-feature vectors corresponding to at least one target information feature dimension of the value modification information respectively, and the value modification information is used for modifying the value attribute value of the article when the article is acquired;
determining the total predicted sales value of each piece of value modification information to be screened based on each feature vector and a preset linear regression model, wherein the linear regression model is obtained by training a linear regression mixed regularization model in advance;
and screening target value modification information from the value modification information to be screened according to the total predicted sales value, and distributing the target value modification information to target users.
2. The method of claim 1, wherein obtaining each of the value modification information to be filtered comprises:
and screening out the value modification information to be screened from the value modification information according to the article types in the historical behavior data of the target user.
3. The method according to claim 1 or 2, wherein the obtaining of the feature vector corresponding to each value modification information to be screened comprises:
acquiring target user portrait data of the target user;
and determining a feature vector of each piece of value modification information to be screened according to historical behavior data in the target user portrait data.
4. The method of claim 3, wherein determining the feature vector of each value modification information to be filtered according to the historical behavior data in the target user representation data comprises:
determining the characteristic value of each target information characteristic dimension of each piece of value modification information to be screened according to historical behavior data in the target user portrait data;
and based on a preset nonlinear transformation algorithm, performing data standardization on each characteristic value of each to-be-screened value modification information to generate a characteristic vector of each to-be-screened value modification information.
5. The method of claim 1, wherein screening target value modification information from each of the value modification information to be screened based on each of the predicted total sales values comprises:
classifying the value modification information to be screened, and determining the article class to which the value modification information to be screened belongs;
and screening target value modification information from the value modification information to be screened belonging to the article category according to the predicted total sales value of the value modification information to be screened belonging to the article category aiming at each article category.
6. The method of claim 1, wherein the linear regression model is pre-trained by:
obtaining the characteristic vectors and the total sales value of at least two sample value modification information as each training sample;
inputting each training sample into the linear regression mixed regularization model for model training, and determining the value of each model parameter in the linear regression mixed regularization model, wherein the dependent variable and the independent variable of the linear regression mixed regularization model are respectively a total predicted sales value and each target information feature dimension corresponding to the feature vector, the regular terms of the linear regression mixed regularization model are an L1 regular term and an L2 regular term, and each model parameter is the weighting weight of the corresponding target information feature dimension;
and constructing the linear regression model by using the fitting term of the linear regression mixed regularization model and the value of each model parameter.
7. The method of claim 6, wherein after the determining the values of the model parameters in the linear regression hybrid regularization model and before the constructing the linear regression model using the fitting term of the linear regression hybrid regularization model and the values of the model parameters, further comprising:
and if a negative value exists in the values of the model parameters, adjusting the negative value to be a positive value smaller than any positive value in the values.
8. The method of claim 1, wherein the target information feature dimension is an information feature dimension that satisfies a preset threshold with respect to a degree of contribution to the value modification information being hit.
9. The method of claim 8, wherein the target information characteristic dimension is a threshold value attribute value that modifies a value attribute value at which the item is acquired, a value attribute value modification value, a number of days in effect, an exposure amount over a set time period, a pickup amount over the set time period, or a consumption amount over the set time period.
10. An information distribution apparatus characterized by comprising:
the characteristic vector acquisition module is used for acquiring characteristic vectors corresponding to the value modification information to be screened respectively, wherein the characteristic vectors comprise sub characteristic vectors corresponding to at least one target information characteristic dimension of the value modification information respectively, and the value modification information is used for modifying a value attribute value when an article is acquired;
the total predicted sales value determining module is used for determining the total predicted sales value of each piece of value modification information to be screened based on each feature vector and a preset linear regression model, wherein the linear regression model is obtained by training a linear regression mixed regularization model in advance;
and the target value modification information distribution module is used for screening out target value modification information from the value modification information to be screened according to the total predicted sales value and distributing the target value modification information to target users.
11. An electronic device, characterized in that the electronic device comprises:
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 information distribution method of any one of claims 1-9.
12. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the information distribution method according to any one of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010092716.4A CN113269574A (en) | 2020-02-14 | 2020-02-14 | Information distribution method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010092716.4A CN113269574A (en) | 2020-02-14 | 2020-02-14 | Information distribution method, device, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113269574A true CN113269574A (en) | 2021-08-17 |
Family
ID=77227356
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010092716.4A Pending CN113269574A (en) | 2020-02-14 | 2020-02-14 | Information distribution method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113269574A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102254270A (en) * | 2010-05-19 | 2011-11-23 | 索尼公司 | Special discount voucher selective supporting device, system, method and program |
| US20170221086A1 (en) * | 2016-01-28 | 2017-08-03 | Sap Se | Framework for generating coupons |
| CN107767179A (en) * | 2017-10-25 | 2018-03-06 | 口碑(上海)信息技术有限公司 | The quality evaluating method and device of electronic ticket |
| JP2018081337A (en) * | 2016-11-14 | 2018-05-24 | カタリナ マーケティング ジャパン株式会社 | Coupon issuing system |
| CN110070382A (en) * | 2018-01-24 | 2019-07-30 | 北京京东尚科信息技术有限公司 | Method and apparatus for generating information |
| CN110223122A (en) * | 2019-06-14 | 2019-09-10 | 江苏云脑数据科技有限公司 | Discount coupon animation effect assessment prediction method |
-
2020
- 2020-02-14 CN CN202010092716.4A patent/CN113269574A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102254270A (en) * | 2010-05-19 | 2011-11-23 | 索尼公司 | Special discount voucher selective supporting device, system, method and program |
| US20170221086A1 (en) * | 2016-01-28 | 2017-08-03 | Sap Se | Framework for generating coupons |
| JP2018081337A (en) * | 2016-11-14 | 2018-05-24 | カタリナ マーケティング ジャパン株式会社 | Coupon issuing system |
| CN107767179A (en) * | 2017-10-25 | 2018-03-06 | 口碑(上海)信息技术有限公司 | The quality evaluating method and device of electronic ticket |
| CN110070382A (en) * | 2018-01-24 | 2019-07-30 | 北京京东尚科信息技术有限公司 | Method and apparatus for generating information |
| CN110223122A (en) * | 2019-06-14 | 2019-09-10 | 江苏云脑数据科技有限公司 | Discount coupon animation effect assessment prediction method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107908740B (en) | Information output method and device | |
| WO2021081962A1 (en) | Recommendation model training method, recommendation method, device, and computer-readable medium | |
| US10181138B2 (en) | System and method for determining retail-business-rule coefficients from current prices | |
| US20200143439A1 (en) | System and method for integrating retail price optimization for revenue and profit with business rules | |
| WO2018053966A1 (en) | Click rate estimation | |
| CN109840730B (en) | Method and device for data prediction | |
| US9846885B1 (en) | Method and system for comparing commercial entities based on purchase patterns | |
| US10515378B2 (en) | Extracting relevant features from electronic marketing data for training analytical models | |
| CN109213936B (en) | Commodity searching method and device | |
| US20210366006A1 (en) | Ranking of business object | |
| US20190080352A1 (en) | Segment Extension Based on Lookalike Selection | |
| WO2020221022A1 (en) | Service object recommendation method | |
| CN111612581A (en) | Method, device and equipment for recommending articles and storage medium | |
| CN111754287A (en) | Article screening method, apparatus, device and storage medium | |
| WO2020150611A1 (en) | Systems and methods for entity performance and risk scoring | |
| CN111966886A (en) | Object recommendation method, object recommendation device, electronic equipment and storage medium | |
| CN112598472A (en) | Product recommendation method, device, system, medium and program product | |
| CN107679916A (en) | For obtaining the method and device of user interest degree | |
| CN110598183A (en) | Flow distribution method, device, equipment and storage medium | |
| CN113407854A (en) | Application recommendation method, device and equipment and computer readable storage medium | |
| CN115983900A (en) | Method, apparatus, device, medium, and program product for constructing user marketing strategy | |
| CN110852768A (en) | Dynamic pricing method and system, equipment and storage medium | |
| CN112348590A (en) | Method and device for determining value of article, electronic equipment and storage medium | |
| CN113778979A (en) | Method and device for determining live broadcast click rate | |
| CN113763089A (en) | Article recommendation method and device and computer-readable storage medium |
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 |