WO2019169977A1 - 一种信息转化率的预测、信息推荐方法和装置 - Google Patents

一种信息转化率的预测、信息推荐方法和装置 Download PDF

Info

Publication number
WO2019169977A1
WO2019169977A1 PCT/CN2019/073365 CN2019073365W WO2019169977A1 WO 2019169977 A1 WO2019169977 A1 WO 2019169977A1 CN 2019073365 W CN2019073365 W CN 2019073365W WO 2019169977 A1 WO2019169977 A1 WO 2019169977A1
Authority
WO
WIPO (PCT)
Prior art keywords
recommended
information
rate
conversion
probability
Prior art date
Application number
PCT/CN2019/073365
Other languages
English (en)
French (fr)
Inventor
周志超
熊军
周峰
蒋建
黄国进
郑岩
冯健
Original Assignee
阿里巴巴集团控股有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 阿里巴巴集团控股有限公司 filed Critical 阿里巴巴集团控股有限公司
Publication of WO2019169977A1 publication Critical patent/WO2019169977A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION 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
    • G06Q30/00Commerce
    • G06Q30/02Marketing; Price estimation or determination; Fundraising
    • G06Q30/0201Market modelling; Market analysis; Collecting market data
    • G06Q30/0202Market predictions or forecasting for commercial activities
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION 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
    • G06Q30/00Commerce
    • G06Q30/02Marketing; Price estimation or determination; Fundraising
    • G06Q30/0207Discounts or incentives, e.g. coupons or rebates
    • G06Q30/0211Determining the effectiveness of discounts or incentives
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION 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
    • G06Q30/00Commerce
    • G06Q30/02Marketing; Price estimation or determination; Fundraising
    • G06Q30/0207Discounts or incentives, e.g. coupons or rebates
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION 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
    • G06Q30/00Commerce
    • G06Q30/02Marketing; Price estimation or determination; Fundraising
    • G06Q30/0241Advertisements
    • G06Q30/0242Determining effectiveness of advertisements

Definitions

  • the present application relates to the field of computer technology, and in particular, to a method for predicting information conversion rate, and a method and device for recommending information.
  • the information recommender In the information recommendation, the information recommender usually predicts the conversion rate of the recommended information (English name: Conversion Rate, CVR for short) before recommending the information, and determines which information is recommended to the user according to the predicted conversion rate. In the case of recommending a coupon to a user, the information recommender usually predicts the conversion rate of each coupon after recommending the plurality of coupons to the user before recommending the plurality of coupons to the user, so as to facilitate the conversion rate. Coupons are recommended to users.
  • the conversion rate of the recommended information (English name: Conversion Rate, CVR for short) before recommending the information, and determines which information is recommended to the user according to the predicted conversion rate.
  • the information recommender In the case of recommending a coupon to a user, the information recommender usually predicts the conversion rate of each coupon after recommending the plurality of coupons to the user before recommending the plurality of coupons to the user, so as to facilitate the conversion rate. Coupons are recommended to users.
  • the click rate of the information when predicting the conversion rate of information, the click rate of the information can be predicted, and the conversion rate of the information can be measured by the click rate of the information.
  • the click rate of the information if the click rate of the information is high, it can be regarded as a high conversion rate of the information.
  • the accuracy of measuring the conversion rate of information according to the click rate of the information is low, and the information recommendation cannot be effectively performed to the user.
  • the embodiment of the present application provides a method for predicting information conversion rate, a method for recommending information, and a device for solving the problem that the accuracy of the information is low according to the click rate of the information, and the information cannot be effectively recommended to the user.
  • a method for predicting information conversion rate including:
  • a prediction device for information conversion rate including:
  • Determining a unit determining information to be recommended and characteristic information of the user
  • the first prediction unit predicts, according to the to-be-recommended information and the feature information, a conversion probability corresponding to at least two factors that affect the conversion process of the information to be recommended;
  • a second prediction unit predicting a conversion rate of the information to be recommended according to the at least two conversion probabilities.
  • an electronic device comprising:
  • a memory arranged to store computer executable instructions that, when executed, cause the processor to perform the following operations:
  • a computer readable storage medium storing one or more programs, the one or more programs, when executed by an electronic device comprising a plurality of applications, causing the The electronic device performs the following methods:
  • a method for information recommendation including:
  • an information recommendation apparatus including:
  • a first determining unit configured to determine a plurality of to-be-recommended information and feature information of the user
  • the second determining unit determines, according to the plurality of to-be-recommended information and the feature information, a conversion probability corresponding to at least two factors that affect the conversion process of the plurality of to-be-recommended information;
  • a prediction unit predicting a conversion rate of each of the to-be-recommended information according to the at least two conversion probabilities
  • the information recommendation unit performs information recommendation to the user according to the predicted conversion rate of each of the to-be-recommended information.
  • an electronic device comprising:
  • a memory arranged to store computer executable instructions that, when executed, cause the processor to perform the following operations:
  • a computer readable storage medium storing one or more programs, when the one or more programs are executed by an electronic device including a plurality of applications, The electronic device performs the following methods:
  • a method for predicting a coupon conversion rate comprising:
  • the conversion rate of the coupon to be recommended is predicted according to at least two of the click rate, the receiving rate, and the verification rate.
  • a device for predicting a coupon conversion rate comprising:
  • Determining a unit determining a coupon to be recommended and characteristic information of the user
  • the first prediction unit predicts at least two of a click rate, a receiving rate, and a verification rate of the to-be-recommended coupon during the conversion process according to the ticket to be recommended and the feature information;
  • the second prediction unit predicts a conversion rate of the to-be-recommended coupon according to at least two of the click rate, the receiving rate, and the verification rate.
  • an electronic device comprising:
  • a memory arranged to store computer executable instructions that, when executed, cause the processor to perform the following operations:
  • the conversion rate of the coupon to be recommended is predicted according to at least two of the click rate, the receiving rate, and the verification rate.
  • a computer readable storage medium storing one or more programs, when the one or more programs are executed by an electronic device including a plurality of applications, The electronic device performs the following methods:
  • the conversion rate of the coupon to be recommended is predicted according to at least two of the click rate, the receiving rate, and the verification rate.
  • a method of recommending a voucher comprising:
  • a coupon is recommended to the user based on the conversion rate of the plurality of coupons to be recommended.
  • an apparatus for recommending a voucher comprising:
  • Determining a unit determining a plurality of coupons to be recommended and characteristic information of the user
  • the first prediction unit predicts at least two of a click rate, a receiving rate, and a verification rate of the plurality of to-be-recommended coupons in the conversion process according to the plurality of coupons to be recommended and the feature information;
  • the second prediction unit predicts a conversion rate of the plurality of coupons to be recommended according to at least two of the click rate, the receiving rate, and the verification rate;
  • the recommendation unit recommends the voucher to the user according to the conversion rate of the plurality of coupons to be recommended.
  • an electronic device comprising:
  • a memory arranged to store computer executable instructions that, when executed, cause the processor to perform the following operations:
  • a coupon is recommended to the user based on the conversion rate of the plurality of coupons to be recommended.
  • a computer readable storage medium storing one or more programs, when the one or more programs are executed by an electronic device including a plurality of applications, The electronic device performs the following methods:
  • a coupon is recommended to the user based on the conversion rate of the plurality of coupons to be recommended.
  • the technical solution provided by the embodiment of the present application determines the information to be recommended and the feature information of the user; and predicts, according to the information to be recommended and the feature information, a conversion probability corresponding to at least two factors affecting the conversion process of the information to be recommended; And predicting a conversion rate of the information to be recommended according to a conversion probability corresponding to the at least two factors.
  • the prediction information conversion rate can be predicted according to the conversion probability of at least two factors affecting the information conversion process, the predicted information conversion rate can more fully reflect the true information conversion rate, thereby improving information conversion.
  • the accuracy of the prediction of the rate in turn, can effectively recommend information to the user.
  • FIG. 1 is a schematic flow chart of a method for predicting information conversion rate according to an embodiment of the present application
  • FIG. 2 is a schematic diagram of a prediction of a coupon conversion rate and a recommended coupon of one embodiment of the present application
  • FIG. 3 is a schematic flow chart of an information recommendation method according to an embodiment of the present application.
  • FIG. 4 is a schematic flow chart of a method for predicting a coupon conversion rate according to an embodiment of the present application
  • FIG. 5 is a schematic flow chart of a method for recommending coupons according to an embodiment of the present application.
  • FIG. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
  • FIG. 7 is a schematic structural diagram of an apparatus for predicting information conversion rate according to an embodiment of the present application.
  • FIG. 8 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
  • FIG. 9 is a schematic structural diagram of an information recommendation apparatus according to an embodiment of the present application.
  • FIG. 10 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
  • FIG. 11 is a schematic structural diagram of a device for predicting a coupon conversion rate according to an embodiment of the present application.
  • FIG. 12 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
  • FIG. 13 is a schematic structural diagram of an apparatus for recommending coupons according to an embodiment of the present application.
  • the conversion rate of information can be understood as the conversion efficiency of the information from exposure to write-off.
  • the information recommender can predict the conversion rate of the information according to the click-through rate of the information before recommending the information to the user, so as to compare the conversion rate. High information is recommended to the user. Among them, the higher the click-through rate of the information, the higher the conversion rate can be considered. However, the click rate usually represents the user’s clicks or views on the information. The conversion rate of the information through the click rate does not effectively and comprehensively reflect the conversion rate of the information, that is, the conversion rate based on the click rate prediction is less accurate.
  • the embodiment of the present application provides a method for predicting information conversion rate, a method for recommending information, and a device for predicting information conversion rate, including: determining information to be recommended and feature information of a user; And the feature information, predicting a conversion probability corresponding to the at least two factors that affect the conversion process of the information to be recommended; and predicting a conversion rate of the information to be recommended according to the conversion probability corresponding to the at least two factors.
  • the prediction conversion rate can be predicted according to the conversion probability of at least two factors affecting the information conversion process, the prediction is not based only on the click rate of the information, and thus the predicted information conversion rate can be more comprehensive. Reflecting the true information conversion rate, thereby improving the prediction accuracy of information conversion rate, and thus effectively recommending information to users.
  • the technical solution provided by the embodiment of the present application can be applied to recommend information to the user in the case of a known user.
  • the technical solution provided by the embodiment of the present application may be used to determine the conversion rate of the information that needs to be recommended, and recommend the information to the user according to the conversion rate. If the number of information to be recommended is 1, the information may be recommended to the user according to the conversion rate. If the number of information to be recommended is multiple, the information may be recommended to the user according to the conversion rate. .
  • FIG. 1 is a schematic flow chart of a method for predicting information conversion rate according to an embodiment of the present application. The method is as follows.
  • S102 Determine information to be recommended and feature information of the user.
  • the web server or the server of the application may recommend information to the user based on the webpage or the application, and the user may be regarded as a recommendation.
  • the target user of the information Before recommending information to the user, the information to be recommended recommended to the user and the feature information of the user may be determined.
  • the to-be-recommended information may be information that is to be recommended to the user, but may not be recommended to the user in the end.
  • the information to be recommended may be electronic information, and may be other electronic information such as an electronic coupon or a web address link.
  • the number of the information to be recommended may be one or multiple.
  • the user's characteristic information may be the user's personal information, for example, the user's height, age, weight, etc., or may be the user's behavior data, for example, the restaurant that the user frequently goes to, the user's favorite operation, etc., here is no longer an example. Description.
  • the feature information of the user may be determined according to the webpage currently browsed by the user or the opened application. For example, when the user browses the webpage, the server of the website may obtain information such as the webpage browsing record of the user, and if the user has logged into the website, the user may also Obtain the personal information that the user fills in when registering the website. For example, when the user opens the application, the server of the application can obtain the history record of the user using the application. If the user has logged in to the application, the user name of the user can also be obtained.
  • S104 After determining the to-be-recommended information recommended by the user and the feature information of the user, S104 may be performed.
  • S104 Prediction, according to the to-be-recommended information and the feature information, a conversion probability corresponding to at least two factors that affect the conversion process of the information to be recommended.
  • a conversion probability corresponding to at least two factors affecting the conversion process of the information to be recommended may be predicted, that is, at least two conversion probabilities that affect the conversion rate of the information to be recommended are predicted.
  • the entire conversion process (ie, the conversion link) of the to-be-recommended information may include: exposing the to-be-recommended information, viewing the to-be-recommended information, obtaining the to-be-recommended information, and using the to-be-recommended information.
  • the information to be recommended may be the information recommended by the user in the information recommendation direction.
  • the information to be recommended may be the user clicking or browsing the information to be recommended, and the information to be recommended may be the user receiving or obtaining the information to be recommended, and the information to be recommended may be used. It is the last operation performed by the user on the recommended information. For example, if the information to be recommended is a coupon, the information to be recommended may be that the user uses the coupon to make a payment. If the information to be recommended is a web link, the information to be recommended may be the user. Forward the URL link, no longer one by one.
  • the viewing, obtaining, and using the information to be recommended may affect the conversion rate of the information to be recommended. Therefore, the at least two factors affecting the conversion process of the information to be recommended may include: viewing the information to be recommended, obtaining the information to be recommended, and Use at least two of the information to be recommended.
  • the at least two conversion probabilities that affect the conversion process of the information to be recommended may include: a first probability of viewing the information to be recommended, a second probability of obtaining the information to be recommended, and a third probability of using the third probability of the information to be recommended. At least two.
  • the first probability of the information to be recommended may be: the ratio of the number of users M to the N to be recommended by the N users after the information to be recommended is recommended to the N users; and the second probability of obtaining the information to be recommended may be
  • the ratio of the number of users P to L M, N, and L P are all non-negative integers.
  • the conversion probability corresponding to the at least two factors that affect the conversion process of the information to be recommended is predicted according to the to-be-recommended information and the feature information, and may include:
  • the conversion rate model including at least two of a first probability model, a second probability model, and a third probability model
  • the embodiment of the present application may be pre-trained to obtain a conversion rate model for predicting information to be recommended, and the conversion rate model may specifically include at least two of a first probability model, a second probability model, and a third probability model, where The first probability model is used to predict a first probability of the information to be recommended, the second probability model is used to predict a second probability of the information to be recommended, and the third probability model is used to predict a third probability of the information to be recommended.
  • the attribute information of the information to be recommended may be based on at least two of the first probability model, the second probability model, and the third probability model included in the conversion rate model.
  • the recommendation method and the feature information of the user are input as input, and at least two of the first probability and the second probability third probability of the information to be recommended are respectively obtained.
  • the attribute information of the to-be-recommended information may include a category of the information to be recommended (for example, the information to be recommended is a network link or a coupon, etc.), the amount of the information to be recommended, the information recommender to which the information to be recommended belongs, and the like.
  • the recommendation information recommendation method may include a short message, an email, a website link, or an application notification information, and is not specifically limited.
  • the training when the conversion rate model is obtained, the training may be specifically obtained by the following methods, including:
  • the historical data includes: attribute information of the information to be recommended, at least two of viewing data, acquisition data, and usage data of the information to be recommended, where the viewing data includes Clicking on the feature information of the user of the information to be recommended, the acquired data includes feature information of the user who obtains the information to be recommended, and the usage data includes feature information of the user who uses the information to be recommended;
  • the historical data is trained based on a preset model to obtain the conversion rate model.
  • the coupon may be an electronic coupon, and may specifically be a coupon, a full coupon, etc., and is not specifically limited herein.
  • At least two factors affecting the conversion process of the coupon may include at least two of a click coupon, a coupon, and a counterfeit coupon
  • at least two conversion probabilities affecting the coupon conversion process may include: Click rate (corresponding to the first probability of viewing the information to be recommended), the rate of receipt of the ticket (corresponding to the second probability of obtaining the information to be recommended), and the write-off rate of the ticket (corresponding to the third probability of using the information to be recommended) At least two of them.
  • the conversion rate model for predicting the coupon conversion rate may include: a click rate model (corresponding to the first probability model), a receiving rate model (corresponding to the second probability model), and a write-off rate model (corresponding to the third probability model) At least two of them.
  • historical data of the coupon can be acquired.
  • the historical data of the voucher may include attribute information of the voucher, and the attribute information may be a category of the voucher, for example, a voucher or a full voucher, or may be an amount of the voucher or an amount of the voucher, or may be a voucher Information recommenders, etc., are not illustrated here.
  • the historical data of the vouchers may also include a recommendation manner of the vouchers, and the recommended manner may be a short message, a mail, a website link, or a notification information of the application, and is not specifically limited.
  • the historical data of the coupon may also include click data of the coupon (view data corresponding to the information to be recommended), receipt data of the coupon (acquisition data corresponding to the information to be recommended), and a core of the coupon under different attribute information and recommendation methods. At least two of the pin data (corresponding to the usage data of the information to be recommended).
  • the click data of the voucher may include the feature information of the user who clicks on the voucher, and the feature information may be the personal information of the user who clicks on the voucher, such as the height, weight, age, etc. of the user, or may be the behavior data of the user who clicks on the voucher. For example, restaurants that users like to go to, sports that they like, and so on.
  • the receipt data of the voucher may include the feature information of the user who collects the voucher.
  • the feature information may include the personal information of the user who collects the voucher and the behavior data. For details, refer to the description of the personal information and behavior data of the user who clicks on the voucher. Repeat the description.
  • the write-off data of the voucher may include feature information of the user who issues the voucher, and the feature information may include personal information of the user who issues the voucher and behavior data.
  • the historical data may be trained based on a preset model, and a conversion rate model is obtained.
  • the preset model may be a deep learning model or a neural network model, and may be determined according to actual needs, and is not specifically limited herein. For example, if the training based on the deep learning model is determined according to actual conditions, the result is more Good, you can use the deep learning model for training.
  • the conversion rate model including the click rate model, the receiving rate model, and the verification rate model
  • the click rate model, the receiving rate model, and the write-off can be separately trained.
  • Rate model the preset models used in training the click-through rate model, the receiving rate model, and the verification rate model may be the same or different.
  • the attribute information of the coupon included in the historical data, the recommendation method, and the feature information of the user who clicks the coupon under different recommendation modes may be used as an input variable, and the user may have different attribute information for the attribute information.
  • the click rate of the coupons in different recommended modes is used as an output variable, and the training is based on the preset model, and finally the click rate model for determining the coupon hit rate is obtained.
  • the machine learning model can be specifically used, and the click rate model is obtained through continuous data iterative training.
  • At least two of the click rate, the receiving rate, and the verification rate of the coupon conversion process may be determined based on the conversion rate model of the coupon.
  • step S106 After determining the conversion rate model of the information to be recommended based on the method described above, and predicting the conversion probability corresponding to the at least two factors affecting the conversion process of the information to be recommended according to the conversion rate model, step S106 may be performed.
  • S106 Predict a conversion rate of the information to be recommended according to a conversion probability corresponding to the at least two factors.
  • the conversion rate of the information to be recommended may be further predicted.
  • predicting the conversion rate of the information to be recommended according to the conversion probability corresponding to the at least two factors may include:
  • a product of the predicted at least two of the first probability, the second probability, and the third probability is determined as a conversion rate of the information to be recommended.
  • the product of the click rate of the coupon and the write-off rate may be determined as the conversion rate of the coupon; if the acceptance rate and the write-off rate of the coupon are predicted, the The product of the coupon collection rate and the write-off rate is determined as the conversion rate of the coupon; if the click rate, the acceptance rate, and the write-off rate of the coupon are predicted, the product of the click rate, the receipt rate, and the write-off rate of the coupon may be determined as The conversion rate of the coupon, of course, can also be the product of any two conversion probabilities as the conversion rate of the coupon.
  • the write-off rate of the coupon has a greater influence on the conversion rate of the coupon. Therefore, when predicting at least two conversion probabilities affecting the conversion process of the coupon, at least two of the conversion probabilities may be prioritized. Including the write-off rate of the voucher, when further predicting the conversion rate of the voucher, the verification rate of the voucher may be used as a necessary condition for predicting the conversion rate of the voucher, and at least one of the click rate and the collection rate of the voucher. The product is used as the conversion rate of the coupon.
  • the conversion probability corresponding to the three factors affecting the coupon conversion process, namely, the click rate, the acceptance rate, and the write-off rate, and the predicted click rate and the acceptance rate of the coupon.
  • the product of the write-off rate as the conversion rate of the coupon.
  • the method further includes:
  • the information recommendation When the information recommendation is performed to the user, if the number of pieces of information to be recommended is 1, it may be determined whether the information to be recommended is recommended to the user. For example, if the conversion rate of the information to be recommended is high, the information may be recommended to the user, and vice versa.
  • the information recommendation to the user according to the conversion rate of the information to be recommended may include:
  • the information to be recommended in the range, and the to-be-recommended information with the highest conversion rate among the determined information to be recommended is recommended to the user.
  • the recommended information may be sorted, and one of the to-be-recommended information is selected and recommended to the user according to the sorting result.
  • the information to be recommended is described as a voucher, and the specific sorting method of the plurality of recommended coupons may include at least the following two types:
  • One method is to sort multiple coupons according to the size of the conversion rate, and select one of the coupons with the highest conversion rate to recommend to the user.
  • one of the coupons may be randomly selected for recommendation to the user, or one of the coupons may be recommended to the user according to a preset rule. For example, a coupon with a larger amount may be selected for recommendation to the user.
  • all the coupons with the same conversion rate can also be recommended to the user, and the user can determine which one to use according to the actual situation.
  • Another method is to predict the click rate, the picking rate, and the write-off rate according to the click rate, the receiving rate, and the verification rate, respectively, after predicting at least two of the click rate, the receiving rate, and the write-off rate of the coupon, determining the click rate and receiving At least two of the rate and the write-off rate are in the set of coupons, and the coupons with the highest conversion rate among the coupons are recommended to the user.
  • the predicted click rate, acceptance rate, and write-off rate of the six coupons are: a1, a2, and a3 (corresponding to A's click rate, acceptance rate and write-off rate), b1, b2, b3, c1, c2, c3, d1, d2, d3, e1, e2, e3, f1, f2, f3, where, according to the click rate
  • the result of sorting the six coupons by the size of the write-off rate is: c3, b3, d3, f3, a3, e1.
  • the coupon may be recommended to the user in a corresponding recommended manner.
  • FIG. 2 is a schematic diagram of a prediction of a coupon conversion rate and a coupon for an embodiment of the present application.
  • Fig. 2 illustrates an example in which the conversion rate of the coupon is calculated based on the click rate, the collection rate, and the verification rate of the coupon.
  • the click rate model, the collection rate model, and the verification rate model of the voucher can be separately trained according to the history click data, the historical collection data, and the historical verification data of the voucher.
  • the historical click data can be modeled based on the preset machine learning model A, and finally the click rate model is obtained.
  • the specific method refer to the method of training the click rate model described above, and no longer Repeat the description.
  • the historical collection data can be trained based on the preset machine learning model B to obtain a collection rate model
  • the historical verification data is trained based on the preset learning model C to obtain a verification rate model.
  • the attribute information of the coupon to be recommended to the user and the recommendation method, the user's characteristic information, and the click rate of the coupon based on the click rate model may be used according to the receiving rate model.
  • the predicted rate of acceptance model predicts the write-off rate of the voucher based on the write-off rate model. Thereafter, the product of the predicted click rate, the collection rate, and the verification rate of the coupon can be used as the conversion rate of the coupon.
  • the number of coupons to be recommended to the user is plural.
  • the plurality of coupons can be sorted according to the conversion rate, and the coupon with the highest conversion rate can be obtained. And recommend the coupon to the user in the corresponding exposure channel. For example, if the voucher is an application-related voucher, the voucher can be recommended to the user in the manner of the notification information of the application.
  • the full reduction coupon D can be recommended to the user in the manner of the notification message of the application.
  • the plurality of coupons may also be sorted according to at least two of a click rate, a receiving rate, and a verification rate, and at least two of the click rate, the receiving rate, and the verification rate are determined to be Set the coupons within the sorting range, and recommend the coupons with the highest conversion rate among the determined coupons to the user, and the description will not be repeated here.
  • the technical solution provided by the embodiment of the present application determines the information to be recommended and the feature information of the user; and predicts, according to the information to be recommended and the feature information, a conversion probability corresponding to at least two factors affecting the conversion process of the information to be recommended; And predicting a conversion rate of the information to be recommended according to a conversion probability corresponding to the at least two factors.
  • the prediction information conversion rate can be predicted according to the conversion probability of at least two factors affecting the information conversion process, the predicted information conversion rate can more fully reflect the true information conversion rate, thereby improving information conversion.
  • the accuracy of the prediction of the rate in turn, can effectively recommend information to the user.
  • FIG. 3 is a schematic flow chart of an information recommendation method according to an embodiment of the present application.
  • the embodiment of the present application is described as an example of preparing a plurality of to-be-recommended information to the user.
  • the information recommendation method is as follows.
  • S302 Determine a plurality of to-be-recommended information and feature information of the user.
  • S304 Predict a conversion probability corresponding to at least two factors affecting the conversion process of the plurality of to-be-recommended information according to the plurality of to-be-recommended information and the feature information.
  • a conversion probability corresponding to at least two factors affecting the conversion process of the information to be recommended may be predicted.
  • S306 Predict a conversion rate of the plurality of to-be-recommended information according to a conversion probability corresponding to the at least two factors.
  • S308 Perform information recommendation to the user according to the conversion rate of the plurality of to-be-recommended information.
  • the information recommendation to the user according to the conversion rate of the plurality of to-be-recommended information may include:
  • the conversion probability corresponding to the at least two factors that affect the plurality of to-be-recommended information conversion processes includes: viewing the first probability of the information to be recommended, and acquiring the information of the to-be-recommended information At least two of a second probability and a third probability of using the information to be recommended;
  • the information to be recommended in the range, and the to-be-recommended information with the highest conversion rate among the determined information to be recommended is recommended to the user.
  • the technical solution provided by the embodiment of the present application determines a plurality of to-be-recommended information and feature information of the user; and predicts at least two types of processes that affect the conversion process of the plurality of to-be-recommended information according to the plurality of to-be-recommended information and the feature information. a conversion probability corresponding to the factor; predicting a conversion rate of the plurality of to-be-recommended information according to the conversion probability corresponding to the at least two factors; and performing information recommendation to the user according to the conversion rate of the plurality of to-be-recommended information.
  • the conversion rate of the information can be predicted according to the conversion probability of at least two factors affecting the information conversion process, the information conversion rate with higher accuracy can be predicted, and the information recommendation can be effectively performed to the user.
  • the above examples illustrate the prediction method of information conversion rate and the implementation process of information recommendation.
  • the implementation process of the embodiment of the present application is described below by taking the scenario of the information conversion rate prediction method and the information recommendation method applied to the coupon as an example.
  • the to-be-recommended information may be a voucher, and at least two factors affecting the voucher conversion process include: at least two of a click coupon, a voucher, and a verification voucher, and at least two conversion probabilities that affect the voucher conversion process include: a click rate. At least two of the rate, the rate of receipt, and the write-off rate.
  • the voucher may be an electronic voucher, and may specifically be a coupon, a coupon, a full coupon, or the like.
  • the method of predicting the coupon conversion rate is as follows.
  • S402 Determine a coupon to be recommended and feature information of the user.
  • S404 predict, according to the coupon to be recommended and the feature information, at least two of a click rate, a receiving rate, and a verification rate of the coupon to be recommended during the conversion process.
  • S406 predict a conversion rate of the to-be-recommended coupon according to at least two of the click rate, the receiving rate, and the verification rate.
  • predicting at least two of a click rate, a receiving rate, and a verification rate of the to-be-recommended coupon during the conversion process including:
  • the conversion rate model including at least two of a click rate model, a yield rate model, and a write-off rate model;
  • the conversion rate model is obtained by the following methods, including:
  • the historical data includes: a recommendation manner of the coupon to be recommended, in the click data, the receiving data, and the verification data of the coupon to be recommended in the recommended manner
  • the click data includes feature information of a user who clicks on the ticket to be recommended
  • the receiving data includes feature information of a user who receives the ticket to be recommended
  • the verification data includes using the Characteristic information of the user of the coupon to be recommended
  • the historical data is trained based on a preset model to obtain the conversion rate model.
  • predicting a conversion rate of the to-be-recommended coupon according to at least two of the click rate, the receiving rate, and the verification rate including:
  • the product of the click rate and the verification rate is determined as the conversion rate of the coupon to be recommended
  • the product of the receiving rate and the verification rate is determined as the conversion rate of the coupon to be recommended
  • the product of the click rate, the receiving rate, and the verification rate is determined as the conversion rate of the ticket to be recommended.
  • the method further includes:
  • a coupon is recommended to the user based on the conversion rate.
  • recommending a voucher to the user according to the conversion rate including:
  • the technical solution provided by the embodiment of the present application determines the coupon to be recommended and the feature information of the user; and predicts the click rate and the receiving rate of the ticket to be recommended in the conversion process according to the ticket to be recommended and the feature information. And at least two of the verification rates; predicting a conversion rate of the coupon to be recommended according to at least two of the click rate, the collection rate, and the verification rate.
  • the predicted conversion rate of the coupon can be predicted according to at least two conversion probabilities affecting the click rate, the receiving rate, and the write-off rate of the coupon conversion process, the predicted coupon conversion rate can be more comprehensively reflected.
  • the real coupon conversion rate thereby increasing the forecast accuracy of the coupon conversion rate, and thus effectively recommending coupons to users.
  • FIG. 5 is a schematic flow chart of a method for recommending coupons according to an embodiment of the present application.
  • the method of recommending coupons is as follows.
  • S502 Determine a plurality of coupons to be recommended and feature information of the user.
  • S504 predict, according to the plurality of coupons to be recommended and the feature information, at least two of a click rate, a receiving rate, and a verification rate of the plurality of to-be-recommended coupons in the conversion process.
  • S506 predict a conversion rate of the plurality of coupons to be recommended according to at least two of the click rate, the receiving rate, and the verification rate.
  • S508 Recommend a voucher to the user according to the conversion rate of the plurality of coupons to be recommended.
  • recommending the coupon to the user including:
  • the technical solution provided by the embodiment of the present application determines a plurality of coupons to be recommended and feature information of the user; and predicts, according to the plurality of coupons to be recommended and the feature information, the plurality of coupons to be recommended in the conversion process At least two of a click rate, a receiving rate, and a write-off rate; predicting a conversion rate of the plurality of coupons to be recommended according to at least two of the click rate, the receiving rate, and the verification rate And recommending a voucher to the user according to the conversion rate of the plurality of coupons to be recommended.
  • the conversion rate of the voucher can be predicted according to at least two conversion probabilities affecting the click rate, the receiving rate, and the verification rate of the voucher conversion process, the higher accuracy can be predicted.
  • the coupon conversion rate in turn, can effectively recommend coupons to users.
  • FIG. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
  • the electronic device includes a processor, optionally including an internal bus, a network interface, and a memory.
  • the memory may include a memory, such as a high-speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk memory.
  • RAM high-speed random access memory
  • non-volatile memory such as at least one disk memory.
  • the electronic device may also include hardware required for other services.
  • the processor, the network interface, and the memory may be interconnected by an internal bus, which may be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended) Industry Standard Architecture, extending the industry standard structure) bus.
  • the bus can be divided into an address bus, a data bus, a control bus, and the like. For ease of representation, only one double-headed arrow is shown in Figure 6, but it does not mean that there is only one bus or one type of bus.
  • the program can include program code, the program code including computer operating instructions.
  • the memory can include both memory and non-volatile memory and provides instructions and data to the processor.
  • the processor reads the corresponding computer program from the non-volatile memory into memory and then runs to form a prediction device for information conversion rate at a logical level.
  • the processor executes the program stored in the memory and is specifically used to perform the following operations:
  • the method performed by the apparatus for predicting information conversion rate disclosed in the embodiment shown in FIG. 6 of the present application may be applied to a processor or implemented by a processor.
  • the processor may be an integrated circuit chip with signal processing capabilities.
  • each step of the above method may be completed by an integrated logic circuit of hardware in a processor or an instruction in a form of software.
  • the above processor may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; or may be a digital signal processor (DSP), dedicated integration.
  • ASIC Application Specific Integrated Circuit
  • FPGA Field-Programmable Gate Array
  • other programmable logic device discrete gate or transistor logic device, discrete hardware component.
  • the general purpose processor may be a microprocessor or the processor or any conventional processor or the like.
  • the steps of the method disclosed in the embodiments of the present application may be directly implemented by the hardware decoding processor, or may be performed by a combination of hardware and software modules in the decoding processor.
  • the software module can be located in a conventional storage medium such as random access memory, flash memory, read only memory, programmable read only memory or electrically erasable programmable memory, registers, and the like.
  • the storage medium is located in the memory, and the processor reads the information in the memory and combines the hardware to complete the steps of the above method.
  • the electronic device can also perform the method of FIG. 1 and realize the function of the information conversion rate prediction device in the embodiment shown in FIG. 1 , which is not described herein again.
  • the electronic device of the present application does not exclude other implementation manners, such as a logic device or a combination of software and hardware, etc., that is, the execution body of the following processing flow is not limited to each logical unit. It can also be hardware or logic.
  • the embodiment of the present application further provides a computer readable storage medium storing one or more programs, the one or more programs including instructions, when the portable electronic device is included in a plurality of applications When executed, the portable electronic device can be configured to perform the method of the embodiment shown in FIG. 1 and specifically for performing the following operations:
  • FIG. 7 is a schematic structural diagram of an information conversion rate predicting apparatus 70 according to an embodiment of the present application.
  • the information conversion rate prediction apparatus 70 may include: a determining unit 71, a first prediction unit 72, and a second prediction unit 73, where:
  • Determining unit 71 determining information to be recommended and feature information of the user
  • the first prediction unit 72 predicts, according to the to-be-recommended information and the feature information, a conversion probability corresponding to at least two factors that affect the conversion process of the information to be recommended;
  • the second prediction unit 73 predicts a conversion rate of the to-be-recommended information according to a conversion probability corresponding to the at least two factors.
  • the at least two factors affecting the to-be-recommended information conversion process include: viewing the to-be-recommended information, obtaining the to-be-recommended information, and using the to-be-recommended information;
  • the conversion probability corresponding to the at least two factors includes: a first probability of viewing the information to be recommended, a second probability of acquiring the information to be recommended, and a third probability of using the information to be recommended.
  • the first prediction unit 72 predicts a conversion probability corresponding to at least two factors that affect the conversion process of the information to be recommended, including:
  • the conversion rate model including at least two of a first probability model, a second probability model, and a third probability model
  • the first prediction unit 72 trains to obtain the conversion rate model by:
  • the historical data includes: attribute information and recommendation manner of the information to be recommended, viewing data, obtaining data, and using the information to be recommended in different attribute information and recommendation manners At least two of the data, the viewing data includes feature information of a user who clicks on the information to be recommended, the acquired data includes feature information of a user who obtains the information to be recommended, and the usage data includes using the Feature information of the user who recommended the information;
  • the historical data is trained based on a preset model to obtain the conversion rate model.
  • the second prediction unit 73 predicts a conversion rate of the to-be-recommended information according to the conversion probability corresponding to the at least two factors, including:
  • first probability and the third probability are predicted, determining a product of the first probability and the second probability as a conversion rate of the information to be recommended;
  • the information conversion rate prediction apparatus 70 further includes: a recommendation unit 74, wherein:
  • the recommendation unit 74 After the second prediction unit 73 predicts the conversion rate of the to-be-recommended information, the recommendation unit 74 performs information recommendation to the user according to the conversion rate of the to-be-recommended information.
  • the recommending unit 74 performs information recommendation to the user according to the conversion rate of the information to be recommended, including:
  • the information to be recommended in the range, and the to-be-recommended information with the highest conversion rate among the determined information to be recommended is recommended to the user.
  • the information conversion rate predicting device 70 can also perform the method of FIG. 1 or FIG. 2 and realize the function of the information conversion rate predicting device in the embodiment shown in FIG. 1 and FIG. 2, which is not described herein again.
  • FIG. 8 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
  • the electronic device includes a processor, optionally including an internal bus, a network interface, and a memory.
  • the memory may include a memory, such as a high-speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk memory.
  • RAM high-speed random access memory
  • non-volatile memory such as at least one disk memory.
  • the electronic device may also include hardware required for other services.
  • the processor, the network interface, and the memory may be interconnected by an internal bus, which may be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended) Industry Standard Architecture, extending the industry standard structure) bus.
  • the bus can be divided into an address bus, a data bus, a control bus, and the like. For ease of representation, only one double-headed arrow is shown in Figure 8, but it does not mean that there is only one bus or one type of bus.
  • the program can include program code, the program code including computer operating instructions.
  • the memory can include both memory and non-volatile memory and provides instructions and data to the processor.
  • the processor reads the corresponding computer program from the non-volatile memory into the memory and then runs to form an information recommendation device at a logical level.
  • the processor executes the program stored in the memory and is specifically used to perform the following operations:
  • the method performed by the information recommendation apparatus disclosed in the embodiment shown in FIG. 8 of the present application may be applied to a processor or implemented by a processor.
  • the processor may be an integrated circuit chip with signal processing capabilities.
  • each step of the above method may be completed by an integrated logic circuit of hardware in a processor or an instruction in a form of software.
  • the above processor may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; or may be a digital signal processor (DSP), dedicated integration.
  • ASIC Application Specific Integrated Circuit
  • FPGA Field-Programmable Gate Array
  • other programmable logic device discrete gate or transistor logic device, discrete hardware component.
  • the general purpose processor may be a microprocessor or the processor or any conventional processor or the like.
  • the steps of the method disclosed in the embodiments of the present application may be directly implemented by the hardware decoding processor, or may be performed by a combination of hardware and software modules in the decoding processor.
  • the software module can be located in a conventional storage medium such as random access memory, flash memory, read only memory, programmable read only memory or electrically erasable programmable memory, registers, and the like.
  • the storage medium is located in the memory, and the processor reads the information in the memory and combines the hardware to complete the steps of the above method.
  • the electronic device can also perform the method of FIG. 3 and implement the functions of the information recommendation device in the embodiment shown in FIG. 3, which is not described herein again.
  • the electronic device of the present application does not exclude other implementation manners, such as a logic device or a combination of software and hardware, etc., that is, the execution body of the following processing flow is not limited to each logical unit. It can also be hardware or logic.
  • the embodiment of the present application further provides a computer readable storage medium storing one or more programs, the one or more programs including instructions, when the portable electronic device is included in a plurality of applications When executed, the portable electronic device can be configured to perform the method of the embodiment shown in FIG. 1 and specifically for performing the following operations:
  • FIG. 9 is a schematic structural diagram of an information recommendation predicting apparatus 90 according to an embodiment of the present application.
  • the information recommendation apparatus 90 may include: a determining unit 91, a first prediction unit 92, a second prediction unit 93, and a recommendation unit 94, where:
  • the determining unit 91 determines a plurality of to-be-recommended information and feature information of the user;
  • the first prediction unit 92 predicts a conversion probability corresponding to at least two factors affecting the conversion process of the plurality of to-be-recommended information according to the plurality of to-be-recommended information and the feature information;
  • the second prediction unit 93 predicts a conversion rate of the plurality of to-be-recommended information according to a conversion probability corresponding to the at least two factors;
  • the recommendation unit 94 performs information recommendation to the user according to the conversion rate of the plurality of to-be-recommended information.
  • the recommending unit 94 performs information recommendation to the user according to the conversion rate of the plurality of to-be-recommended information, including:
  • the conversion probability corresponding to the at least two factors that affect the multiple information to be recommended conversion process includes: viewing a first probability of the information to be recommended, obtaining a second probability of the information to be recommended, and using the At least two of the third probabilities of the information to be recommended;
  • the recommendation unit 94 performs information recommendation to the user according to the conversion rate of the plurality of to-be-recommended information, including:
  • the information to be recommended in the range, and the to-be-recommended information with the highest conversion rate among the determined information to be recommended is recommended to the user.
  • the information recommendation device 90 can also perform the method of FIG. 3 and implement the functions of the information recommendation device in the embodiment shown in FIG. 3. The embodiments of the present application are not described herein again.
  • FIG. 10 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
  • the electronic device includes a processor, optionally including an internal bus, a network interface, and a memory.
  • the memory may include a memory, such as a high-speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk memory.
  • RAM high-speed random access memory
  • non-volatile memory such as at least one disk memory.
  • the electronic device may also include hardware required for other services.
  • the processor reads the corresponding computer program from the non-volatile memory into memory and then runs to form a predictive device for the coupon conversion rate at a logical level.
  • the processor executes the program stored in the memory and is specifically used to perform the following operations:
  • the conversion rate of the coupon to be recommended is predicted according to at least two of the click rate, the receiving rate, and the verification rate.
  • the method described above for the voucher conversion rate prediction apparatus disclosed in the embodiment shown in FIG. 10 of the present application may be applied to a processor or implemented by a processor.
  • the specific connection structure and the implemented functions of other hardware in FIG. 10 can be referred to the related description in the electronic device shown in FIG. 6, and the description is not repeated here.
  • the electronic device can also perform the method of FIG. 4 and implement the function of the device for predicting the conversion rate of the voucher in the embodiment shown in FIG. 4, which is not described herein again.
  • the electronic device of the present application does not exclude other implementation manners, such as a logic device or a combination of software and hardware, etc., that is, the execution body of the following processing flow is not limited to each logical unit. It can also be hardware or logic.
  • the embodiment of the present application further provides a computer readable storage medium storing one or more programs, the one or more programs including instructions, when the portable electronic device is included in a plurality of applications When executed, the portable electronic device can be configured to perform the method of the embodiment shown in FIG. 4, and is specifically configured to perform the following operations:
  • the conversion rate of the coupon to be recommended is predicted based on at least two of the click rate, the receiving rate, and the verification rate.
  • FIG. 11 is a schematic structural diagram of a coupon conversion rate prediction apparatus 110 according to an embodiment of the present application.
  • the coupon conversion rate prediction apparatus 110 may include: a determining unit 111, a first prediction unit 112, and a second prediction unit 113, where:
  • Determining unit 111 determining a coupon to be recommended and feature information of the user
  • the first prediction unit 112 predicts at least two of the click rate, the receiving rate, and the verification rate of the to-be-recommended coupon during the conversion process according to the coupon to be recommended and the feature information;
  • the second prediction unit 113 predicts a conversion rate of the to-be-recommended coupon according to at least two of the click rate, the receiving rate, and the verification rate.
  • the first prediction unit 112 predicts at least two of a click rate, a receiving rate, and a verification rate of the to-be-recommended coupon during the conversion process according to the ticket to be recommended and the feature information.
  • Species including:
  • the conversion rate model including at least two of a click rate model, a yield rate model, and a write-off rate model;
  • the first prediction unit 112 trains the conversion rate model by:
  • the historical data includes: a recommendation manner of the coupon to be recommended, in the click data, the receiving data, and the verification data of the coupon to be recommended in the recommended manner
  • the click data includes feature information of a user who clicks on the ticket to be recommended
  • the receiving data includes feature information of a user who receives the ticket to be recommended
  • the verification data includes using the Characteristic information of the user of the coupon to be recommended
  • the historical data is trained based on a preset model to obtain the conversion rate model.
  • the second prediction unit 113 predicts a conversion rate of the to-be-recommended coupon according to at least two of the click rate, the receiving rate, and the verification rate, including:
  • the product of the click rate and the verification rate is determined as the conversion rate of the coupon to be recommended
  • the product of the receiving rate and the verification rate is determined as the conversion rate of the coupon to be recommended
  • the product of the click rate, the receiving rate, and the verification rate is determined as the conversion rate of the ticket to be recommended.
  • the coupon conversion rate prediction apparatus 110 further includes: a recommendation unit 114, where:
  • the recommendation unit 114 after the second prediction unit 113 predicts the conversion rate of the coupon to be recommended, recommends the coupon to the user according to the conversion rate.
  • the recommending unit 114 according to the conversion rate, recommending a voucher to the user, including:
  • the voucher conversion rate prediction device 110 can also perform the method of FIG. 4 and implement the function of the voucher conversion rate prediction device in the embodiment shown in FIG. 4, which is not described herein again.
  • FIG. 12 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
  • the electronic device includes a processor, optionally including an internal bus, a network interface, and a memory.
  • the memory may include a memory, such as a high-speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk memory.
  • RAM high-speed random access memory
  • the electronic device may also include hardware required for other services.
  • the processor reads the corresponding computer program from the non-volatile memory into memory and then runs to form a device for recommending coupons on a logical level.
  • the processor executes the program stored in the memory and is specifically used to perform the following operations:
  • a coupon is recommended to the user based on the conversion rate of the plurality of coupons to be recommended.
  • the method performed by the apparatus for recommending coupons disclosed in the embodiment shown in FIG. 12 of the present application may be applied to a processor or implemented by a processor.
  • the specific connection structure and the functions of the other hardware in FIG. 12 can be referred to the related description in the electronic device shown in FIG. 8, and the description is not repeated here.
  • the electronic device can also perform the method of FIG. 5 and implement the functions of the device for recommending the voucher in the embodiment shown in FIG. 5, which is not described herein again.
  • the electronic device of the present application does not exclude other implementation manners, such as a logic device or a combination of software and hardware, etc., that is, the execution body of the following processing flow is not limited to each logical unit. It can also be hardware or logic.
  • the embodiment of the present application further provides a computer readable storage medium storing one or more programs, the one or more programs including instructions, when the portable electronic device is included in a plurality of applications When executed, the portable electronic device can be caused to perform the method of the embodiment shown in FIG. 5, and is specifically configured to perform the following operations:
  • a coupon is recommended to the user based on the conversion rate of the plurality of coupons to be recommended.
  • FIG. 13 is a schematic structural diagram of an apparatus 130 for recommending coupons according to an embodiment of the present application.
  • the apparatus 130 for recommending coupons may include: a determining unit 131, a first prediction unit 132, a second prediction unit 133, and a recommendation unit 134, where:
  • the determining unit 131 determines a plurality of coupons to be recommended and feature information of the user
  • the first prediction unit 132 predicts at least two of the click rate, the receiving rate, and the verification rate of the plurality of to-be-recommended coupons during the conversion process according to the plurality of coupons to be recommended and the feature information;
  • the second prediction unit 133 predicts a conversion rate of the plurality of coupons to be recommended according to at least two of the click rate, the receiving rate, and the verification rate;
  • the recommending unit 134 recommends the coupon to the user according to the conversion rate of the plurality of coupons to be recommended.
  • the recommending unit 134 according to the conversion rate of the plurality of coupons to be recommended, recommending the coupon to the user, including:
  • the apparatus 130 for recommending coupons may also perform the method of FIG. 5 and implement the functions of the apparatus for recommending coupons in the embodiment shown in FIG. 5, and details are not described herein again.
  • the system, device, module or unit illustrated in the above embodiments may be implemented by a computer chip or an entity, or by a product having a certain function.
  • a typical implementation device is a computer.
  • the computer can be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or A combination of any of these devices.
  • Computer readable media includes both permanent and non-persistent, removable and non-removable media.
  • Information storage can be implemented by any method or technology.
  • the information can be computer readable instructions, data structures, modules of programs, or other data.
  • Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read only memory. (ROM), electrically erasable programmable read only memory (EEPROM), flash memory or other memory technology, compact disk read only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, Magnetic tape cartridges, magnetic tape storage or other magnetic storage devices or any other non-transportable media can be used to store information that can be accessed by a computing device.
  • computer readable media does not include temporary storage of computer readable media, such as modulated data signals and carrier waves.

Landscapes

  • Business, Economics & Management (AREA)
  • Strategic Management (AREA)
  • Engineering & Computer Science (AREA)
  • Accounting & Taxation (AREA)
  • Development Economics (AREA)
  • Finance (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Economics (AREA)
  • Game Theory and Decision Science (AREA)
  • Marketing (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Communication Control (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

一种信息转化率的预测、信息推荐方法和装置,该信息转化率的预测方法包括:确定待推荐信息以及用户的特征信息(S102);根据所述待推荐信息以及所述特征信息,预测影响所述待推荐信息转化过程的至少两种因素对应的转化概率(S104);根据所述至少两种因素对应的转化概率,预测所述待推荐信息的转化率(S106)。

Description

一种信息转化率的预测、信息推荐方法和装置
相关申请的交叉引用
本专利申请要求于2018年03月07日提交的、申请号为201810187238.8、发明名称为“一种信息转化率的预测、信息推荐方法和装置”的中国专利申请的优先权,该申请的全文以引用的方式并入本文中。
技术领域
本申请涉及计算机技术领域,尤其涉及一种信息转化率的预测、信息推荐方法和装置。
背景技术
在信息推荐中,信息推荐方在进行信息推荐之前,通常会预测其所推荐的信息的转化率(英文全称:Conversion Rate,简称CVR),并根据预测的转化率确定向用户推荐哪些信息。以向用户推荐优惠券为例,信息推荐方在向用户推荐多个优惠券之前,通常会预测将多个优惠券推荐给用户后每一个优惠券的转化率,以便于将转化率较高的优惠券推荐给用户。
通常,在对信息的转化率进行预测时,可以对信息的点击率进行预测,并通过信息的点击率衡量信息的转化率。其中,若信息的点击率较高,则可以视为信息的转化率较高。然而,在实际应用中,根据信息的点击率来衡量信息的转化率的准确度较低,不能有效地向用户进行信息推荐。
发明内容
本申请实施例提供一种信息转化率的预测、信息推荐方法和装置,用于解决在根据信息的点击率衡量信息的转化率时准确度较低,不能有效地向用户进行信息推荐的问题。
为解决上述技术问题,本申请实施例是这样实现的:
第一方面,提出一种信息转化率的预测方法,包括:
确定待推荐信息以及用户的特征信息;
根据所述待推荐信息以及所述特征信息,预测影响所述待推荐信息转化过程的至少 两种因素对应的转化概率;
根据所述至少两种因素对应的转化概率,预测所述待推荐信息的转化率。
第二方面,提出了一种信息转化率的预测装置,包括:
确定单元,确定待推荐信息以及用户的特征信息;
第一预测单元,根据所述待推荐信息以及所述特征信息,预测影响所述待推荐信息转化过程的至少两种因素对应的转化概率;
第二预测单元;根据所述至少两种转化概率,预测所述待推荐信息的转化率。
第三方面,提出一种电子设备,该电子设备包括:
处理器;以及
被安排成存储计算机可执行指令的存储器,该可执行指令在被执行时使该处理器执行以下操作:
确定待推荐信息以及用户的特征信息;
根据所述待推荐信息以及所述特征信息,预测影响所述待推荐信息转化过程的至少两种因素对应的转化概率;
根据所述至少两种因素对应的转化概率,预测所述待推荐信息的转化率。
第四方面,提出一种计算机可读存储介质,所述计算机可读存储介质存储一个或多个程序,所述一个或多个程序当被包括多个应用程序的电子设备执行时,使得所述电子设备执行以下方法:
确定待推荐信息以及用户的特征信息;
根据所述待推荐信息以及所述特征信息,预测影响所述待推荐信息转化过程的至少两种因素对应的转化概率;
根据所述至少两种因素对应的转化概率,预测所述待推荐信息的转化率。
第五方面,提出一种信息推荐方法,包括:
确定多个待推荐信息以及用户的特征信息;
根据所述多个待推荐信息以及所述特征信息,预测影响所述多个待推荐信息转化过程的至少两种因素对应的转化概率;
根据所述至少两种因素对应的转化概率,预测所述多个待推荐信息的转化率;
根据所述多个待推荐信息的转化率,向所述用户进行信息推荐。
第六方面,提出一种信息推荐装置,包括:
第一确定单元,确定多个待推荐信息以及用户的特征信息;
第二确定单元,根据所述多个待推荐信息以及所述特征信息,确定影响所述多个待推荐信息转化过程的至少两种因素对应的转化概率;
预测单元,根据所述至少两种转化概率,预测每一个所述待推荐信息的转化率;
信息推荐单元,根据预测的每一个所述待推荐信息的转化率,向所述用户进行信息推荐。
第七方面,提出一种电子设备,该电子设备包括:
处理器;以及
被安排成存储计算机可执行指令的存储器,该可执行指令在被执行时使该处理器执行以下操作:
确定多个待推荐信息以及用户的特征信息;
根据所述多个待推荐信息以及所述特征信息,预测影响所述多个待推荐信息转化过程的至少两种因素对应的转化概率;
根据所述至少两种因素对应的转化概率,预测所述多个待推荐信息的转化率;
根据所述多个待推荐信息的转化率,向所述用户进行信息推荐。
第八方面,提出一种计算机可读存储介质,所述计算机可读存储介质存储一个或多个程序,所述一个或多个程序当被包括多个应用程序的电子设备执行时,使得所述电子设备执行以下方法:
确定多个待推荐信息以及用户的特征信息;
根据所述多个待推荐信息以及所述特征信息,预测影响所述多个待推荐信息转化过程的至少两种因素对应的转化概率;
根据所述至少两种因素对应的转化概率,预测所述多个待推荐信息的转化率;
根据所述多个待推荐信息的转化率,向所述用户进行信息推荐。
第九方面,提出一种券转化率的预测方法,包括:
确定待推荐的券以及用户的特征信息;
根据所述待推荐的券以及所述特征信息,预测所述待推荐的券在转化过程中的点击率、领取率以及核销率中的至少两种;
根据所述点击率、所述领取率以及所述核销率中的至少两种,预测所述待推荐的券的转化率。
第十方面,提出了一种券转化率的预测装置,包括:
确定单元,确定待推荐的券以及用户的特征信息;
第一预测单元,根据所述待推荐的券以及所述特征信息,预测所述待推荐的券在转化过程中的点击率、领取率以及核销率中的至少两种;
第二预测单元,根据所述点击率、所述领取率以及所述核销率中的至少两种,预测所述待推荐的券的转化率。
第十一方面,提出一种电子设备,该电子设备包括:
处理器;以及
被安排成存储计算机可执行指令的存储器,该可执行指令在被执行时使该处理器执行以下操作:
确定待推荐的券以及用户的特征信息;
根据所述待推荐的券以及所述特征信息,预测所述待推荐的券在转化过程中的点击率、领取率以及核销率中的至少两种;
根据所述点击率、所述领取率以及所述核销率中的至少两种,预测所述待推荐的券的转化率。
第十二方面,提出一种计算机可读存储介质,所述计算机可读存储介质存储一个或多个程序,所述一个或多个程序当被包括多个应用程序的电子设备执行时,使得所述电子设备执行以下方法:
确定待推荐的券以及用户的特征信息;
根据所述待推荐的券以及所述特征信息,预测所述待推荐的券在转化过程中的点击率、领取率以及核销率中的至少两种;
根据所述点击率、所述领取率以及所述核销率中的至少两种,预测所述待推荐的券的转化率。
第十三方面,提出一种推荐券的方法,包括:
确定多个待推荐的券以及用户的特征信息;
根据所述多个待推荐的券以及所述特征信息,预测所述多个待推荐的券在转化过程中的点击率、领取率以及核销率中的至少两种;
根据所述点击率、所述领取率以及所述核销率中的至少两种,预测所述多个待推荐的券的转化率;
根据所述多个待推荐的券的转化率,向所述用户推荐券。
第十四方面,提出一种推荐券的装置,包括:
确定单元,确定多个待推荐的券以及用户的特征信息;
第一预测单元,根据所述多个待推荐的券以及所述特征信息,预测所述多个待推荐的券在转化过程中的点击率、领取率以及核销率中的至少两种;
第二预测单元,根据所述点击率、所述领取率以及所述核销率中的至少两种,预测所述多个待推荐的券的转化率;
推荐单元,根据所述多个待推荐的券的转化率,向所述用户推荐券。
第十五方面,提出一种电子设备,该电子设备包括:
处理器;以及
被安排成存储计算机可执行指令的存储器,该可执行指令在被执行时使该处理器执行以下操作:
确定多个待推荐的券以及用户的特征信息;
根据所述多个待推荐的券以及所述特征信息,预测所述多个待推荐的券在转化过程中的点击率、领取率以及核销率中的至少两种;
根据所述点击率、所述领取率以及所述核销率中的至少两种,预测所述多个待推荐的券的转化率;
根据所述多个待推荐的券的转化率,向所述用户推荐券。
第十六方面,提出一种计算机可读存储介质,所述计算机可读存储介质存储一个或多个程序,所述一个或多个程序当被包括多个应用程序的电子设备执行时,使得所述电子设备执行以下方法:
确定多个待推荐的券以及用户的特征信息;
根据所述多个待推荐的券以及所述特征信息,预测所述多个待推荐的券在转化过程中的点击率、领取率以及核销率中的至少两种;
根据所述点击率、所述领取率以及所述核销率中的至少两种,预测所述多个待推荐的券的转化率;
根据所述多个待推荐的券的转化率,向所述用户推荐券。
本申请实施例采用的上述至少一个技术方案能够达到以下有益效果:
本申请实施例提供的技术方案,确定待推荐信息以及用户的特征信息;根据所述待推荐信息以及所述特征信息,预测影响所述待推荐信息转化过程的至少两种因素对应的转化概率;根据所述至少两种因素对应的转化概率,预测所述待推荐信息的转化率。这样,由于在预测信息转化率时,可以根据影响信息转化过程的至少两种因素的转化概率进行预测,因此,预测得到的信息转化率可以更加全面的反映真实的信息转化率,从而提高信息转化率的预测准确度,进而可以有效地向用户进行信息推荐。
附图说明
为了更清楚地说明本申请实施例或现有技术中的技术方案,下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍,显而易见地,下面描述中的附图仅仅是本申请中记载的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动性的前提下,还可以根据这些附图获得其他的附图。
图1是本申请的一个实施例信息转化率的预测方法的流程示意图;
图2是本申请的一个实施例券转化率的预测和推荐券的示意图;
图3是本申请的一个实施例信息推荐方法的流程示意图;
图4是本申请的一个实施例券转化率的预测方法的流程示意图;
图5是本申请的一个实施例推荐券的方法的流程示意图;
图6是本申请的一个实施例电子设备的结构示意图;
图7是本申请的一个实施例信息转化率的预测装置的结构示意图;
图8是本申请的一个实施例电子设备的结构示意图;
图9是本申请的一个实施例信息推荐装置的结构示意图;
图10是本申请的一个实施例电子设备的结构示意图;
图11是本申请的一个实施例券转化率的预测装置的结构示意图;
图12是本申请的一个实施例电子设备的结构示意图;
图13是本申请的一个实施例推荐券的装置的结构示意图。
具体实施方式
信息的转化率可以理解为信息从曝光到核销的过程中的转化效率,通常,信息推荐方在向用户推荐信息之前,可以根据信息的点击率预测信息的转化率,以便于将转化率较高的信息推荐给用户。其中,信息的点击率越高,可以视为其转化率越高。然而,点击率通常代表用户对信息的点击或查看次数,通过点击率预测信息的转化率并不能有效、全面地体现信息的转化率,即根据点击率预测得到的转化率的准确率较低。
为了解决上述技术问题,本申请实施例提供一种信息转化率的预测、信息推荐方法和装置,该信息转化率的预测方法包括:确定待推荐信息以及用户的特征信息;根据所述待推荐信息以及所述特征信息,预测影响所述待推荐信息转化过程的至少两种因素对应的转化概率;根据所述至少两种因素对应的转化概率,预测所述待推荐信息的转化率。
这样,由于在预测信息转化率时,可以根据影响信息转化过程的至少两种因素的转化概率进行预测,并不是仅根据信息的点击率进行预测,因此,预测得到的信息转化率可以更加全面的反映真实的信息转化率,从而提高信息转化率的预测准确度,进而可以有效地向用户进行信息推荐。
为了使本技术领域的人员更好地理解本申请中的技术方案,下面将结合本申请实施例中的附图,对本申请实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例仅仅是本申请一部分实施例,而不是全部的实施例。基于本申请中的实施例,本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例,都应当属于本申请保护的范围。
需要说明的是,本申请实施例提供的技术方案可以应用于在已知用户的情况下,向该用户推荐信息。在向用户推荐信息之前,可以使用本申请实施例提供的技术方案确定需要推荐的信息的转化率,并根据转化率向用户推荐信息。其中,若待推荐的信息的个数为1,则可以根据转化率确定是否向用户推荐该信息,若待推荐的信息的个数为多个,则可以根据转化率确定向用户推荐哪一个信息。
以下结合附图,详细说明本申请各实施例提供的技术方案。
图1是本申请的一个实施例信息转化率的预测方法的流程示意图。所述方法如下所述。
S102:确定待推荐信息以及用户的特征信息。
在S102中,在用户浏览网页或打开某个应用时,网页服务端或应用的服务端(可以视为信息推荐方)可以基于该网页或应用向该用户进行信息推荐,该用户可以视为推荐信息的目标用户。在向用户推荐信息之前,可以确定向用户推荐的待推荐信息以及该用户的特征信息。
本申请实施例中,所述待推荐信息可以是准备向用户推荐,但最终不一定会推荐给用户的信息。所述待推荐信息可以是电子信息,具体可以是电子券、也可以网址链接等其他电子信息。所述待推荐信息的个数可以是一个,也可以是多个。本申请实施例在确定待推荐信息的转化率时,可以以确定其中一个待推荐信息的转化率为例进行说明。
用户的特征信息可以是用户的个人信息,例如,用户的身高、年龄、体重等,也可以是用户的行为数据,例如,用户经常去的饭馆、用户喜欢的运行等,这里不再一一举例说明。
用户的特征信息可以根据用户当前浏览的网页或打开的应用确定得到,例如,用户在浏览网页时,网站的服务端可以获取到用户的网页浏览记录等信息,若用户已登录网站,则还可以获取到用户在注册该网站时填写的个人信息。再例如,用户在打开应用时,应用的服务端可以获取到用户使用该应用的历史记录,若用户已登录该应用,还可以获取到用户的用户名等信息。
在确定向用户推荐的待推荐信息以及用户的特征信息后,可以执行S104。
S104:根据所述待推荐信息以及所述特征信息,预测影响所述待推荐信息转化过程的至少两种因素对应的转化概率。
在S104中,可以根据确定的待推荐信息以及用户的特征信息,预测影响待推荐信息转化过程的至少两种因素对应的转化概率,即预测得到影响待推荐信息转化率的至少两种转化概率。
本申请实施例中,待推荐信息在从曝光到核销的整个转化过程(即转化链路)依次可以包括:曝光待推荐信息、查看待推荐信息、获取待推荐信息以及使用待推荐信息。其中,曝光待推荐信息可以是信息推荐方向用户展示待推荐信息,查看待推荐信息可以是用户点击或浏览待推荐信息,获取待推荐信息可以是用户领取或获得待推荐信息,使用待推荐信息可以是用户对待推荐信息进行的最后操作,例如,若待推荐信息是优惠券,则使用待推荐信息可以是用户使用优惠券进行支付,若待推荐信息是网址链接,则使用待推荐信息可以是用户转发该网址链接,这里不再一一举例说明。
本申请实施例中,由于查看、获取以及使用待推荐信息会影响待推荐信息的转化率,因此,影响待推荐信息转化过程的至少两种因素可以包括:查看待推荐信息、获取待推荐信息以及使用待推荐信息中的至少两种。这样,影响待推荐信息转化过程的至少两种转化概率可以包括:查看待推荐信息的第一概率、获取待推荐信息的第二概率以及使用待推荐信息的第三概率这三种转化概率中的至少两种。
查看待推荐信息的第一概率可以是,将待推荐信息推荐给N个用户后,N个用户中查看待推荐信息的用户个数M与N的比值;获取待推荐信息的第二概率可以是,查看待推荐信息的M个用户中,获取待推荐信息的用户个数L与M的比值;使用待推荐信息的第三概率可以是获取待推荐信息的L个用户中,使用待推荐信息的用户个数P与L的比值。其中,M、N、L个P均为非负整数。
本申请实施例中,根据所述待推荐信息以及所述特征信息,预测影响所述待推荐信息转化过程的至少两种因素对应的转化概率,可以包括:
确定预先训练得到的转化率模型,所述转化率模型包括第一概率模型、第二概率模型以及第三概率模型中的至少两种;
根据所述待推荐信息以及所述特征信息,通过所述转化率模型确定所述第一概率、所述第二概率以及所述第三概率中的至少两种。
本申请实施例可以预先训练得到用于预测待推荐信息的转化率模型,所述转化率模型具体可以包括第一概率模型,第二概率模型以及第三概率模型中的至少两种,其中,所述第一概率模型用于预测待推荐信息的第一概率,所述第二概率模型用于预测待 推荐信息的第二概率,所述第三概率模型用于预测待推荐信息的第三概率。
这样,在确定待推荐信息以及用户的特征信息后,可以基于转化率模型中包含的第一概率模型、第二概率模型以及第三概率模型中的至少两种,以待推荐信息的属性信息以及推荐方式、用户的特征信息作为输入,分别得到待推荐信息的第一概率、第二概率第三概率中的至少两种。
所述待推荐信息的属性信息可以包括待推荐信息的类别(例如,待推荐信息是网络链接或抵价券等)、待推荐信息的金额,待推荐信息所属的信息推荐方等,所述待推荐信息的推荐方式可以包括短信、邮件、网址链接或应用的通知信息等,不做具体限定。在确定所述待推荐信息时,可以确定得到所述待推荐信息的属性信息以及推荐方式。
本申请实施例中,在训练得到所述转化率模型时,具体可以通过以下方式训练得到,包括:
获取所述待推荐信息的历史数据,所述历史数据包括:所述待推荐信息的属性信息,所述待推荐信息的查看数据、获取数据以及使用数据中的至少两种,所述查看数据包括点击所述待推荐信息的用户的特征信息,所述获取数据包括获取所述待推荐信息的用户的特征信息,所述使用数据包括使用所述待推荐信息的用户的特征信息;
基于预设模型对所述历史数据进行训练,得到所述转化率模型。
以下以待推荐信息为券这一具体场景为例详细说明。其中,券可以是电子券,具体可以是抵价券、满减券等,这里不做具体限定。
在待推荐信息为券时,影响券转化过程的至少两种因素可以包括:点击券、领取券以及核销券中的至少两种,影响券转化过程的至少两种转化概率可以包括:券的点击率(对应于查看待推荐信息的第一概率)、券的领取率(对应于获取待推荐信息的第二概率)以及券的核销率(对应于使用待推荐信息的第三概率)中的至少两种。
用于预测得到券转化率的转化率模型可以包括:点击率模型(对应于第一概率模型)、领取率模型(对应于第二概率模型)以及核销率模型(对应于第三概率模型)中的至少两种。
在训练得到券的转化率模型时,具体地,首先,可以获取券的历史数据。券的历史数据可以包括券的属性信息,所述属性信息可以是券的类别,例如,是抵价券还是满减券,也可以是券的还可以是券的金额,还可以是券所属的信息推荐方等,这里不再一一举例说明。
券的历史数据也可以包括券的推荐方式,所述推荐方式可以是短信、邮件、网址链接或应用的通知信息等,不做具体限定。
券的历史数据还可以包括在不同的属性信息和推荐方式下,券的点击数据(对应于待推荐信息的查看数据)、券的领取数据(对应于待推荐信息的获取数据)以及券的核销数据(对应于待推荐信息的使用数据)中的至少两种。其中,券的点击数据可以包括点击券的用户的特征信息,该特征信息可以是点击券的用户的个人信息,例如用户的身高、体重、年龄等,也可以是点击券的用户的行为数据,例如,用户喜欢去的饭馆、喜欢的运动等。
券的领取数据可以包括领取券的用户的特征信息,该特征信息可以包括领取券的用户的个人信息以及行为数据,具体可以参见上述对点击券的用户的个人信息以及行为数据的描述,这里不再重复描述。同样的,券的核销数据可以包括核销券的用户的特征信息,该特征信息可以包括核销券的用户的个人信息以及行为数据。
其次,在获取得到券的上述历史数据后,可以基于预设模型对所述历史数据进行训练,并得到转化率模型。其中,所述预设模型可以是深度学习模型,也可以是神经网络模型等,具体可以根据实际需要确定,这里不做具体限定,例如,如果根据实际情况确定基于深度学习模型进行训练的结果更佳,则可以使用深度学习模型进行训练。
以转化率模型包含点击率模型、领取率模型以及核销率模型为例,在基于预设模型对上述历史数据训练得到转化率模型时,可以分别训练得到点击率模型、领取率模型以及核销率模型。其中,训练点击率模型、领取率模型以及核销率模型时使用的预设模型可以相同,也可以不同。
以训练点击率模型为例,具体地,可以将历史数据中包含的券的属性信息、推荐方式以及在不同的推荐方式下点击券的用户的特征信息作为输入变量,将用户对不同属性信息的券在不同推荐方式下的点击率作为输出变量,基于预设模型进行训练,最终得到用于确定券点击率的点击率模型。其中,在基于预设模型进行训练时,具体可以使用机器学习模型,通过不断的数据迭代训练得到点击率模型。
基于上述记载的方法训练得到券的点击率模型、领取率模型以及核销率模型后,可以确定得到券的转化率模型。
这样,在确定券的转化率模型后,在确定待推荐的券以及用户的特征信息后,可以基于券的转化率模型确定影响券转化过程的点击率、领取率以及核销率中的至少两 种。
在基于上述记载的方法确定待推荐信息的转化率模型,并根据转化率模型预测影响待推荐信息转化过程的至少两种因素对应的转化概率后,可以执行步骤S106。
S106:根据所述至少两种因素对应的转化概率,预测所述待推荐信息的转化率。
在S106中,在预测得到影响待推荐信息转化过程的至少两种因素对应的转化概率后,可以进一步预测得到待推荐信息的转化率。
本申请实施例中,根据所述至少两种因素对应的转化概率,预测所述待推荐信息的转化率,可以包括:
将预测的所述第一概率、所述第二概率以及所述第三概率中至少两种概率的乘积确定为所述待推荐信息的转化率。
以下仍以待推荐信息为券这一具体场景为例详细说明。
具体地,若预测得到券的点击率和核销率,则可以将券的点击率与核销率的乘积确定为券的转化率;若预测得到券的领取率和核销率,则可以将券的领取率与核销率的乘积确定为券的转化率;若预测得到券的点击率、领取率和核销率,则可以将券的点击率、领取率与核销率的乘积确定为券的转化率,当然,也可以将其中任意两个转化概率的乘积作为券的转化率。
需要说明的是,由于在实际应用中,券的核销率对券的转化率的影响比较大,因此,在预测影响券转化过程的至少两种转化概率时,至少两种转化概率中可以优先包括券的核销率,在进一步预测券的转化率时,可以将券的核销率作为预测券的转化率的必要条件,并将其与券的点击率以及领取率中的至少一种的乘积作为券的转化率。
此外,为了更准确地预测得到券的转化率,可以预测影响券转化过程的三种因素对应的转化概率,即点击率、领取率和核销率,并将预测的券的点击率、领取率和核销率的乘积作为券的转化率。
在本申请的另一实施例中,在预测得到待推荐信息的转化率后,所述方法还包括:
根据所述待推荐信息的转化率,向所述用户进行信息推荐。
在向用户进行信息推荐时,如果待推荐信息的个数为1,则可以确定是否向用户推荐所述待推荐信息。例如,待推荐信息的转化率高,则可以向用户推荐该信息,反之, 则可以不推荐该信息。
如果待推荐信息的个数为多个,则根据所述待推荐信息的转化率,向所述用户进行信息推荐,可以包括:
根据所述转化率的大小对多个所述待推荐信息进行排序,并将转化率最大的待推荐信息推荐给所述用户;或,
分别按照第一概率、第二概率以及第三概率中的至少两种对多个所述待推荐信息进行排序;确定第一概率、第二概率以及第三概率中至少两种均在设定排序范围内的待推荐信息,并将确定的待推荐信息中转化率最大的待推荐信息推荐给所述用户。
也就是说,如果待推荐信息的个数为多个,则可以对待推荐信息进行排序,根据排序结果选择其中一个待推荐信息推荐给用户。以待推荐的信息为券进行说明,对待推荐的多个券的具体排序方法至少可以包含以下两种:
一种方法是根据转化率的大小,将多个券进行排序,选择转化率最大的其中一个券推荐给用户。其中,若多个券的转化率相同,则可以随机选择其中一个券推荐给用户,或按照预设规则选择其中一个券推荐给用户,例如,可以选择金额较大的券推荐给用户。当然,也可以将转化率相同的券全部推荐给用户,由用户根据实际情况确定使用哪个。
另一种方法是在预测得到券的点击率、领取率以及核销率中的至少两种后,可以分别根据点击率、领取率以及核销率对多个券进行排序,确定点击率、领取率以及核销率中的至少两种均在设定排序范围内的券,并将这些券中转化率最大的券推荐给用户。
例如,待推荐的券有六张,分别为A、B、C、D、E、F,预测得到的六张券的点击率、领取率以及核销率分别为:a1、a2、a3(对应A的点击率、领取率和核销率),b1、b2、b3,c1、c2、c3,d1、d2、d3,e1、e2、e3,f1、f2、f3,其中,根据点击率的大小对六张券进行排序的结果为:b1、a1、d1、c1、e1、f1,根据领取率的大小对六张券进行排序的结果为:d2、b2、f2、e1、c1、a1,根据核销率的大小对六张券进行排序的结果为:c3、b3、d3、f3、a3、e1。可知,券B和券D的点击率、领取率以及核销率的排序均在前三,那么,可以选择券B和券D中转化率最大的券推荐给用户。
在基于上述任一种方法确定向用户推荐的券后,可以将该券以对应的推荐方式推荐给用户。
图2是本申请的一个实施例券转化率的预测和推荐券的示意图。图2以根据券 的点击率、领取率以及核销率预测券的转化率为例进行说明。
图2中,在向用户推荐券之前,可以根据券的历史点击数据、历史领取数据以及历史核销数据,分别训练得到券的点击率模型、领取率模型以及核销率模型。其中,以历史点击数据为例,可以基于预设的机器学习模型A对历史点击数据进行模型训练,最终得到点击率模型,具体方法可以参见上述记载的训练得到点击率模型的方法,这里不再重复描述。
同样,可以基于预设的机器学习模型B对历史领取数据进行训练,得到领取率模型,基于预设的及其学习模型C对历史核销数据进行训练,得到核销率模型。
在得到点击率模型、领取率模型以及核销率模型后,可以根据准备推荐给用户的券的属性信息以及推荐方式,用户的特征信息,基于点击率模型预测券的点击率,根据领取率模型预测领取率模型,根据核销率模型预测券的核销率。之后,可以将预测得到的券的点击率、领取率以及核销率的乘积作为券的转化率。
图2中,准备向用户推荐的券的个数为多个,在按照上述记载的方法确定多个券的转化率后,可以按照转化率的大小对多个券排序,得到转化率最大的券,并将该券以对应的曝光渠道推荐给用户。例如,该券是与应用相关的券,则可以以应用的通知信息的方式将券推荐给用户。
例如,计划向用户推荐十个满减券A、B、C、D、E、F、G、H、I和J,基于图2所示的方法预测得到十个满减券的转化率依次为:a、b、c、d、e、f、g、h、i和j,其中,d的值最大,那么,可以以应用的通知消息的方式向用户推荐满减券D。
应理解,图2中还可以仅预测券的点击率和核销率,根据点击率和核销率预测券的转化率,或预测券的领取率和核销率,根据领取率和核销率预测券的转化率。在对多个券进行排序时,也可以分别按照点击率、领取率以及核销率中的至少两种对多个券进行排序,确定点击率、领取率以及核销率中至少两种均在设定排序范围内的券,并将确定的券中转化率最大的券推荐给用户,这里不再重复描述。
本申请实施例提供的技术方案,确定待推荐信息以及用户的特征信息;根据所述待推荐信息以及所述特征信息,预测影响所述待推荐信息转化过程的至少两种因素对应的转化概率;根据所述至少两种因素对应的转化概率,预测所述待推荐信息的转化率。这样,由于在预测信息转化率时,可以根据影响信息转化过程的至少两种因素的转化概率进行预测,因此,预测得到的信息转化率可以更加全面的反映真实的信息转化率,从 而提高信息转化率的预测准确度,进而可以有效地向用户进行信息推荐。
图3是本申请的一个实施例信息推荐方法的流程示意图。本申请实施例以准备向用户推荐多个待推荐信息为例进行说明。所述信息推荐方法如下所述。
S302:确定多个待推荐信息以及用户的特征信息。
S304:根据所述多个待推荐信息以及所述特征信息,预测影响所述多个待推荐信息转化过程的至少两种因素对应的转化概率。
在S304中,可以分别针对其中一个待推荐信息,预测影响待推荐信息转化过程的至少两种因素对应的转化概率。
S306:根据所述至少两种因素对应的转化概率,预测所述多个待推荐信息的转化率。
S308:根据所述多个待推荐信息的转化率,向所述用户进行信息推荐。
在本申请的一个实施例中,根据所述多个待推荐信息的转化率,向所述用户进行信息推荐,可以包括:
根据所述转化率的大小对多个所述待推荐信息进行排序,并将转化率最大的待推荐信息推荐给所述用户。
在本申请的另一个实施例中,影响所述多个待推荐信息转化过程的至少两种因素对应的转化概率包括:查看所述待推荐信息的第一概率、获取所述待推荐信息的第二概率以及使用所述待推荐信息的第三概率中的至少两种;
根据所述多个待推荐信息的转化率,向所述用户进行信息推荐,包括:
分别按照第一概率、第二概率以及第三概率中的至少两种对所述多个待推荐信息进行排序;确定第一概率、第二概率以及第三概率中至少两种均在设定排序范围内的待推荐信息,并将确定的待推荐信息中转化率最大的待推荐信息推荐给所述用户。
图3所示实施例相关步骤的具体实现可参考图1所示实施例中对应的步骤的具体实现,本说明书一个或多个实施例在此不再赘述。
本申请实施例提供的技术方案,确定多个待推荐信息以及用户的特征信息;根据所述多个待推荐信息以及所述特征信息,预测影响所述多个待推荐信息转化过程的至少两种因素对应的转化概率;根据所述至少两种因素对应的转化概率,预测所述多个待推荐信息的转化率;根据所述多个待推荐信息的转化率,向所述用户进行信息推荐。这 样,由于可以根据影响信息转化过程的至少两种因素的转化概率预测得到信息的转化率,因此,可以预测得到较高准确度的信息转化率,进而可以有效地向用户进行信息推荐。
以上举例说明了信息转化率的预测方法以及信息推荐的实施过程。为更直观的说明本申请实施例的设计构思和技术效果,下面以上述信息转化率的预测方法和信息推荐方法应用在券这一场景为例,说明本申请实施例的实施过程。具体地,待推荐信息可以为券,影响券转化过程的至少两种因素包括:点击券、领取券以及核销券中的至少两种,影响券转化过程的至少两种转化概率包括:点击率、领取率以及核销率中的至少两种。以上关于信息转化率的预测方法以及信息推荐方法的阐释与说明均可应用于以下券转化率的预测以及推荐券这一具体应用场景,重复内容或不再赘述。
图4是本申请的一个实施例券转化率的预测方法的流程示意图。本申请实施例中,券可以是电子券,具体可以是优惠券、抵价券、满减券等。所述券转化率的预测方法如下所述。
S402:确定待推荐的券以及用户的特征信息。
S404:根据所述待推荐的券以及所述特征信息,预测所述待推荐的券在转化过程中的点击率、领取率以及核销率中的至少两种。
S406:根据所述点击率、所述领取率以及所述核销率中的至少两种,预测所述待推荐的券的转化率。
进一步地,根据所述待推荐的券以及所述特征信息,预测所述待推荐的券在转化过程中的点击率、领取率以及核销率中的至少两种,包括:
确定预先训练得到的转化率模型,所述转化率模型包括点击率模型、领取率模型以及核销率模型中的至少两种;
根据所述待推荐的券以及所述特征信息,通过所述转化率模型确定所述点击率、所述领取率以及所述核销率中的至少两种。
本申请实施例中,所述转化率模型通过以下方式训练得到,包括:
获取所述待推荐的券的历史数据,所述历史数据包括:所述待推荐的券的推荐方式,在所述推荐方式下所述待推荐的券的点击数据、领取数据以及核销数据中的至少两种,所述点击数据包括点击所述待推荐的券的用户的特征信息,所述领取数据包括领取所述待推荐的券的用户的特征信息,所述核销数据包括使用所述待推荐的券的用户的 特征信息;
基于预设模型对所述历史数据进行训练,得到所述转化率模型。
进一步地,根据所述点击率、所述领取率以及所述核销率中的至少两种,预测所述待推荐的券的转化率,包括:
将所述点击率、所述领取率以及所述核销率中至少两种的乘积确定为所述待推荐的券的转化率,其中:
若预测得到所述点击率以及所述核销率,则将所述点击率以及所述核销率的乘积确定为所述待推荐的券的转化率;
若预测得到所述领取率以及所述核销率,则将所述领取率以及所述核销率的乘积确定为所述待推荐的券的转化率;
若预测得到所述点击率、所述领取率以及所述核销率,则将所述点击率、所述领取率以及所述核销率的乘积确定为所述待推荐的券的转化率。
更进一步地,在预测得到所述待推荐的券的转化率后,所述方法还包括:
根据所述转化率向所述用户推荐券。
具体地,根据所述转化率向所述用户推荐券,包括:
若所述待推荐的券的个数为多个,则:
根据所述转化率的大小对多个所述待推荐的券进行排序,并将转化率最大的券推荐给所述用户;或,
分别按照点击率、领取率以及核销率中的至少两种对多个所述待推荐的券进行排序;确定点击率、领取率以及核销率中至少两种均在设定排序范围内的券,将确定的券中转化率最大的券推荐给所述用户。
图4所示实施例相关步骤的具体实现可参考图1所示实施例中对应的步骤的具体实现,本说明书一个或多个实施例在此不再赘述。
本申请实施例提供的技术方案,确定待推荐的券以及用户的特征信息;根据所述待推荐的券以及所述特征信息,预测所述待推荐的券在转化过程中的点击率、领取率以及核销率中的至少两种;根据所述点击率、所述领取率以及所述核销率中的至少两种,预测所述待推荐的券的转化率。这样,由于在预测券的转化率时,可以根据影响券转化 过程的点击率、领取率以及核销率中的至少两种转化概率进行预测,因此,预测得到的券转化率可以更加全面的反映真实的券转化率,从而提高券转化率的预测准确度,进而可以有效地向用户推荐券。
图5是本申请的一个实施例推荐券的方法的流程示意图。所述推荐券的方法如下所述。
S502:确定多个待推荐的券以及用户的特征信息。
S504:根据所述多个待推荐的券以及所述特征信息,预测所述多个待推荐的券在转化过程中的点击率、领取率以及核销率中的至少两种。
S506:根据所述点击率、所述领取率以及所述核销率中的至少两种,预测所述多个待推荐的券的转化率.
S508:根据所述多个待推荐的券的转化率,向所述用户推荐券。
进一步地,根据所述多个待推荐的券的转化率,向所述用户推荐券,包括:
根据转化率的大小对所述多个待推荐的券进行排序,并将转化率最大的券推荐给所述用户;或,
分别按照点击率、领取率以及核销率中的至少两种对所述多个待推荐的券进行排序;确定点击率、领取率以及核销率中至少两种均在设定排序范围内的券,将确定的券中转化率最大的券推荐给所述用户。
图5所示实施例相关步骤的具体实现可参考图1所示实施例中对应的步骤的具体实现,本说明书一个或多个实施例在此不再赘述。
本申请实施例提供的技术方案,确定多个待推荐的券以及用户的特征信息;根据所述多个待推荐的券以及所述特征信息,预测所述多个待推荐的券在转化过程中的点击率、领取率以及核销率中的至少两种;根据所述点击率、所述领取率以及所述核销率中的至少两种,预测所述多个待推荐的券的转化率;根据所述多个待推荐的券的转化率,向所述用户推荐券。这样,在向用户推荐券时,由于可以根据影响券转化过程的点击率、领取率以及核销率中的至少两种转化概率预测得到券的转化率,因此,可以预测得到较高准确度的券转化率,进而可以有效地向用户推荐券。
上述对本说明书特定实施例进行了描述。其它实施例在所附权利要求书的范围内。在一些情况下,在权利要求书中记载的动作或步骤可以按照不同于实施例中的顺序 来执行并且仍然可以实现期望的结果。另外,在附图中描绘的过程不一定要求示出的特定顺序或者连续顺序才能实现期望的结果。在某些实施方式中,多任务处理和并行处理也是可以的或者可能是有利的。
图6是本申请的一个实施例电子设备的结构示意图。请参考图6,在硬件层面,该电子设备包括处理器,可选地还包括内部总线、网络接口、存储器。其中,存储器可能包含内存,例如高速随机存取存储器(Random-Access Memory,RAM),也可能还包括非易失性存储器(non-volatile memory),例如至少1个磁盘存储器等。当然,该电子设备还可能包括其他业务所需要的硬件。
处理器、网络接口和存储器可以通过内部总线相互连接,该内部总线可以是ISA(Industry Standard Architecture,工业标准体系结构)总线、PCI(Peripheral Component Interconnect,外设部件互连标准)总线或EISA(Extended Industry Standard Architecture,扩展工业标准结构)总线等。所述总线可以分为地址总线、数据总线、控制总线等。为便于表示,图6中仅用一个双向箭头表示,但并不表示仅有一根总线或一种类型的总线。
存储器,用于存放程序。具体地,程序可以包括程序代码,所述程序代码包括计算机操作指令。存储器可以包括内存和非易失性存储器,并向处理器提供指令和数据。
处理器从非易失性存储器中读取对应的计算机程序到内存中然后运行,在逻辑层面上形成信息转化率的预测装置。处理器,执行存储器所存放的程序,并具体用于执行以下操作:
确定待推荐信息以及用户的特征信息;
根据所述待推荐信息以及所述特征信息,预测影响所述待推荐信息转化过程的至少两种因素对应的转化概率;
根据所述至少两种因素对应的转化概率,预测所述待推荐信息的转化率。
上述如本申请图6所示实施例揭示的信息转化率的预测装置执行的方法可以应用于处理器中,或者由处理器实现。处理器可能是一种集成电路芯片,具有信号的处理能力。在实现过程中,上述方法的各步骤可以通过处理器中的硬件的集成逻辑电路或者软件形式的指令完成。上述的处理器可以是通用处理器,包括中央处理器(Central Processing Unit,CPU)、网络处理器(Network Processor,NP)等;还可以是数字信号处理器(Digital Signal Processor,DSP)、专用集成电路(Application Specific Integrated Circuit,ASIC)、现场可编程门阵列(Field-Programmable Gate Array,FPGA)或者其 他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件。可以实现或者执行本申请实施例中的公开的各方法、步骤及逻辑框图。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。结合本申请实施例所公开的方法的步骤可以直接体现为硬件译码处理器执行完成,或者用译码处理器中的硬件及软件模块组合执行完成。软件模块可以位于随机存储器,闪存、只读存储器,可编程只读存储器或者电可擦写可编程存储器、寄存器等本领域成熟的存储介质中。该存储介质位于存储器,处理器读取存储器中的信息,结合其硬件完成上述方法的步骤。
该电子设备还可执行图1的方法,并实现信息转化率的预测装置在图1所示实施例中的功能,本申请实施例在此不再赘述。
当然,除了软件实现方式之外,本申请的电子设备并不排除其他实现方式,比如逻辑器件抑或软硬件结合的方式等等,也就是说以下处理流程的执行主体并不限定于各个逻辑单元,也可以是硬件或逻辑器件。
本申请实施例还提出了一种计算机可读存储介质,该计算机可读存储介质存储一个或多个程序,该一个或多个程序包括指令,该指令当被包括多个应用程序的便携式电子设备执行时,能够使该便携式电子设备执行图1所示实施例的方法,并具体用于执行以下操作:
确定待推荐信息以及用户的特征信息;
根据所述待推荐信息以及所述特征信息,预测影响所述待推荐信息转化过程的至少两种因素对应的转化概率;
根据所述至少两种因素对应的转化概率,预测所述待推荐信息的转化率。
图7是本申请的一个实施例信息转化率的预测装置70的结构示意图。请参考图7,在一种软件实施方式中,所述信息转化率的预测装置70可包括:确定单元71、第一预测单元72和第二预测单元73,其中:
确定单元71,确定待推荐信息以及用户的特征信息;
第一预测单元72,根据所述待推荐信息以及所述特征信息,预测影响所述待推荐信息转化过程的至少两种因素对应的转化概率;
第二预测单元73,根据所述至少两种因素对应的转化概率,预测所述待推荐信息的转化率。
可选地,影响所述待推荐信息转化过程的至少两种因素包括:查看所述待推荐信息、获取所述待推荐信息以及使用所述待推荐信息中的至少两种;
所述至少两种因素对应的转化概率包括:查看所述待推荐信息的第一概率、获取所述待推荐信息的第二概率以及使用所述待推荐信息的第三概率中的至少两种。
可选地,所述第一预测单元72,根据所述待推荐信息以及所述特征信息,预测影响所述待推荐信息转化过程的至少两种因素对应的转化概率,包括:
确定预先训练得到的转化率模型,所述转化率模型包括第一概率模型、第二概率模型以及第三概率模型中的至少两种;
根据所述待推荐信息以及所述特征信息,通过所述转化率模型确定所述第一概率、所述第二概率以及所述第三概率中的至少两种。
可选地,所述第一预测单元72通过以下方式训练得到所述转化率模型,包括:
获取所述待推荐信息的历史数据,所述历史数据包括:所述待推荐信息的属性信息和推荐方式,在不同的属性信息和推荐方式下所述待推荐信息的查看数据、获取数据以及使用数据中的至少两种,所述查看数据包括点击所述待推荐信息的用户的特征信息,所述获取数据包括获取所述待推荐信息的用户的特征信息,所述使用数据包括使用所述待推荐信息的用户的特征信息;
基于预设模型对所述历史数据进行训练,得到所述转化率模型。
可选地,所述第二预测单元73,根据所述至少两种因素对应的转化概率,预测所述待推荐信息的转化率,包括:
将预测的所述第一概率、所述第二概率以及所述第三概率中至少两种概率的乘积确定为所述待推荐信息的转化率,其中:
若预测得到所述第一概率以及所述第三概率,则将所述第一概率以及所述第二概率的乘积确定为所述待推荐信息的转化率;
若预测得到所述第二概率以及所述第三概率,则将所述第二概率以及所述第三概率的乘积确定为所述待推荐信息的转化率;
若预测得到所述第一概率、所述第二概率以及所述第三概率,则将所述第一概率、所述第二概率以及所述第三概率的乘积确定为所述待推荐信息的转化率。
可选地,所述信息转化率的预测装置70还包括:推荐单元74,其中:
所述推荐单元74在所述第二预测单元73预测得到所述待推荐信息的转化率后,根据所述待推荐信息的转化率,向所述用户进行信息推荐。
可选地,所述推荐单元74,根据所述待推荐信息的转化率,向所述用户进行信息推荐,包括:
若所述待推荐信息的个数为多个,则:
根据所述转化率的大小对多个所述待推荐信息进行排序,并将转化率最大的待推荐信息推荐给所述用户;或,
分别按照第一概率、第二概率以及第三概率中的至少两种对多个所述待推荐信息进行排序;确定第一概率、第二概率以及第三概率中至少两种均在设定排序范围内的待推荐信息,并将确定的待推荐信息中转化率最大的待推荐信息推荐给所述用户。
信息转化率的预测装置70还可执行图1或图2的方法,并实现信息转化率的预测装置在图1、图2所示实施例的功能,本申请实施例在此不再赘述。
图8是本申请的一个实施例电子设备的结构示意图。请参考图8,在硬件层面,该电子设备包括处理器,可选地还包括内部总线、网络接口、存储器。其中,存储器可能包含内存,例如高速随机存取存储器(Random-Access Memory,RAM),也可能还包括非易失性存储器(non-volatile memory),例如至少1个磁盘存储器等。当然,该电子设备还可能包括其他业务所需要的硬件。
处理器、网络接口和存储器可以通过内部总线相互连接,该内部总线可以是ISA(Industry Standard Architecture,工业标准体系结构)总线、PCI(Peripheral Component Interconnect,外设部件互连标准)总线或EISA(Extended Industry Standard Architecture,扩展工业标准结构)总线等。所述总线可以分为地址总线、数据总线、控制总线等。为便于表示,图8中仅用一个双向箭头表示,但并不表示仅有一根总线或一种类型的总线。
存储器,用于存放程序。具体地,程序可以包括程序代码,所述程序代码包括计算机操作指令。存储器可以包括内存和非易失性存储器,并向处理器提供指令和数据。
处理器从非易失性存储器中读取对应的计算机程序到内存中然后运行,在逻辑层面上形成信息推荐装置。处理器,执行存储器所存放的程序,并具体用于执行以下操作:
确定多个待推荐信息以及用户的特征信息;
根据所述多个待推荐信息以及所述特征信息,预测影响所述多个待推荐信息转化过程的至少两种因素对应的转化概率;
根据所述至少两种因素对应的转化概率,预测所述多个待推荐信息的转化率;
根据所述多个待推荐信息的转化率,向所述用户进行信息推荐。
上述如本申请图8所示实施例揭示的信息推荐装置执行的方法可以应用于处理器中,或者由处理器实现。处理器可能是一种集成电路芯片,具有信号的处理能力。在实现过程中,上述方法的各步骤可以通过处理器中的硬件的集成逻辑电路或者软件形式的指令完成。上述的处理器可以是通用处理器,包括中央处理器(Central Processing Unit,CPU)、网络处理器(Network Processor,NP)等;还可以是数字信号处理器(Digital Signal Processor,DSP)、专用集成电路(Application Specific Integrated Circuit,ASIC)、现场可编程门阵列(Field-Programmable Gate Array,FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件。可以实现或者执行本申请实施例中的公开的各方法、步骤及逻辑框图。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。结合本申请实施例所公开的方法的步骤可以直接体现为硬件译码处理器执行完成,或者用译码处理器中的硬件及软件模块组合执行完成。软件模块可以位于随机存储器,闪存、只读存储器,可编程只读存储器或者电可擦写可编程存储器、寄存器等本领域成熟的存储介质中。该存储介质位于存储器,处理器读取存储器中的信息,结合其硬件完成上述方法的步骤。
该电子设备还可执行图3的方法,并实现信息推荐装置在图3所示实施例中的功能,本申请实施例在此不再赘述。
当然,除了软件实现方式之外,本申请的电子设备并不排除其他实现方式,比如逻辑器件抑或软硬件结合的方式等等,也就是说以下处理流程的执行主体并不限定于各个逻辑单元,也可以是硬件或逻辑器件。
本申请实施例还提出了一种计算机可读存储介质,该计算机可读存储介质存储一个或多个程序,该一个或多个程序包括指令,该指令当被包括多个应用程序的便携式电子设备执行时,能够使该便携式电子设备执行图1所示实施例的方法,并具体用于执行以下操作:
确定多个待推荐信息以及用户的特征信息;
根据所述多个待推荐信息以及所述特征信息,预测影响所述多个待推荐信息转 化过程的至少两种因素对应的转化概率;
根据所述至少两种因素对应的转化概率,预测所述多个待推荐信息的转化率;
根据所述多个待推荐信息的转化率,向所述用户进行信息推荐。
图9是本申请的一个实施例信息推荐预测装置90的结构示意图。请参考图9,在一种软件实施方式中,所述信息推荐装置90可包括:确定单元91、第一预测单元92、第二预测单元93以及推荐单元94,其中:
确定单元91,确定多个待推荐信息以及用户的特征信息;
第一预测单元92,根据所述多个待推荐信息以及所述特征信息,预测影响所述多个待推荐信息转化过程的至少两种因素对应的转化概率;
第二预测单元93,根据所述至少两种因素对应的转化概率,预测所述多个待推荐信息的转化率;
推荐单元94,根据所述多个待推荐信息的转化率,向所述用户进行信息推荐。
可选地,所述推荐单元94,根据所述多个待推荐信息的转化率,向所述用户进行信息推荐,包括:
根据所述转化率的大小对多个所述待推荐信息进行排序,并将转化率最大的待推荐信息推荐给所述用户。
可选地,影响所述多个待推荐信息转化过程的至少两种因素对应的转化概率包括:查看所述待推荐信息的第一概率、获取所述待推荐信息的第二概率以及使用所述待推荐信息的第三概率中的至少两种;
所述推荐单元94,根据所述多个待推荐信息的转化率,向所述用户进行信息推荐,包括:
分别按照第一概率、第二概率以及第三概率中的至少两种对所述多个待推荐信息进行排序;确定第一概率、第二概率以及第三概率中至少两种均在设定排序范围内的待推荐信息,并将确定的待推荐信息中转化率最大的待推荐信息推荐给所述用户。
信息推荐装置90还可执行图3的方法,并实现信息推荐装置在图3所示实施例中的功能,本申请实施例在此不再赘述。
图10是本申请的一个实施例电子设备的结构示意图。请参考图10,在硬件层面, 该电子设备包括处理器,可选地还包括内部总线、网络接口、存储器。其中,存储器可能包含内存,例如高速随机存取存储器(Random-Access Memory,RAM),也可能还包括非易失性存储器(non-volatile memory),例如至少1个磁盘存储器等。当然,该电子设备还可能包括其他业务所需要的硬件。
图10中,处理器从非易失性存储器中读取对应的计算机程序到内存中然后运行,在逻辑层面上形成券转化率的预测装置。处理器,执行存储器所存放的程序,并具体用于执行以下操作:
确定待推荐的券以及用户的特征信息;
根据所述待推荐的券以及所述特征信息,预测所述待推荐的券在转化过程中的点击率、领取率以及核销率中的至少两种;
根据所述点击率、所述领取率以及所述核销率中的至少两种,预测所述待推荐的券的转化率。
上述如本申请图10所示实施例揭示的券转化率的预测装置执行的方法可以应用于处理器中,或者由处理器实现。此外,图10中其他硬件的具体连接结构以及实现的功能可以参见图6所示的电子设备中的相关记载,这里不再重复描述。
该电子设备还可执行图4的方法,并实现券转化率的预测装置在图4所示实施例中的功能,本申请实施例在此不再赘述。
当然,除了软件实现方式之外,本申请的电子设备并不排除其他实现方式,比如逻辑器件抑或软硬件结合的方式等等,也就是说以下处理流程的执行主体并不限定于各个逻辑单元,也可以是硬件或逻辑器件。
本申请实施例还提出了一种计算机可读存储介质,该计算机可读存储介质存储一个或多个程序,该一个或多个程序包括指令,该指令当被包括多个应用程序的便携式电子设备执行时,能够使该便携式电子设备执行图4所示实施例的方法,并具体用于执行以下操作:
确定待推荐的券以及用户的特征信息;
根据所述待推荐的券以及所述特征信息,预测所述待推荐的券在转化过程中的点击率、领取率以及核销率中的至少两种;
根据所述点击率、所述领取率以及所述核销率中的至少两种,预测所述待推荐 的券的转化率。
图11是本申请的一个实施例券转化率的预测装置110的结构示意图。请参考图11,在一种软件实施方式中,所述券转化率的预测装置110可包括:确定单元111、第一预测单元112和第二预测单元113,其中:
确定单元111,确定待推荐的券以及用户的特征信息;
第一预测单元112,根据所述待推荐的券以及所述特征信息,预测所述待推荐的券在转化过程中的点击率、领取率以及核销率中的至少两种;
第二预测单元113,根据所述点击率、所述领取率以及所述核销率中的至少两种,预测所述待推荐的券的转化率。
可选地,所述第一预测单元112,根据所述待推荐的券以及所述特征信息,预测所述待推荐的券在转化过程中的点击率、领取率以及核销率中的至少两种,包括:
确定预先训练得到的转化率模型,所述转化率模型包括点击率模型、领取率模型以及核销率模型中的至少两种;
根据所述待推荐的券以及所述特征信息,通过所述转化率模型确定所述点击率、所述领取率以及所述核销率中的至少两种。
可选地,所述第一预测单元112通过以下方式训练得到所述转化率模型,包括:
获取所述待推荐的券的历史数据,所述历史数据包括:所述待推荐的券的推荐方式,在所述推荐方式下所述待推荐的券的点击数据、领取数据以及核销数据中的至少两种,所述点击数据包括点击所述待推荐的券的用户的特征信息,所述领取数据包括领取所述待推荐的券的用户的特征信息,所述核销数据包括使用所述待推荐的券的用户的特征信息;
基于预设模型对所述历史数据进行训练,得到所述转化率模型。
可选地,所述第二预测单元113,,根据所述点击率、所述领取率以及所述核销率中的至少两种,预测所述待推荐的券的转化率,包括:
将所述点击率、所述领取率以及所述核销率中至少两种的乘积确定为所述待推荐的券的转化率,其中:
若预测得到所述点击率以及所述核销率,则将所述点击率以及所述核销率的乘积确定为所述待推荐的券的转化率;
若预测得到所述领取率以及所述核销率,则将所述领取率以及所述核销率的乘积确定为所述待推荐的券的转化率;
若预测得到所述点击率、所述领取率以及所述核销率,则将所述点击率、所述领取率以及所述核销率的乘积确定为所述待推荐的券的转化率。
可选地,所述券转化率的预测装置110还包括:推荐单元114,其中:
所述推荐单元114,在所述第二预测单元113预测得到所述待推荐的券的转化率后,根据所述转化率向所述用户推荐券。
可选地,所述推荐单元114,根据所述转化率向所述用户推荐券,包括:
若所述待推荐的券的个数为多个,则:
根据所述转化率的大小对多个所述待推荐的券进行排序,并将转化率最大的券推荐给所述用户;或,
分别按照点击率、领取率以及核销率中的至少两种对多个所述待推荐的券进行排序;确定点击率、领取率以及核销率中至少两种均在设定排序范围内的券,将确定的券中转化率最大的券推荐给所述用户。
券转化率的预测装置110还可执行图4的方法,并实现券转化率的预测装置在图4所示实施例中的功能,本申请实施例在此不再赘述。
图12是本申请的一个实施例电子设备的结构示意图。请参考图12,在硬件层面,该电子设备包括处理器,可选地还包括内部总线、网络接口、存储器。其中,存储器可能包含内存,例如高速随机存取存储器(Random-Access Memory,RAM),也可能还包括非易失性存储器(non-volatile memory),例如至少1个磁盘存储器等。当然,该电子设备还可能包括其他业务所需要的硬件。
图12中,处理器从非易失性存储器中读取对应的计算机程序到内存中然后运行,在逻辑层面上形成推荐券的装置。处理器,执行存储器所存放的程序,并具体用于执行以下操作:
确定多个待推荐的券以及用户的特征信息;
根据所述多个待推荐的券以及所述特征信息,预测所述多个待推荐的券在转化过程中的点击率、领取率以及核销率中的至少两种;
根据所述点击率、所述领取率以及所述核销率中的至少两种,预测所述多个待 推荐的券的转化率;
根据所述多个待推荐的券的转化率,向所述用户推荐券。
上述如本申请图12所示实施例揭示的推荐券的装置执行的方法可以应用于处理器中,或者由处理器实现。此外,图12中其他硬件的具体连接结构以及实现的功能可以参见图8所示的电子设备中的相关记载,这里不再重复描述。
该电子设备还可执行图5的方法,并实现推荐券的装置在图5所示实施例中的功能,本申请实施例在此不再赘述。
当然,除了软件实现方式之外,本申请的电子设备并不排除其他实现方式,比如逻辑器件抑或软硬件结合的方式等等,也就是说以下处理流程的执行主体并不限定于各个逻辑单元,也可以是硬件或逻辑器件。
本申请实施例还提出了一种计算机可读存储介质,该计算机可读存储介质存储一个或多个程序,该一个或多个程序包括指令,该指令当被包括多个应用程序的便携式电子设备执行时,能够使该便携式电子设备执行图5所示实施例的方法,并具体用于执行以下操作:
确定多个待推荐的券以及用户的特征信息;
根据所述多个待推荐的券以及所述特征信息,预测所述多个待推荐的券在转化过程中的点击率、领取率以及核销率中的至少两种;
根据所述点击率、所述领取率以及所述核销率中的至少两种,预测所述多个待推荐的券的转化率;
根据所述多个待推荐的券的转化率,向所述用户推荐券。
图13是本申请的一个实施例推荐券的装置130的结构示意图。请参考图13,在一种软件实施方式中,所述推荐券的装置130可包括:确定单元131、第一预测单元132、第二预测单元133和推荐单元134,其中:
确定单元131,确定多个待推荐的券以及用户的特征信息;
第一预测单元132,根据所述多个待推荐的券以及所述特征信息,预测所述多个待推荐的券在转化过程中的点击率、领取率以及核销率中的至少两种;
第二预测单元133,根据所述点击率、所述领取率以及所述核销率中的至少两种,预测所述多个待推荐的券的转化率;
推荐单元134,根据所述多个待推荐的券的转化率,向所述用户推荐券。
可选地,所述推荐单元134,根据所述多个待推荐的券的转化率,向所述用户推荐券,包括:
根据转化率的大小对所述多个待推荐的券进行排序,并将转化率最大的券推荐给所述用户;或,
分别按照点击率、领取率以及核销率中的至少两种对所述多个待推荐的券进行排序;确定点击率、领取率以及核销率中至少两种均在设定排序范围内的券,将确定的券中转化率最大的券推荐给所述用户。
推荐券的装置130还可执行图5的方法,并实现推荐券的装置在图5所示实施例中的功能,本申请实施例在此不再赘述。
总之,以上所述仅为本申请的较佳实施例而已,并非用于限定本申请的保护范围。凡在本申请的精神和原则之内,所作的任何修改、等同替换、改进等,均应包含在本申请的保护范围之内。
上述实施例阐明的系统、装置、模块或单元,具体可以由计算机芯片或实体实现,或者由具有某种功能的产品来实现。一种典型的实现设备为计算机。具体的,计算机例如可以为个人计算机、膝上型计算机、蜂窝电话、相机电话、智能电话、个人数字助理、媒体播放器、导航设备、电子邮件设备、游戏控制台、平板计算机、可穿戴设备或者这些设备中的任何设备的组合。
计算机可读介质包括永久性和非永久性、可移动和非可移动媒体可以由任何方法或技术来实现信息存储。信息可以是计算机可读指令、数据结构、程序的模块或其他数据。计算机的存储介质的例子包括,但不限于相变内存(PRAM)、静态随机存取存储器(SRAM)、动态随机存取存储器(DRAM)、其他类型的随机存取存储器(RAM)、只读存储器(ROM)、电可擦除可编程只读存储器(EEPROM)、快闪记忆体或其他内存技术、只读光盘只读存储器(CD-ROM)、数字多功能光盘(DVD)或其他光学存储、磁盒式磁带,磁带磁磁盘存储或其他磁性存储设备或任何其他非传输介质,可用于存储可以被计算设备访问的信息。按照本文中的界定,计算机可读介质不包括暂存电脑可读媒体(transitory media),如调制的数据信号和载波。
还需要说明的是,术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含,从而使得包括一系列要素的过程、方法、商品或者设备不仅包括那些要素, 而且还包括没有明确列出的其他要素,或者是还包括为这种过程、方法、商品或者设备所固有的要素。在没有更多限制的情况下,由语句“包括一个……”限定的要素,并不排除在包括所述要素的过程、方法、商品或者设备中还存在另外的相同要素。
本说明书中的各个实施例均采用递进的方式描述,各个实施例之间相同相似的部分互相参见即可,每个实施例重点说明的都是与其他实施例的不同之处。尤其,对于系统实施例而言,由于其基本相似于方法实施例,所以描述的比较简单,相关之处参见方法实施例的部分说明即可。

Claims (30)

  1. 一种信息转化率的预测方法,包括:
    确定待推荐信息以及用户的特征信息;
    根据所述待推荐信息以及所述特征信息,预测影响所述待推荐信息转化过程的至少两种因素对应的转化概率;
    根据所述至少两种因素对应的转化概率,预测所述待推荐信息的转化率。
  2. 如权利要求1所述的方法,
    影响所述待推荐信息转化过程的至少两种因素包括:查看所述待推荐信息、获取所述待推荐信息以及使用所述待推荐信息中的至少两种;
    所述至少两种因素对应的转化概率包括:查看所述待推荐信息的第一概率、获取所述待推荐信息的第二概率以及使用所述待推荐信息的第三概率中的至少两种。
  3. 如权利要求2所述的方法,根据所述待推荐信息以及所述特征信息,预测影响所述待推荐信息转化过程的至少两种因素对应的转化概率,包括:
    确定预先训练得到的转化率模型,所述转化率模型包括第一概率模型、第二概率模型以及第三概率模型中的至少两种;
    根据所述待推荐信息以及所述特征信息,通过所述转化率模型确定所述第一概率、所述第二概率以及所述第三概率中的至少两种。
  4. 如权利要求3所述的方法,所述转化率模型通过以下方式训练得到,包括:
    获取所述待推荐信息的历史数据,所述历史数据包括:所述待推荐信息的属性信息和推荐方式,在不同的属性信息和推荐方式下所述待推荐信息的查看数据、获取数据以及使用数据中的至少两种,所述查看数据包括点击所述待推荐信息的用户的特征信息,所述获取数据包括获取所述待推荐信息的用户的特征信息,所述使用数据包括使用所述待推荐信息的用户的特征信息;
    基于预设模型对所述历史数据进行训练,得到所述转化率模型。
  5. 如权利要求4所述的方法,根据所述至少两种因素对应的转化概率,预测所述待推荐信息的转化率,包括:
    将预测的所述第一概率、所述第二概率以及所述第三概率中至少两种概率的乘积确定为所述待推荐信息的转化率,其中:
    若预测得到所述第一概率以及所述第三概率,则将所述第一概率以及所述第二概率的乘积确定为所述待推荐信息的转化率;
    若预测得到所述第二概率以及所述第三概率,则将所述第二概率以及所述第三概率 的乘积确定为所述待推荐信息的转化率;
    若预测得到所述第一概率、所述第二概率以及所述第三概率,则将所述第一概率、所述第二概率以及所述第三概率的乘积确定为所述待推荐信息的转化率。
  6. 如权利要求1至5任一项所述的方法,在预测得到所述待推荐信息的转化率后,所述方法还包括:
    根据所述待推荐信息的转化率,向所述用户进行信息推荐。
  7. 如权利要求6所述的方法,根据所述待推荐信息的转化率,向所述用户进行信息推荐,包括:
    若所述待推荐信息的个数为多个,则:
    根据所述转化率的大小对多个所述待推荐信息进行排序,并将转化率最大的待推荐信息推荐给所述用户;或,
    分别按照第一概率、第二概率以及第三概率中的至少两种对多个所述待推荐信息进行排序;确定第一概率、第二概率以及第三概率中至少两种均在设定排序范围内的待推荐信息,并将确定的待推荐信息中转化率最大的待推荐信息推荐给所述用户。
  8. 一种信息推荐方法,包括:
    确定多个待推荐信息以及用户的特征信息;
    根据所述多个待推荐信息以及所述特征信息,预测影响所述多个待推荐信息转化过程的至少两种因素对应的转化概率;
    根据所述至少两种因素对应的转化概率,预测所述多个待推荐信息的转化率;
    根据所述多个待推荐信息的转化率,向所述用户进行信息推荐。
  9. 如权利要求8所述的方法,根据所述多个待推荐信息的转化率,向所述用户进行信息推荐,包括:
    根据所述转化率的大小对多个所述待推荐信息进行排序,并将转化率最大的待推荐信息推荐给所述用户。
  10. 如权利要求8所述的方法,影响所述多个待推荐信息转化过程的至少两种因素对应的转化概率包括:查看所述待推荐信息的第一概率、获取所述待推荐信息的第二概率以及使用所述待推荐信息的第三概率中的至少两种;
    根据所述多个待推荐信息的转化率,向所述用户进行信息推荐,包括:
    分别按照第一概率、第二概率以及第三概率中的至少两种对所述多个待推荐信息进行排序;确定第一概率、第二概率以及第三概率中至少两种均在设定排序范围内的待推荐信息,并将确定的待推荐信息中转化率最大的待推荐信息推荐给所述用户。
  11. 一种券转化率的预测方法,包括:
    确定待推荐的券以及用户的特征信息;
    根据所述待推荐的券以及所述特征信息,预测所述待推荐的券在转化过程中的点击率、领取率以及核销率中的至少两种;
    根据所述点击率、所述领取率以及所述核销率中的至少两种,预测所述待推荐的券的转化率。
  12. 如权利要求11所示的方法,根据所述待推荐的券以及所述特征信息,预测所述待推荐的券在转化过程中的点击率、领取率以及核销率中的至少两种,包括:
    确定预先训练得到的转化率模型,所述转化率模型包括点击率模型、领取率模型以及核销率模型中的至少两种;
    根据所述待推荐的券以及所述特征信息,通过所述转化率模型确定所述点击率、所述领取率以及所述核销率中的至少两种。
  13. 如权利要求12所述的方法,所述转化率模型通过以下方式训练得到,包括:
    获取所述待推荐的券的历史数据,所述历史数据包括:所述待推荐的券的推荐方式,在所述推荐方式下所述待推荐的券的点击数据、领取数据以及核销数据中的至少两种,所述点击数据包括点击所述待推荐的券的用户的特征信息,所述领取数据包括领取所述待推荐的券的用户的特征信息,所述核销数据包括使用所述待推荐的券的用户的特征信息;
    基于预设模型对所述历史数据进行训练,得到所述转化率模型。
  14. 如权利要求13所述的方法,根据所述点击率、所述领取率以及所述核销率中的至少两种,预测所述待推荐的券的转化率,包括:
    将所述点击率、所述领取率以及所述核销率中至少两种的乘积确定为所述待推荐的券的转化率,其中:
    若预测得到所述点击率以及所述核销率,则将所述点击率以及所述核销率的乘积确定为所述待推荐的券的转化率;
    若预测得到所述领取率以及所述核销率,则将所述领取率以及所述核销率的乘积确定为所述待推荐的券的转化率;
    若预测得到所述点击率、所述领取率以及所述核销率,则将所述点击率、所述领取率以及所述核销率的乘积确定为所述待推荐的券的转化率。
  15. 如权利要求11至14任一项所述的方法,在预测得到所述待推荐的券的转化率后,所述方法还包括:
    根据所述转化率向所述用户推荐券。
  16. 如权利要求15所述的方法,根据所述转化率向所述用户推荐券,包括:
    若所述待推荐的券的个数为多个,则:
    根据所述转化率的大小对多个所述待推荐的券进行排序,并将转化率最大的券推荐给所述用户;或,
    分别按照点击率、领取率以及核销率中的至少两种对多个所述待推荐的券进行排序;确定点击率、领取率以及核销率中至少两种均在设定排序范围内的券,将确定的券中转化率最大的券推荐给所述用户。
  17. 一种推荐券的方法,包括:
    确定多个待推荐的券以及用户的特征信息;
    根据所述多个待推荐的券以及所述特征信息,预测所述多个待推荐的券在转化过程中的点击率、领取率以及核销率中的至少两种;
    根据所述点击率、所述领取率以及所述核销率中的至少两种,预测所述多个待推荐的券的转化率;
    根据所述多个待推荐的券的转化率,向所述用户推荐券。
  18. 如权利要求17所述的方法,根据所述多个待推荐的券的转化率,向所述用户推荐券,包括:
    根据转化率的大小对所述多个待推荐的券进行排序,并将转化率最大的券推荐给所述用户;或,
    分别按照点击率、领取率以及核销率中的至少两种对所述多个待推荐的券进行排序;确定点击率、领取率以及核销率中至少两种均在设定排序范围内的券,将确定的券中转化率最大的券推荐给所述用户。
  19. 一种信息转化率的预测装置,包括:
    确定单元,确定待推荐信息以及用户的特征信息;
    第一预测单元,根据所述待推荐信息以及所述特征信息,预测影响所述待推荐信息转化过程的至少两种因素对应的转化概率;
    第二预测单元;根据所述至少两种转化概率,预测所述待推荐信息的转化率。
  20. 一种电子设备,包括:
    处理器;以及
    被安排成存储计算机可执行指令的存储器,该可执行指令在被执行时使该处理器执行以下操作:
    确定待推荐信息以及用户的特征信息;
    根据所述待推荐信息以及所述特征信息,预测影响所述待推荐信息转化过程的至少两种因素对应的转化概率;
    根据所述至少两种因素对应的转化概率,预测所述待推荐信息的转化率。
  21. 一种计算机可读存储介质,所述计算机可读存储介质存储一个或多个程序,所述一个或多个程序当被包括多个应用程序的电子设备执行时,使得所述电子设备执行以下方法:
    确定待推荐信息以及用户的特征信息;
    根据所述待推荐信息以及所述特征信息,预测影响所述待推荐信息转化过程的至少两种因素对应的转化概率;
    根据所述至少两种因素对应的转化概率,预测所述待推荐信息的转化率。
  22. 一种信息推荐装置,包括:
    确定单元,确定多个待推荐信息以及用户的特征信息;
    第一预测单元,根据所述多个待推荐信息以及所述特征信息,预测影响所述多个待推荐信息转化过程的至少两种因素对应的转化概率;
    第二预测单元,根据所述至少两种因素对应的转化概率,预测所述多个待推荐信息的转化率;
    推荐单元,根据所述多个待推荐信息的转化率,向所述用户进行信息推荐。
  23. 一种电子设备,包括:
    处理器;以及
    被安排成存储计算机可执行指令的存储器,该可执行指令在被执行时使该处理器执行以下操作:
    确定多个待推荐信息以及用户的特征信息;
    根据所述多个待推荐信息以及所述特征信息,预测影响所述多个待推荐信息转化过程的至少两种因素对应的转化概率;
    根据所述至少两种因素对应的转化概率,预测所述多个待推荐信息的转化率;
    根据所述多个待推荐信息的转化率,向所述用户进行信息推荐。
  24. 一种计算机可读存储介质,所述计算机可读存储介质存储一个或多个程序,所述一个或多个程序当被包括多个应用程序的电子设备执行时,使得所述电子设备执行以下方法:
    确定多个待推荐信息以及用户的特征信息;
    根据所述多个待推荐信息以及所述特征信息,预测影响所述多个待推荐信息转化过程的至少两种因素对应的转化概率;
    根据所述至少两种因素对应的转化概率,预测所述多个待推荐信息的转化率;
    根据所述多个待推荐信息的转化率,向所述用户进行信息推荐。
  25. 一种券转化率的预测装置,包括:
    确定单元,确定待推荐的券以及用户的特征信息;
    第一预测单元,根据所述待推荐的券以及所述特征信息,预测所述待推荐的券在转化过程中的点击率、领取率以及核销率中的至少两种;
    第二预测单元,根据所述点击率、所述领取率以及所述核销率中的至少两种,预测所述待推荐的券的转化率。
  26. 一种电子设备,包括:
    处理器;以及
    被安排成存储计算机可执行指令的存储器,该可执行指令在被执行时使该处理器执行以下操作:
    确定待推荐的券以及用户的特征信息;
    根据所述待推荐的券以及所述特征信息,预测所述待推荐的券在转化过程中的点击率、领取率以及核销率中的至少两种;
    根据所述点击率、所述领取率以及所述核销率中的至少两种,预测所述待推荐的券的转化率。
  27. 一种计算机可读存储介质,所述计算机可读存储介质存储一个或多个程序,所述一个或多个程序当被包括多个应用程序的电子设备执行时,使得所述电子设备执行以下方法:
    确定待推荐的券以及用户的特征信息;
    根据所述待推荐的券以及所述特征信息,预测所述待推荐的券在转化过程中的点击率、领取率以及核销率中的至少两种;
    根据所述点击率、所述领取率以及所述核销率中的至少两种,预测所述待推荐的券的转化率。
  28. 一种券推荐装置,包括:
    确定单元,确定多个待推荐的券以及用户的特征信息;
    第一预测单元,根据所述多个待推荐的券以及所述特征信息,预测所述多个待推荐的券在转化过程中的点击率、领取率以及核销率中的至少两种;
    第二预测单元,根据所述点击率、所述领取率以及所述核销率中的至少两种,预测所述多个待推荐的券的转化率;
    推荐单元,根据所述多个待推荐的券的转化率,向所述用户推荐券。
  29. 一种电子设备,包括:
    处理器;以及
    被安排成存储计算机可执行指令的存储器,该可执行指令在被执行时使该处理器执行以下操作:
    确定多个待推荐的券以及用户的特征信息;
    根据所述多个待推荐的券以及所述特征信息,预测所述多个待推荐的券在转化过程中的点击率、领取率以及核销率中的至少两种;
    根据所述点击率、所述领取率以及所述核销率中的至少两种,预测所述多个待推荐的券的转化率;
    根据所述多个待推荐的券的转化率,向所述用户推荐券。
  30. 一种计算机可读存储介质,所述计算机可读存储介质存储一个或多个程序,所述一个或多个程序当被包括多个应用程序的电子设备执行时,使得所述电子设备执行以下方法:
    确定多个待推荐的券以及用户的特征信息;
    根据所述多个待推荐的券以及所述特征信息,预测所述多个待推荐的券在转化过程中的点击率、领取率以及核销率中的至少两种;
    根据所述点击率、所述领取率以及所述核销率中的至少两种,预测所述多个待推荐的券的转化率;
    根据所述多个待推荐的券的转化率,向所述用户推荐券。
PCT/CN2019/073365 2018-03-07 2019-01-28 一种信息转化率的预测、信息推荐方法和装置 WO2019169977A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201810187238.8 2018-03-07
CN201810187238.8A CN108510313A (zh) 2018-03-07 2018-03-07 一种信息转化率的预测、信息推荐方法和装置

Publications (1)

Publication Number Publication Date
WO2019169977A1 true WO2019169977A1 (zh) 2019-09-12

Family

ID=63376213

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/073365 WO2019169977A1 (zh) 2018-03-07 2019-01-28 一种信息转化率的预测、信息推荐方法和装置

Country Status (3)

Country Link
CN (1) CN108510313A (zh)
TW (1) TWI777010B (zh)
WO (1) WO2019169977A1 (zh)

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111160959A (zh) * 2019-12-20 2020-05-15 深圳前海微众银行股份有限公司 一种用户点击转化预估方法及装置
CN111460290A (zh) * 2020-03-27 2020-07-28 喜丈(上海)网络科技有限公司 信息的推荐方法、装置、设备和存储介质
CN111859289A (zh) * 2020-06-17 2020-10-30 北京嘀嘀无限科技发展有限公司 交通工具的交易转化率预估方法、装置、电子设备及介质
CN112381607A (zh) * 2020-11-12 2021-02-19 杭州时趣信息技术有限公司 一种网络商品排序方法、装置、设备及介质
CN112508613A (zh) * 2020-12-09 2021-03-16 天津狮拓信息技术有限公司 一种商品推荐方法、装置、电子设备和可读存储介质
CN112667892A (zh) * 2020-12-25 2021-04-16 北京达佳互联信息技术有限公司 一种信息推荐方法、装置、服务器和存储介质
CN113360770A (zh) * 2021-06-30 2021-09-07 北京百度网讯科技有限公司 内容推荐的方法、装置、设备以及存储介质
CN114422584A (zh) * 2021-12-16 2022-04-29 北京百度网讯科技有限公司 资源的推送方法、设备和存储介质

Families Citing this family (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108510313A (zh) * 2018-03-07 2018-09-07 阿里巴巴集团控股有限公司 一种信息转化率的预测、信息推荐方法和装置
CN109697636A (zh) * 2018-12-27 2019-04-30 拉扎斯网络科技(上海)有限公司 一种商户推荐方法、商户推荐装置、电子设备和介质
CN111382346B (zh) * 2018-12-28 2023-09-01 第四范式(北京)技术有限公司 用于推荐内容的方法及系统
CN109740066B (zh) * 2019-01-28 2021-05-28 北京三快在线科技有限公司 信息推荐方法、信息推荐装置、存储介质及电子设备
CN110046932A (zh) * 2019-03-29 2019-07-23 阿里巴巴集团控股有限公司 一种数据处理方法、装置及设备
CN110188917A (zh) * 2019-04-16 2019-08-30 上海上湖信息技术有限公司 客户转化率分析方法及装置、可读存储介质
CN110569427B (zh) * 2019-08-07 2022-03-01 智者四海(北京)技术有限公司 一种多目标排序模型训练、用户行为预测方法及装置
CN110490389B (zh) * 2019-08-27 2023-07-21 腾讯科技(深圳)有限公司 点击率预测方法、装置、设备及介质
CN112819497B (zh) * 2019-11-18 2023-10-10 百度在线网络技术(北京)有限公司 转化率预测方法、装置、设备和存储介质
CN111192108A (zh) * 2019-12-16 2020-05-22 北京淇瑀信息科技有限公司 一种用于产品推荐的排序方法、装置和电子设备
CN111401974A (zh) * 2020-06-03 2020-07-10 北京每日优鲜电子商务有限公司 信息发送方法、装置、电子设备和计算机可读介质
CN111798280B (zh) * 2020-09-08 2020-12-15 腾讯科技(深圳)有限公司 多媒体信息推荐方法、装置和设备及存储介质
CN113344647B (zh) * 2021-07-14 2023-05-23 杭州网易云音乐科技有限公司 一种信息推荐的方法及装置
TWI810670B (zh) * 2021-10-04 2023-08-01 中華電信股份有限公司 服務優惠推薦之裝置、方法及執行其方法之電腦程式產品

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102479366A (zh) * 2010-11-25 2012-05-30 阿里巴巴集团控股有限公司 一种商品推荐方法及系统
CN102663519A (zh) * 2012-04-01 2012-09-12 浙江盘石信息技术有限公司 网络广告投放中媒体选择的优化系统及方法
CN103618774A (zh) * 2013-11-19 2014-03-05 北京奇虎科技有限公司 一种基于网络行为的资源推荐方法及装置、系统
CN104598511A (zh) * 2014-10-23 2015-05-06 腾讯科技(深圳)有限公司 搜索结果推荐方法、装置及系统
CN107391680A (zh) * 2017-07-24 2017-11-24 北京京东尚科信息技术有限公司 内容推荐方法、装置和设备
CN107688967A (zh) * 2017-08-24 2018-02-13 平安科技(深圳)有限公司 客户购买意向的预测方法及终端设备
CN108510313A (zh) * 2018-03-07 2018-09-07 阿里巴巴集团控股有限公司 一种信息转化率的预测、信息推荐方法和装置

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102479366A (zh) * 2010-11-25 2012-05-30 阿里巴巴集团控股有限公司 一种商品推荐方法及系统
CN102663519A (zh) * 2012-04-01 2012-09-12 浙江盘石信息技术有限公司 网络广告投放中媒体选择的优化系统及方法
CN103618774A (zh) * 2013-11-19 2014-03-05 北京奇虎科技有限公司 一种基于网络行为的资源推荐方法及装置、系统
CN104598511A (zh) * 2014-10-23 2015-05-06 腾讯科技(深圳)有限公司 搜索结果推荐方法、装置及系统
CN107391680A (zh) * 2017-07-24 2017-11-24 北京京东尚科信息技术有限公司 内容推荐方法、装置和设备
CN107688967A (zh) * 2017-08-24 2018-02-13 平安科技(深圳)有限公司 客户购买意向的预测方法及终端设备
CN108510313A (zh) * 2018-03-07 2018-09-07 阿里巴巴集团控股有限公司 一种信息转化率的预测、信息推荐方法和装置

Cited By (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111160959A (zh) * 2019-12-20 2020-05-15 深圳前海微众银行股份有限公司 一种用户点击转化预估方法及装置
CN111160959B (zh) * 2019-12-20 2023-06-27 深圳前海微众银行股份有限公司 一种用户点击转化预估方法及装置
CN111460290B (zh) * 2020-03-27 2023-06-02 喜丈(上海)网络科技有限公司 信息的推荐方法、装置、设备和存储介质
CN111460290A (zh) * 2020-03-27 2020-07-28 喜丈(上海)网络科技有限公司 信息的推荐方法、装置、设备和存储介质
CN111859289A (zh) * 2020-06-17 2020-10-30 北京嘀嘀无限科技发展有限公司 交通工具的交易转化率预估方法、装置、电子设备及介质
CN111859289B (zh) * 2020-06-17 2024-04-19 北京嘀嘀无限科技发展有限公司 交通工具的交易转化率预估方法、装置、电子设备及介质
CN112381607A (zh) * 2020-11-12 2021-02-19 杭州时趣信息技术有限公司 一种网络商品排序方法、装置、设备及介质
CN112381607B (zh) * 2020-11-12 2023-11-24 杭州时趣信息技术有限公司 一种网络商品排序方法、装置、设备及介质
CN112508613B (zh) * 2020-12-09 2024-03-19 天津狮拓信息技术有限公司 一种商品推荐方法、装置、电子设备和可读存储介质
CN112508613A (zh) * 2020-12-09 2021-03-16 天津狮拓信息技术有限公司 一种商品推荐方法、装置、电子设备和可读存储介质
CN112667892A (zh) * 2020-12-25 2021-04-16 北京达佳互联信息技术有限公司 一种信息推荐方法、装置、服务器和存储介质
CN112667892B (zh) * 2020-12-25 2024-01-19 北京达佳互联信息技术有限公司 一种信息推荐方法、装置、服务器和存储介质
CN113360770A (zh) * 2021-06-30 2021-09-07 北京百度网讯科技有限公司 内容推荐的方法、装置、设备以及存储介质
CN113360770B (zh) * 2021-06-30 2024-04-12 北京百度网讯科技有限公司 内容推荐的方法、装置、设备以及存储介质
CN114422584A (zh) * 2021-12-16 2022-04-29 北京百度网讯科技有限公司 资源的推送方法、设备和存储介质

Also Published As

Publication number Publication date
CN108510313A (zh) 2018-09-07
TWI777010B (zh) 2022-09-11
TW201939379A (zh) 2019-10-01

Similar Documents

Publication Publication Date Title
WO2019169977A1 (zh) 一种信息转化率的预测、信息推荐方法和装置
JP6940646B2 (ja) 情報推奨方法、情報推奨装置、機器および媒体
US11315155B2 (en) Method and system for exposing data used in ranking search results
WO2019149145A1 (zh) 投诉举报类别的排序方法和装置
US20180081978A1 (en) Method and Apparatus for Processing Information
US9064212B2 (en) Automatic event categorization for event ticket network systems
US20190311395A1 (en) Estimating click-through rate
US8315895B1 (en) Method and system for obtaining review updates within a review and rating system
US11275748B2 (en) Influence score of a social media domain
CN106295832A (zh) 产品信息推送方法及装置
CN108885624A (zh) 信息推荐系统及方法
CN113688313A (zh) 一种预测模型的训练方法、信息推送的方法及装置
CN110334356A (zh) 文章质量的确定方法、文章筛选方法、以及相应的装置
CN114817538B (zh) 文本分类模型的训练方法、文本分类方法及相关设备
WO2023000491A1 (zh) 一种应用推荐方法、装置、设备及计算机可读存储介质
CN105761154A (zh) 一种社会化推荐方法及装置
CN110334936B (zh) 一种信贷资质评分模型的构建方法、装置和设备
CN113535817B (zh) 特征宽表生成及业务处理模型的训练方法和装置
CN105138897B (zh) 确定用户权限的方法及装置
CN113159870A (zh) 推送信息的展示方法、装置及计算机设备
US10671932B1 (en) Software application selection models integration
US20140045165A1 (en) Methods and apparatus for training people on the use of sentiment and predictive capabilities resulting therefrom
CN108229572B (zh) 一种参数寻优方法及计算设备
US10031970B1 (en) Search engine optimization in social question and answer systems
CN112446777A (zh) 一种信用评估方法、装置、设备及存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19764097

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19764097

Country of ref document: EP

Kind code of ref document: A1