CN111080413A - E-commerce platform commodity recommendation method and device, server and storage medium - Google Patents
E-commerce platform commodity recommendation method and device, server and storage medium Download PDFInfo
- Publication number
- CN111080413A CN111080413A CN201911326148.3A CN201911326148A CN111080413A CN 111080413 A CN111080413 A CN 111080413A CN 201911326148 A CN201911326148 A CN 201911326148A CN 111080413 A CN111080413 A CN 111080413A
- Authority
- CN
- China
- Prior art keywords
- user
- click
- preset
- recommended
- prediction model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0623—Item investigation
- G06Q30/0625—Directed, with specific intent or strategy
Landscapes
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the invention provides a method, a device, a server and a storage medium for recommending E-commerce platform commodities. The recommendation method of the E-commerce platform commodities comprises the following steps: acquiring user behavior data of a first user; determining a user representation of the first user based on the user behavior data; training a preset recall model based on a first preset algorithm; training a click prediction model based on a second preset algorithm; determining a recommended commodity set according to the user portrait and the preset recall model, wherein the recommended commodity set comprises a plurality of recommended commodities; sequencing a plurality of recommended commodities in the recommended commodity set according to the user behavior data and the click prediction model; and returning the sequencing result to a commodity recommendation list and displaying the commodity recommendation list to the first user. The effect of providing different recommended commodities for different users to improve the purchase rate of the users is achieved.
Description
Technical Field
The embodiment of the invention relates to the technical field of electronic commerce, in particular to a method, a device, a server and a storage medium for recommending e-commerce platform commodities.
Background
With the rapid growth of the internet, more and more users are beginning to make purchases and consume on the internet.
The user needs to search for the product on the e-commerce platform and then order and purchase the favorite product. At present, in order to improve the purchase rate of a user, when the user opens an e-commerce platform, goods are recommended to the user, and the user can select a good of interest from the recommended goods to purchase.
However, the current electronic commerce system recommends the same items to the users, and each user sees the same ordered list of items. The interests of each user are different, and the conventional scheme cannot provide thousands of friendly shopping experiences for the users and improve the sales performance intelligently.
Disclosure of Invention
The embodiment of the invention provides a method, a device, a server and a storage medium for recommending E-commerce platform commodities, so as to provide different recommended commodities for different users and improve the purchase rate of the users.
In a first aspect, an embodiment of the present invention provides a method for recommending e-commerce platform goods, including:
acquiring user behavior data of a first user;
determining a user representation of the first user based on the user behavior data;
training a preset recall model based on a first preset algorithm;
training a click prediction model based on a second preset algorithm;
determining a recommended commodity set according to the user portrait and the preset recall model, wherein the recommended commodity set comprises a plurality of recommended commodities;
sequencing a plurality of recommended commodities in the recommended commodity set according to the user behavior data and the click prediction model;
and returning the sequencing result to a commodity recommendation list and displaying the commodity recommendation list to the first user.
Optionally, before the sorting the plurality of recommended commodities in the recommended commodity set according to the user behavior data and the click prediction model, the method includes:
establishing a plurality of preset click models;
shunting the user to each preset click model through an ABTest mechanism to obtain a prediction result of each preset click model;
and determining an optimal preset click model as the click prediction model according to the click purchase result of the user.
Optionally, the multiple preset click models include an XGBoost + LR combination algorithm and a Wide & Deep learning algorithm.
Optionally, after the returning of the sorted result to the product recommendation list and displaying to the first user, the method includes:
acquiring click purchase result sets of a plurality of first users of each click prediction model;
judging the prediction effect of each click prediction model based on the click purchase result set;
judging whether the prediction effect of each click prediction model is larger than a preset threshold value or not;
and if the prediction effect of the click prediction model is smaller than the preset threshold value, adjusting the wind control parameter of the click prediction model.
Optionally, the user behavior data includes a data source, and before the sorting of the recommended commodities in the recommended commodity set according to the user behavior data and the click prediction model, the method includes:
and matching a click prediction model corresponding to the data source according to the data source.
Optionally, the user behavior data includes one or more of basic information of purchasing goods, searched goods, time of browsing goods, evaluation of goods, number of repeated purchases, and interval duration of repeated purchases.
Optionally, the preset recall model is obtained by training based on a collaborative filtering algorithm.
In a second aspect, an embodiment of the present invention provides a device for recommending e-commerce platform goods, including:
the acquisition module is used for acquiring user behavior data of a first user;
a representation module to determine a user representation of the first user based on the user behavior data;
the algorithm training module is used for training a preset recall model based on a first preset algorithm and training a click prediction model based on a second preset algorithm;
the recall module is used for determining a recommended commodity set according to the user portrait and the preset recall model, and the recommended commodity set comprises a plurality of recommended commodities;
the sorting module is used for sorting the recommended commodities in the recommended commodity set according to the user behavior data and the click prediction model;
and the display module is used for returning the sequencing result to the commodity recommendation list and displaying the sequencing result to the first user.
In a third aspect, an embodiment of the present invention provides a server, including:
one or more processors;
a storage device for storing one or more programs,
when executed by the one or more processors, cause the one or more processors to implement a method for recommending e-commerce platform merchandise according to any embodiment of the present invention.
In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements a recommendation method for an e-commerce platform good according to any embodiment of the present invention.
The embodiment of the invention obtains the user behavior data of the first user; determining a user representation of the first user based on the user behavior data; training a preset recall model based on a first preset algorithm; training a click prediction model based on a second preset algorithm; determining a recommended commodity set according to the user portrait and the preset recall model, wherein the recommended commodity set comprises a plurality of recommended commodities; sequencing a plurality of recommended commodities in the recommended commodity set according to the user behavior data and the click prediction model; the sorted result is returned to the commodity recommendation list to be displayed to the first user, so that the problems that thousands of friendly shopping experiences cannot be provided for the user, and the sales performance cannot be intelligently improved are solved, and the effects of providing different recommended commodities for different users and improving the purchase rate of the user are achieved.
Drawings
Fig. 1 is a schematic flowchart of a method for recommending e-commerce platform goods according to an embodiment of the present invention;
fig. 2 is a schematic flowchart of a method for recommending e-commerce platform goods according to a second embodiment of the present invention;
fig. 3 is a schematic structural diagram of a recommendation device for e-commerce platform goods according to a third embodiment of the present invention;
fig. 4 is a schematic structural diagram of a server according to a fourth embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It should be further noted that, for the convenience of description, only some of the structures related to the present invention are shown in the drawings, not all of the structures.
Before discussing exemplary embodiments in more detail, it should be noted that some exemplary embodiments are described as processes or methods depicted as flowcharts. Although a flowchart may describe the steps as a sequential process, many of the steps can be performed in parallel, concurrently or simultaneously. In addition, the order of the steps may be rearranged. A process may be terminated when its operations are completed, but may have additional steps not included in the figure. A process may correspond to a method, a function, a procedure, a subroutine, a subprogram, etc.
Furthermore, the terms "first," "second," and the like may be used herein to describe various orientations, actions, steps, elements, or the like, but the orientations, actions, steps, or elements are not limited by these terms. These terms are only used to distinguish one direction, action, step or element from another direction, action, step or element. For example, the first information may be referred to as second information, and similarly, the second information may be referred to as first information, without departing from the scope of the present application. The first information and the second information are both information, but they are not the same information. The terms "first", "second", etc. are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of the present invention, "a plurality" means at least two, e.g., two, three, etc., unless specifically limited otherwise.
Example one
Fig. 1 is a schematic flow chart of a method for recommending e-commerce platform goods according to an embodiment of the present invention, which is applicable to a scenario of recommending goods for different users.
As shown in fig. 1, a method for recommending e-commerce platform goods according to an embodiment of the present invention includes:
and S110, acquiring user behavior data of the first user.
Wherein the first user is a user who purchases using the e-commerce platform. In this embodiment, the first user does not refer to a specific user. The user behavior data refers to data of the user using the e-commerce platform. Specifically, the user behavior data includes, but is not limited to, one or more of basic information of purchasing goods, searched goods, time of browsing goods, evaluation of goods, number of repeated purchases, and interval duration of repeated purchases. For example, the user a purchases a digital camera, and the basic information of the purchased goods may include information of the brand, model, etc. of the digital camera, which is not limited herein.
S120, determining a user portrait of the first user based on the user behavior data.
The user representation is a representation in which information of a user is labeled. Specifically, the tag of the first user is determined through the user behavior data of the first user. For example, if most or all of the items purchased by the first user are men's clothing, the user of the first user may be depicted as a man; most of the articles browsed by the first user are mobile phones, computers and the like, and the user portrait of the first user is a favorite electronic product. Optionally, the preset user portrait may be formulated in advance, and when the user portrait of the first user is determined according to the user behavior data, which of the preset user portraits is in accordance with the first user may be determined according to the user behavior data, so as to determine the user portrait of the first user, which is not limited herein.
Alternatively, The user profile may be generated using an online learning algorithm FTRL (follow The regulated leader Proximal).
S130, training a preset recall model based on a first preset algorithm;
the first preset algorithm is an algorithm for training a preset recall model. In this embodiment, the first preset algorithm may be a recall algorithm based on collaborative filtering, and is not limited herein.
And S140, training a click prediction model based on a second preset algorithm.
The second preset algorithm is an algorithm for training the click prediction model. In this embodiment, the second preset algorithm may be one or more. Specifically, the second preset algorithm may be an XGBoost + LR combination algorithm and/or a Wide & Deep learning algorithm, which is not limited herein.
S150, determining a recommended commodity set according to the user portrait and the preset recall model, wherein the recommended commodity set comprises a plurality of recommended commodities.
The preset recall model is a model for recommending commodities according to the user portrait. In this embodiment, optionally, the preset recall model is obtained by training based on a collaborative filtering algorithm. Collaborative filtering refers to using past behavior or opinion of an existing user group to predict what a current user is most likely to like or be interested in. Illustratively, the user portraits of user a and user B are both males who like digital products, and when user B is interested in the camera lens, user a is considered to be also interested in the camera lens. The recommended commodity set refers to a set of a plurality of recommended commodities, such as a camera body, a camera lens, a mobile phone, a computer, and the like, and is not limited herein. The recommended commodity set is determined through the preset recall model and the user portrait of the first user, the recommended commodity set not only comprises commodities preferred by the first user, but also comprises commodities possibly preferred by the first user, and the recommendable commodity range is wider.
S160, sequencing the recommended commodities in the recommended commodity set according to the user behavior data and the click prediction model.
The click prediction model is a model for predicting the possibility that the first user clicks to purchase a plurality of recommended goods. Specifically, the click prediction model is obtained by mass behavior data training of mass users. In this embodiment, optionally, the click prediction model may be a model obtained by training through an XGBoost (eXtreme Gradient boost) combined algorithm plus LR (Logistic Regression model); or a model obtained by training through Wide & Deep learning algorithm, which is not limited herein. Xgboost can be used to construct new feature variables, and LR can assemble the original and new features to construct a model and calculate the significance and weighting coefficients of each feature. Wide & Deep combines LR and DNN, Wide part is LR, Deep part is DNN, and the results of the two are combined and output. The plurality of recommended commodities are ranked through the user behavior data and the predicted click model of the first user, and therefore personalized recommendation can be conducted on the first user. Due to the fact that the user behavior data of different users are different, the effect of providing different recommended commodities is achieved.
Optionally, the training algorithm of the click prediction model may be implanted in the form of a plug-in, so that different training algorithms may be switched as needed to train the click prediction model, and the switching manner is simple.
In an optional embodiment, before sorting the recommended commodities in the recommended commodity set according to the user behavior data and the click prediction model, the method may include:
establishing a plurality of preset click models;
shunting the user to each preset click model through an ABTest mechanism to obtain a prediction result of each preset click model;
and determining an optimal preset click model as the click prediction model according to the click purchase result of the user.
The preset click models refer to training models trained through various training algorithms. In this embodiment, optionally, the plurality of preset click models include an XGBoost + LR combination algorithm and a Wide & Deep learning algorithm, which is not limited herein. The ABTest mechanism is used for distributing massive users to a plurality of preset click models as evenly as possible, so that input parameters of different preset click models are equal. Specifically, a large number of users can be averagely distributed into a plurality of preset click models according to a plurality of classifications of gender, age, region, occupation and the like. And determining an optimal preset click model as a click prediction model according to the actual click purchase result of the user, namely determining the most accurate preset click model as the click prediction model. Alternatively, the click purchase result may include purchase and non-purchase, and the click purchase result may be determined by determining whether the user purchases the product. The higher the number or proportion of purchases, the better the click purchase results.
And S170, returning the sequencing result to a commodity recommendation list and displaying the commodity recommendation list to the first user.
The commodity recommendation list is a list in which a plurality of recommended commodities are displayed in a visual form. The sorted result is returned to the commodity list and displayed to the first user, the first user can purchase according to the recommended commodities, the previously sorted recommended commodities are commodities with high predicted purchase possibility of the predicted click model, and the purchase rate of the user is improved. In addition, the plurality of recommended commodities are sequenced by establishing the click prediction model, and manual intervention is not needed.
In an alternative embodiment, the click prediction model is multiple, and after the sorted result is returned to the item recommendation list and displayed to the first user, the method may include:
acquiring click purchase result sets of a plurality of first users of each click prediction model;
judging the prediction effect of each click prediction model based on the click purchase result set;
judging whether the prediction effect of each click prediction model is larger than a preset threshold value or not;
and if the prediction effect of the click prediction model is smaller than the preset threshold value, adjusting the wind control parameter of the click prediction model.
The click purchase result refers to the embodiment that the first user purchases based on the sorted result. The click purchase result set refers to a set of click purchase results of a plurality of first users. Specifically, one purchase result set corresponds to one click prediction model. Illustratively, two click prediction models A and B are provided, when a plurality of click prediction models are used for sorting, different users are distributed into the model A or the model B according to a preset strategy, and the click purchase results of all first users distributed to the model A form a click purchase result set of the model A; and (4) shunting all the click purchase results of the first users to the B model to form a click purchase result set of the B model. Alternatively, the click purchase result set may be counted for a day. Specifically, the prediction effect of the click prediction model may be determined according to the click purchase result set, and the determination may be performed according to whether all the first users of each click prediction model purchase recommended commodities, which is not limited herein. In this embodiment, the predicted effect may be, optionally, a click purchase rate or a conversion rate, etc., that is, an overall click purchase rate or a conversion rate of all the first users of each click prediction model, which is not limited herein. Specifically, the click purchase rate refers to a ratio of the user to click and purchase in the recommended product. Optionally, the click purchase rate may be calculated by a ratio of the number of recommended commodities purchased by all the first users of each click prediction model to the total number of recommended commodities of all the first users of each click prediction model; or may be calculated by the ratio of the number of recommended items purchased by all first users of each click prediction model to the total number of recommended items viewable by all first users of each click prediction model, which is not limited herein. Optionally, click through rate may also be used for risk control. Illustratively, when the click rate is below a value, the model is deactivated and other models are invoked for sorting. The preset threshold is set as required, and may be any value from 0 to 1. The wind control parameters refer to parameters for adjusting the shunting users to click the prediction model. In this embodiment, the wind control parameters may shunt user-to-model policies, proportions, or quantities. Specifically, when the prediction effect of a certain click prediction model is smaller than a preset threshold, for example, the click rate is smaller than the preset threshold, the parameters of the click prediction model are adjusted, for example, the number of users who are distributed to the click prediction model is reduced or the distribution of users to the click prediction model is stopped. Illustratively, when the prediction effect of the click prediction model A is smaller than a preset threshold, the user is stopped from being shunted to the click prediction model A, or the number of users shunted to the click prediction model A is reduced. The number of users shunted to the A prediction model is reduced, and the users can continuously observe for a period of time, so that whether the prediction effect is smaller than a preset threshold value is an accidental situation or not is judged.
Optionally, after a new algorithm is online, the new algorithm may be scheduled to train a preset recall model and click a prediction model according to a configuration policy without restarting a service.
According to the technical scheme of the embodiment of the invention, user behavior data of a first user is obtained; determining a user representation of the first user based on the user behavior data; training a preset recall model based on a first preset algorithm; training a click prediction model based on a second preset algorithm; determining a recommended commodity set according to the user portrait and the preset recall model, wherein the recommended commodity set comprises a plurality of recommended commodities; sequencing a plurality of recommended commodities in the recommended commodity set according to the user behavior data and the click prediction model; and returning the sequencing result to a commodity recommendation list to be displayed to the first user, determining different recommended commodities for the users according to different user behavior data of different users and a preset recall model, sequencing the recommended commodities of the different users according to a click prediction model, sequencing the commodities with high predicted purchasing possibility in the front, and achieving the technical effects of providing different recommended commodities for the different users and improving the purchasing rate of the users.
Example two
Fig. 2 is a schematic flow chart of a method for recommending e-commerce platform goods according to a second embodiment of the present invention. The embodiment is further detailed in the technical scheme, and is suitable for scenes of recommending commodities for different users. The method can be executed by a recommendation device of the E-commerce platform commodity, and the device can be realized in a software and/or hardware mode and can be integrated on a server.
As shown in fig. 2, a method for recommending e-commerce platform goods provided by the second embodiment of the present invention includes:
s210, user behavior data of the first user are obtained.
Wherein the first user is a user who purchases using the e-commerce platform. In this embodiment, the first user does not refer to a specific user. The user behavior data refers to data of the user using the e-commerce platform. Specifically, the user behavior data includes, but is not limited to, basic information of purchased goods, searched goods, time of browsing goods, evaluation of goods, number of repeated purchases, interval duration of repeated purchases, and the like, and is not limited herein. For example, the user a purchases a digital camera, and the basic information of the purchased goods may include information of the brand, model, etc. of the digital camera, which is not limited herein.
S220, determining a user portrait of the first user based on the user behavior data.
The user representation is a representation in which information of a user is labeled. Specifically, the tag of the first user is determined through the user behavior data of the first user. For example, if most or all of the items purchased by the first user are men's clothing, the user of the first user may be depicted as a man; most of the articles browsed by the first user are mobile phones, computers and the like, and the user portrait of the first user is a favorite electronic product. Optionally, the preset user portrait may be formulated in advance, and when the user portrait of the first user is determined according to the user behavior data, which of the preset user portraits is in accordance with the first user may be determined according to the user behavior data, so as to determine the user portrait of the first user, which is not limited herein.
In this embodiment, the user behavior data includes a data source.
S230, training a preset recall model based on a first preset algorithm;
the first preset algorithm is an algorithm for training a preset recall model. In this embodiment, the first preset algorithm may be a recall algorithm based on collaborative filtering, and is not limited herein.
S240, training a click prediction model based on a second preset algorithm.
The second preset algorithm is an algorithm for training the click prediction model. In this embodiment, the second preset algorithm may be one or more. Specifically, the second preset algorithm may be an XGBoost + LR combination algorithm and/or a Wide & Deep learning algorithm, which is not limited herein.
S250, determining a recommended commodity set according to the user portrait and the preset recall model, wherein the recommended commodity set comprises a plurality of recommended commodities.
The preset recall model is a model for recommending commodities according to the user portrait. In this embodiment, optionally, the preset recall model is obtained by training based on a collaborative filtering algorithm. Collaborative filtering refers to using past behavior or opinion of an existing user group to predict what a current user is most likely to like or be interested in. Illustratively, the user portraits of user a and user B are both males who like digital products, and when user B is interested in the camera lens, user a is considered to be also interested in the camera lens. The recommended commodity set refers to a set of a plurality of recommended commodities, such as a camera body, a camera lens, a mobile phone, a computer, and the like, and is not limited herein. The recommended commodity set is determined through the preset recall model and the user portrait of the first user, the recommended commodity set not only comprises commodities preferred by the first user, but also comprises commodities possibly preferred by the first user, and the recommendable commodity range is wider.
And S260, matching a click prediction model corresponding to the data source according to the data source.
The data source refers to a source of user behavior data. In this embodiment, the data source refers to which terminal operating system or channel the user behavior data source originates from, for example, an android system, an IOS (Input output system) system or a Windows system, an APP channel, a wechat community program, and a web page channel. And matching and predicting a more accurate click prediction model according to different data sources, so that the predicted result can be more accurate. And performing multi-model accurate matching of different scenes by using a hierarchical ABtest method.
S270, sequencing the recommended commodities in the recommended commodity set according to the user behavior data and the click prediction model.
The click prediction model is a model for predicting the possibility that the first user clicks to purchase a plurality of recommended goods. Specifically, the click prediction model is obtained by mass behavior data training of mass users. In this embodiment, optionally, the click prediction model may be a model obtained by training through an XGBoost (eXtreme Gradient boost) combined algorithm plus LR (Logistic Regression model); or a model obtained by training through Wide & Deep learning algorithm, which is not limited herein. Xgboost can be used to construct new feature variables, and LR can assemble the original and new features to construct a model and calculate the significance and weighting coefficients of each feature. The wide and Deep combine LR and DNN, the wide part is LR, the Deep part is DNN (Deep Neural Networks), and the results of the two are combined and output. The plurality of recommended commodities are ranked through the user behavior data and the predicted click model of the first user, and therefore personalized recommendation can be conducted on the first user. Due to the fact that the user behavior data of different users are different, the effect of providing different recommended commodities is achieved.
And S280, returning the sequencing result to a commodity recommendation list and displaying the commodity recommendation list to the first user.
The commodity recommendation list is a list in which a plurality of recommended commodities are displayed in a visual form. The sorted result is returned to the commodity list and displayed to the first user, the first user can purchase according to the recommended commodities, the previously sorted recommended commodities are commodities with high predicted purchase possibility of the predicted click model, and the purchase rate of the user is improved. In addition, the plurality of recommended commodities are sequenced by establishing the click prediction model, and manual intervention is not needed.
According to the technical scheme of the embodiment of the invention, user behavior data of a first user is obtained; determining a user representation of the first user based on the user behavior data; training a preset recall model based on a first preset algorithm; training a click prediction model based on a second preset algorithm; determining a recommended commodity set according to the user portrait and the preset recall model, wherein the recommended commodity set comprises a plurality of recommended commodities; sequencing a plurality of recommended commodities in the recommended commodity set according to the user behavior data and the click prediction model; and returning the sequencing result to a commodity recommendation list to be displayed to the first user, determining different recommended commodities for the users according to different user behavior data of different users and a preset recall model, sequencing the recommended commodities of the different users according to a click prediction model, sequencing the commodities with high predicted purchasing possibility in the front, and achieving the technical effects of providing different recommended commodities for the different users and improving the purchasing rate of the users.
EXAMPLE III
Fig. 3 is a schematic structural diagram of a device for recommending e-commerce platform goods according to a third embodiment of the present invention, where the third embodiment of the present invention is applicable to a scenario in which goods are recommended for different users, and the device may be implemented in a software and/or hardware manner and may be integrated on a server.
As shown in fig. 3, the apparatus for recommending e-commerce platform goods provided in this embodiment may include an obtaining module 310, a representation module 320, an algorithm training module 330, a recall module 340, a sorting module 350, and a display module 360, wherein:
an obtaining module 310, configured to obtain user behavior data of a first user;
a portrayal module 320 for determining a user portrayal of the first user based on the user behavior data;
an algorithm training module 330 for training a preset recall model based on a first preset algorithm; and training a click prediction model based on a second preset algorithm.
A recall module 340, configured to determine a recommended commodity set according to the user representation and the preset recall model, where the recommended commodity set includes a plurality of recommended commodities;
a sorting module 350, configured to sort, according to the user behavior data and the click prediction model, a plurality of recommended commodities in the recommended commodity set;
and the display module 360 is configured to return the sorted result to the commodity recommendation list and display the commodity recommendation list to the first user.
Optionally, the apparatus further comprises:
the preset click model establishing module is used for establishing a plurality of preset click models;
the algorithm scheduling module is used for shunting the user to each preset click model through an ABtest mechanism to obtain a prediction result of each preset click model; and determining an optimal preset click model as the click prediction model according to the click purchase result of the user.
Optionally, the algorithm scheduling module may be further configured to schedule the new algorithm to train the preset recall model and the click prediction model according to the configuration policy without restarting the service when the new algorithm is online.
Optionally, the multiple preset click models include an XGBoost + LR combination algorithm and a Wide & Deep learning algorithm.
Optionally, the obtaining module 310 is further configured to obtain a click purchase result of the first user;
the device also includes:
the prediction effect judging module is used for judging the prediction effect of the click prediction model based on the click purchase result;
the judging module is used for judging whether the prediction effect is larger than a preset threshold value or not;
and the adjusting module is used for adjusting the wind control parameter of the click prediction model if the prediction effect is smaller than the preset threshold value.
Optionally, the apparatus further comprises:
and the matching module is used for matching the click prediction model corresponding to the data source according to the data source.
Optionally, the user behavior data includes one or more of basic information of purchasing goods, searched goods, time of browsing goods, evaluation of goods, number of repeated purchases, and interval duration of repeated purchases.
Optionally, the preset recall model is obtained by training based on a collaborative filtering algorithm. The algorithm training module 330 performs model training on the big data cluster by using a machine learning method, uses data fed back by clicking and purchasing of a user as a sample, clicks prediction or purchases prediction as a target, and continuously performs iterative training to automatically update model parameters.
The E-commerce platform commodity recommendation device provided by the embodiment of the invention can execute the E-commerce platform commodity recommendation method provided by any embodiment of the invention, and has corresponding functional modules and beneficial effects of the execution method. Reference may be made to the description of any method embodiment of the invention not specifically described in this embodiment.
Example four
Fig. 4 is a schematic structural diagram of a server according to a fourth embodiment of the present invention. FIG. 4 illustrates a block diagram of an exemplary server 612 suitable for use in implementing embodiments of the present invention. The server 612 shown in fig. 4 is only an example, and should not bring any limitation to the function and the scope of the use of the embodiments of the present invention.
As shown in fig. 4, the server 612 is in the form of a general-purpose server. The components of server 612 may include, but are not limited to: one or more processors 616, a memory device 628, and a bus 618 that couples the various system components including the memory device 628 and the processors 616.
The server 612 typically includes a variety of computer system readable media. Such media can be any available media that is accessible by server 612 and includes both volatile and nonvolatile media, removable and non-removable media.
A program/utility 640 having a set (at least one) of program modules 642 may be stored, for example, in storage 628, such program modules 642 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each of which examples or some combination thereof may comprise an implementation of a network environment. The program modules 642 generally perform the functions and/or methods of the described embodiments of the present invention.
The server 612 may also communicate with one or more external devices 614 (e.g., keyboard, pointing terminal, display 624, etc.), with one or more terminals that enable a user to interact with the server 612, and/or with any terminals (e.g., network card, modem, etc.) that enable the server 612 to communicate with one or more other computing terminals. Such communication may occur via input/output (I/O) interfaces 622. Further, server 612 may communicate with one or more networks (e.g., a Local Area Network (LAN), Wide Area Network (WAN), and/or a public Network such as the internet) via Network adapter 620. As shown in FIG. 4, the network adapter 620 communicates with the other modules of the server 612 via the bus 618. It should be appreciated that although not shown, other hardware and/or software modules may be used in conjunction with the server 612, including but not limited to: microcode, end drives, Redundant processors, external disk drive Arrays, RAID (Redundant Arrays of Independent Disks) systems, tape drives, and data backup storage systems, among others.
The processor 616 executes various functional applications and data processing by running programs stored in the storage device 628, for example, implementing a method for recommending e-commerce platform goods provided by any embodiment of the present invention, the method may include:
acquiring user behavior data of a first user;
determining a user representation of the first user based on the user behavior data;
training a preset recall model based on a first preset algorithm;
training a click prediction model based on a second preset algorithm;
determining a recommended commodity set according to the user portrait and the preset recall model, wherein the recommended commodity set comprises a plurality of recommended commodities;
sequencing a plurality of recommended commodities in the recommended commodity set according to the user behavior data and the click prediction model;
and returning the sequencing result to a commodity recommendation list and displaying the commodity recommendation list to the first user.
According to the technical scheme of the embodiment of the invention, user behavior data of a first user is obtained; determining a user representation of the first user based on the user behavior data; training a preset recall model based on a first preset algorithm; training a click prediction model based on a second preset algorithm; determining a recommended commodity set according to the user portrait and the preset recall model, wherein the recommended commodity set comprises a plurality of recommended commodities; sequencing a plurality of recommended commodities in the recommended commodity set according to the user behavior data and the click prediction model; and returning the sequencing result to a commodity recommendation list to be displayed to the first user, determining different recommended commodities for the users according to different user behavior data of different users and a preset recall model, sequencing the recommended commodities of the different users according to a click prediction model, sequencing the commodities with high predicted purchasing possibility in the front, and achieving the technical effects of providing different recommended commodities for the different users and improving the purchasing rate of the users.
EXAMPLE five
An embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements a method for recommending e-commerce platform goods, where the method includes:
acquiring user behavior data of a first user;
determining a user representation of the first user based on the user behavior data;
training a preset recall model based on a first preset algorithm;
training a click prediction model based on a second preset algorithm;
determining a recommended commodity set according to the user portrait and the preset recall model, wherein the recommended commodity set comprises a plurality of recommended commodities;
sequencing a plurality of recommended commodities in the recommended commodity set according to the user behavior data and the click prediction model;
and returning the sequencing result to a commodity recommendation list and displaying the commodity recommendation list to the first user.
The computer-readable storage media of embodiments of the invention may take any combination of one or more computer-readable media. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples (a non-exhaustive list) of the computer readable storage medium would include the following: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this document, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.
A computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.
Program code embodied on a storage medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.
Computer program code for carrying out operations for aspects of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, golang, python, Scala, C + +, and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or terminal. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).
According to the technical scheme of the embodiment of the invention, user behavior data of a first user is obtained; determining a user representation of the first user based on the user behavior data; training a preset recall model based on a first preset algorithm; training a click prediction model based on a second preset algorithm; determining a recommended commodity set according to the user portrait and the preset recall model, wherein the recommended commodity set comprises a plurality of recommended commodities; sequencing a plurality of recommended commodities in the recommended commodity set according to the user behavior data and the click prediction model; and returning the sequencing result to a commodity recommendation list to be displayed to the first user, determining different recommended commodities for the users according to different user behavior data of different users and a preset recall model, sequencing the recommended commodities of the different users according to a click prediction model, sequencing the commodities with high predicted purchasing possibility in the front, and achieving the technical effects of providing different recommended commodities for the different users and improving the purchasing rate of the users.
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.
Claims (10)
1. A recommendation method for E-commerce platform commodities is characterized by comprising the following steps:
acquiring user behavior data of a first user;
determining a user representation of the first user based on the user behavior data;
training a preset recall model based on a first preset algorithm;
training a click prediction model based on a second preset algorithm;
determining a recommended commodity set according to the user portrait and the preset recall model, wherein the recommended commodity set comprises a plurality of recommended commodities;
sequencing a plurality of recommended commodities in the recommended commodity set according to the user behavior data and the click prediction model;
and returning the sequencing result to a commodity recommendation list and displaying the commodity recommendation list to the first user.
2. The e-commerce platform commodity recommendation method of claim 1, wherein before the sorting the plurality of recommended commodities in the set of recommended commodities according to the user behavior data and the click prediction model, the method comprises:
establishing a plurality of preset click models;
shunting the user to each preset click model through an ABTest mechanism to obtain a prediction result of each preset click model;
and determining an optimal preset click model as the click prediction model according to the click purchase result of the user.
3. The e-commerce platform commodity recommendation method of claim 2, wherein the plurality of preset click models comprise an XGBoost + LR combinatorial algorithm and a Wide & Deep learning algorithm.
4. The method for recommending e-commerce platform goods according to claim 1, wherein said click prediction model is plural, and after said returning the sorted result to the goods recommendation list and displaying it to said first user, comprises:
acquiring click purchase result sets of a plurality of first users of each click prediction model;
judging the prediction effect of each click prediction model based on the click purchase result set;
judging whether the prediction effect of each click prediction model is larger than a preset threshold value or not;
and if the prediction effect of the click prediction model is smaller than the preset threshold value, adjusting the wind control parameter of the click prediction model.
5. The method for recommending e-commerce platform merchandise of claim 1, wherein said user behavior data comprises a data source, comprising, prior to said sorting a plurality of recommended merchandise in said set of recommended merchandise according to said user behavior data and said click prediction model:
and matching a click prediction model corresponding to the data source according to the data source.
6. The e-commerce platform recommendation method of goods according to claim 1, wherein the user behavior data comprises one or more of basic information of the goods purchased, the goods searched, time of browsing the goods, evaluation of the goods, the number of repeated purchases, and the interval duration of the repeated purchases.
7. The recommendation method for E-commerce platform commodities according to any one of claims 1-6, wherein the preset recall model is trained based on a collaborative filtering algorithm.
8. An e-commerce platform merchandise recommendation device, comprising:
the acquisition module is used for acquiring user behavior data of a first user;
a representation module to determine a user representation of the first user based on the user behavior data;
the algorithm training module is used for training a preset recall model based on a first preset algorithm and training a click prediction model based on a second preset algorithm;
the recall module is used for determining a recommended commodity set according to the user portrait and the preset recall model, and the recommended commodity set comprises a plurality of recommended commodities;
the sorting module is used for sorting the recommended commodities in the recommended commodity set according to the user behavior data and the click prediction model;
and the display module is used for returning the sequencing result to the commodity recommendation list and displaying the sequencing result to the first user.
9. A server, comprising:
one or more processors;
storage means for storing one or more programs;
when executed by the one or more processors, cause the one or more processors to implement the method for recommending e-commerce platform goods according to any of claims 1-7.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out a method of recommending an e-commerce platform good according to any one of claims 1-7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911326148.3A CN111080413A (en) | 2019-12-20 | 2019-12-20 | E-commerce platform commodity recommendation method and device, server and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911326148.3A CN111080413A (en) | 2019-12-20 | 2019-12-20 | E-commerce platform commodity recommendation method and device, server and storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111080413A true CN111080413A (en) | 2020-04-28 |
Family
ID=70316274
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201911326148.3A Pending CN111080413A (en) | 2019-12-20 | 2019-12-20 | E-commerce platform commodity recommendation method and device, server and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111080413A (en) |
Cited By (25)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111582984A (en) * | 2020-05-08 | 2020-08-25 | 广州探途网络技术有限公司 | Commodity recommendation method and device and computing equipment |
CN111652648A (en) * | 2020-06-03 | 2020-09-11 | 陈包容 | Method for intelligently generating personalized combined promotion scheme and system with same |
CN111754300A (en) * | 2020-06-22 | 2020-10-09 | 深圳市分期乐网络科技有限公司 | Commodity recommendation method, commodity recommendation device, commodity recommendation equipment and storage medium |
CN112070226A (en) * | 2020-09-02 | 2020-12-11 | 北京百度网讯科技有限公司 | Training method, device and equipment of online prediction model and storage medium |
CN112116434A (en) * | 2020-10-06 | 2020-12-22 | 广州智物互联科技有限公司 | Commodity searching method and system based on big data and electronic mall and cloud service platform |
CN112150206A (en) * | 2020-09-28 | 2020-12-29 | 京东数字科技控股股份有限公司 | Method and equipment for predicting user interested article |
CN112200601A (en) * | 2020-09-11 | 2021-01-08 | 深圳市法本信息技术股份有限公司 | Item recommendation method and device and readable storage medium |
CN112508613A (en) * | 2020-12-09 | 2021-03-16 | 天津狮拓信息技术有限公司 | Commodity recommendation method and device, electronic equipment and readable storage medium |
CN112528164A (en) * | 2020-12-14 | 2021-03-19 | 建信金融科技有限责任公司 | User collaborative filtering recall method and device |
CN112785397A (en) * | 2021-03-09 | 2021-05-11 | 中国工商银行股份有限公司 | Product recommendation method, device and storage medium |
CN113111253A (en) * | 2020-12-31 | 2021-07-13 | 绍兴亿都信息技术股份有限公司 | Personalized recommendation method, system, equipment and medium based on traditional Chinese medicine knowledge |
CN113378068A (en) * | 2021-07-14 | 2021-09-10 | 聚好看科技股份有限公司 | Content recommendation method and server |
CN113435983A (en) * | 2021-07-21 | 2021-09-24 | 陕西科技大学 | Personalized commodity recommendation method based on machine vision and improved neural network |
CN113487117A (en) * | 2021-08-20 | 2021-10-08 | 山东省计算中心(国家超级计算济南中心) | Method and system for simulating e-commerce user behavior data based on multi-dimensional user portrait |
CN113609392A (en) * | 2021-08-06 | 2021-11-05 | 腾讯科技(深圳)有限公司 | Content recommendation method, content to be recommended determination method and related device |
CN113627967A (en) * | 2020-05-09 | 2021-11-09 | 北京奇虎科技有限公司 | Behavior data processing method, device, equipment and storage medium |
CN113781146A (en) * | 2020-11-17 | 2021-12-10 | 北京沃东天骏信息技术有限公司 | Recommendation method, device, equipment and storage medium of product information |
CN113837843A (en) * | 2021-09-29 | 2021-12-24 | 平安科技(深圳)有限公司 | Product recommendation method, device, medium and electronic equipment |
CN113850649A (en) * | 2021-09-29 | 2021-12-28 | 中国科学院空天信息创新研究院 | Customized recommendation method and recommendation system based on multi-platform user data |
CN114219580A (en) * | 2021-12-31 | 2022-03-22 | 江苏玖益贰信息科技有限公司 | Consumer portrait generation device and portrait analysis method |
CN115017400A (en) * | 2021-11-30 | 2022-09-06 | 荣耀终端有限公司 | Application APP recommendation method and electronic equipment |
WO2022220738A1 (en) * | 2021-04-15 | 2022-10-20 | Grabtaxi Holdings Pte. Ltd. | Communications server apparatus, method and communications system for providing a recommendation of offerings by merchants to a user |
CN115373718A (en) * | 2022-10-26 | 2022-11-22 | 小米汽车科技有限公司 | Updating method and device of online model and electronic equipment |
CN116894709A (en) * | 2023-07-14 | 2023-10-17 | 广州洋葱时尚集团有限公司 | Advertisement commodity recommendation method and system |
CN117196662A (en) * | 2023-03-22 | 2023-12-08 | 山东浪潮爱购云链信息科技有限公司 | User behavior acquisition method, device and medium based on e-commerce platform |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105894332A (en) * | 2016-04-22 | 2016-08-24 | 深圳市永兴元科技有限公司 | Commodity recommendation method, device and system based on user behavior analysis |
CN107705183A (en) * | 2017-09-30 | 2018-02-16 | 深圳乐信软件技术有限公司 | Recommendation method, apparatus, storage medium and the server of a kind of commodity |
CN109767255A (en) * | 2018-12-06 | 2019-05-17 | 东莞团贷网互联网科技服务有限公司 | A method of it is modeled by big data and realizes intelligence operation and precision marketing |
CN109934512A (en) * | 2019-03-28 | 2019-06-25 | 努比亚技术有限公司 | A kind of training method and system of prediction model |
CN110008399A (en) * | 2019-01-30 | 2019-07-12 | 阿里巴巴集团控股有限公司 | A kind of training method and device, a kind of recommended method and device of recommended models |
CN110533515A (en) * | 2019-09-04 | 2019-12-03 | 深圳创新奇智科技有限公司 | A kind of the electric business personalized recommendation method and device of high low latency of handling up |
-
2019
- 2019-12-20 CN CN201911326148.3A patent/CN111080413A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105894332A (en) * | 2016-04-22 | 2016-08-24 | 深圳市永兴元科技有限公司 | Commodity recommendation method, device and system based on user behavior analysis |
CN107705183A (en) * | 2017-09-30 | 2018-02-16 | 深圳乐信软件技术有限公司 | Recommendation method, apparatus, storage medium and the server of a kind of commodity |
CN109767255A (en) * | 2018-12-06 | 2019-05-17 | 东莞团贷网互联网科技服务有限公司 | A method of it is modeled by big data and realizes intelligence operation and precision marketing |
CN110008399A (en) * | 2019-01-30 | 2019-07-12 | 阿里巴巴集团控股有限公司 | A kind of training method and device, a kind of recommended method and device of recommended models |
CN109934512A (en) * | 2019-03-28 | 2019-06-25 | 努比亚技术有限公司 | A kind of training method and system of prediction model |
CN110533515A (en) * | 2019-09-04 | 2019-12-03 | 深圳创新奇智科技有限公司 | A kind of the electric business personalized recommendation method and device of high low latency of handling up |
Cited By (37)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111582984A (en) * | 2020-05-08 | 2020-08-25 | 广州探途网络技术有限公司 | Commodity recommendation method and device and computing equipment |
CN113627967A (en) * | 2020-05-09 | 2021-11-09 | 北京奇虎科技有限公司 | Behavior data processing method, device, equipment and storage medium |
CN111652648A (en) * | 2020-06-03 | 2020-09-11 | 陈包容 | Method for intelligently generating personalized combined promotion scheme and system with same |
CN111754300A (en) * | 2020-06-22 | 2020-10-09 | 深圳市分期乐网络科技有限公司 | Commodity recommendation method, commodity recommendation device, commodity recommendation equipment and storage medium |
CN112070226A (en) * | 2020-09-02 | 2020-12-11 | 北京百度网讯科技有限公司 | Training method, device and equipment of online prediction model and storage medium |
EP3846109A3 (en) * | 2020-09-02 | 2021-11-17 | Beijing Baidu Netcom Science And Technology Co. Ltd. | Method and apparatus for training online prediction model, device and storage medium |
CN112070226B (en) * | 2020-09-02 | 2023-06-27 | 北京百度网讯科技有限公司 | Training method, device, equipment and storage medium of online prediction model |
CN112200601A (en) * | 2020-09-11 | 2021-01-08 | 深圳市法本信息技术股份有限公司 | Item recommendation method and device and readable storage medium |
CN112200601B (en) * | 2020-09-11 | 2024-05-14 | 深圳市法本信息技术股份有限公司 | Item recommendation method, device and readable storage medium |
CN112150206A (en) * | 2020-09-28 | 2020-12-29 | 京东数字科技控股股份有限公司 | Method and equipment for predicting user interested article |
CN112116434B (en) * | 2020-10-06 | 2021-11-16 | 深圳市宝贝团信息技术有限公司 | Commodity searching method and system based on big data and electronic mall and cloud service platform |
CN112116434A (en) * | 2020-10-06 | 2020-12-22 | 广州智物互联科技有限公司 | Commodity searching method and system based on big data and electronic mall and cloud service platform |
CN113781146A (en) * | 2020-11-17 | 2021-12-10 | 北京沃东天骏信息技术有限公司 | Recommendation method, device, equipment and storage medium of product information |
CN112508613B (en) * | 2020-12-09 | 2024-03-19 | 天津狮拓信息技术有限公司 | Commodity recommendation method and device, electronic equipment and readable storage medium |
CN112508613A (en) * | 2020-12-09 | 2021-03-16 | 天津狮拓信息技术有限公司 | Commodity recommendation method and device, electronic equipment and readable storage medium |
CN112528164B (en) * | 2020-12-14 | 2023-02-03 | 中国建设银行股份有限公司 | User collaborative filtering recall method and device |
CN112528164A (en) * | 2020-12-14 | 2021-03-19 | 建信金融科技有限责任公司 | User collaborative filtering recall method and device |
CN113111253A (en) * | 2020-12-31 | 2021-07-13 | 绍兴亿都信息技术股份有限公司 | Personalized recommendation method, system, equipment and medium based on traditional Chinese medicine knowledge |
CN112785397A (en) * | 2021-03-09 | 2021-05-11 | 中国工商银行股份有限公司 | Product recommendation method, device and storage medium |
CN112785397B (en) * | 2021-03-09 | 2024-03-29 | 中国工商银行股份有限公司 | Product recommendation method, device and storage medium |
WO2022220738A1 (en) * | 2021-04-15 | 2022-10-20 | Grabtaxi Holdings Pte. Ltd. | Communications server apparatus, method and communications system for providing a recommendation of offerings by merchants to a user |
CN113378068B (en) * | 2021-07-14 | 2023-07-18 | 聚好看科技股份有限公司 | Content recommendation method and server |
CN113378068A (en) * | 2021-07-14 | 2021-09-10 | 聚好看科技股份有限公司 | Content recommendation method and server |
CN113435983A (en) * | 2021-07-21 | 2021-09-24 | 陕西科技大学 | Personalized commodity recommendation method based on machine vision and improved neural network |
CN113609392B (en) * | 2021-08-06 | 2023-06-06 | 腾讯科技(深圳)有限公司 | Content recommendation method, content to be recommended determining method and related device |
CN113609392A (en) * | 2021-08-06 | 2021-11-05 | 腾讯科技(深圳)有限公司 | Content recommendation method, content to be recommended determination method and related device |
CN113487117A (en) * | 2021-08-20 | 2021-10-08 | 山东省计算中心(国家超级计算济南中心) | Method and system for simulating e-commerce user behavior data based on multi-dimensional user portrait |
CN113487117B (en) * | 2021-08-20 | 2023-10-17 | 山东省计算中心(国家超级计算济南中心) | Method and system for simulating behavior data of electric business based on multi-dimensional user portrait |
CN113837843A (en) * | 2021-09-29 | 2021-12-24 | 平安科技(深圳)有限公司 | Product recommendation method, device, medium and electronic equipment |
CN113837843B (en) * | 2021-09-29 | 2023-11-24 | 平安科技(深圳)有限公司 | Product recommendation method and device, medium and electronic equipment |
CN113850649A (en) * | 2021-09-29 | 2021-12-28 | 中国科学院空天信息创新研究院 | Customized recommendation method and recommendation system based on multi-platform user data |
CN115017400B (en) * | 2021-11-30 | 2023-05-26 | 荣耀终端有限公司 | Application APP recommendation method and electronic equipment |
CN115017400A (en) * | 2021-11-30 | 2022-09-06 | 荣耀终端有限公司 | Application APP recommendation method and electronic equipment |
CN114219580A (en) * | 2021-12-31 | 2022-03-22 | 江苏玖益贰信息科技有限公司 | Consumer portrait generation device and portrait analysis method |
CN115373718A (en) * | 2022-10-26 | 2022-11-22 | 小米汽车科技有限公司 | Updating method and device of online model and electronic equipment |
CN117196662A (en) * | 2023-03-22 | 2023-12-08 | 山东浪潮爱购云链信息科技有限公司 | User behavior acquisition method, device and medium based on e-commerce platform |
CN116894709A (en) * | 2023-07-14 | 2023-10-17 | 广州洋葱时尚集团有限公司 | Advertisement commodity recommendation method and system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111080413A (en) | E-commerce platform commodity recommendation method and device, server and storage medium | |
CN108629665B (en) | Personalized commodity recommendation method and system | |
Golder et al. | Will it ever fly? Modeling the takeoff of really new consumer durables | |
Sismeiro et al. | Modeling purchase behavior at an e-commerce web site: A task-completion approach | |
CN112487278A (en) | Training method of recommendation model, and method and device for predicting selection probability | |
CN109417644B (en) | Revenue optimization for cross-screen advertising | |
WO2020221022A1 (en) | Service object recommendation method | |
CN112365283B (en) | Coupon issuing method and device, terminal equipment and storage medium | |
CN114463091A (en) | Information push model training and information push method, device, equipment and medium | |
CN114168843A (en) | Search word recommendation method, device and storage medium | |
CN113744020A (en) | Commodity file processing method and device, electronic equipment and storage medium | |
CN112598472A (en) | Product recommendation method, device, system, medium and program product | |
CN113689259A (en) | Commodity personalized recommendation method and system based on user behaviors | |
CN113327151A (en) | Commodity object recommendation method and device, computer equipment and storage medium | |
Jiang et al. | A high-dimensional choice model for online retailing | |
CN111680213B (en) | Information recommendation method, data processing method and device | |
CN109255645B (en) | Consumption prediction method and device and electronic equipment | |
CN113822734B (en) | Method and device for generating information | |
CN108319612A (en) | Audient's Media Recommendation Method and system | |
CN113744009B (en) | Target object output method, device, computer readable medium and electronic equipment | |
CN113792952A (en) | Method and apparatus for generating a model | |
CN113159877B (en) | Data processing method, device, system and computer readable storage medium | |
CN113762535A (en) | Article recall method and device | |
CN113378033A (en) | Training method and device for recommendation model | |
CN113256362A (en) | Method and apparatus for outputting information |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |