CN117689435A

CN117689435A - Item recommendation method, device, computer equipment and storage medium thereof

Info

Publication number: CN117689435A
Application number: CN202310816126.5A
Authority: CN
Inventors: 朱佳琪
Original assignee: Bank of China Ltd
Current assignee: Bank of China Ltd
Priority date: 2023-07-04
Filing date: 2023-07-04
Publication date: 2024-03-12

Abstract

The present disclosure relates to the field of information recommendation technologies, and in particular, to an article recommendation method, an apparatus, a computer device, and a storage medium thereof. The method comprises the following steps: determining the current article preference of the target user and an operation feedback result corresponding to the current article preference according to a preference determination mode corresponding to the user type of the target user; according to the operation feedback result corresponding to the current article preference, verifying the accuracy of the current article preference; and if the verification is passed, recommending the target object to the target user based on the current object preference. According to the method and the device, the user does not need to find the interested article through a traversing searching method, the interested article can be obtained only according to the recommended content of the target article, the time wasted when the user searches the article is reduced, and the purchasing experience of the user is optimized.

Description

Item recommendation method, device, computer equipment and storage medium thereof

Technical Field

The present disclosure relates to the field of information recommendation technologies, and in particular, to an article recommendation method, an apparatus, a computer device, and a storage medium thereof.

Background

With the continuous development and popularization of internet technology, more and more articles are put on the internet for users to select and purchase articles at any time and any place through the internet. When a user wants to purchase an article on the internet, the user needs to find the interested article by traversing the searching method, so as to complete the purchasing operation of the article.

However, as the number of selectable items on the internet increases, the time taken for the user to find the item of interest by traversing the search method increases, resulting in a waste of time and a poor purchase experience for the user.

Disclosure of Invention

In view of the foregoing, it is desirable to provide an item recommendation method, apparatus, computer device, and storage medium thereof that enable a target user to quickly determine an item of interest.

In a first aspect, the present application provides a method of item recommendation. The method comprises the following steps:

determining the current article preference of the target user and an operation feedback result corresponding to the current article preference according to a preference determination mode corresponding to the user type of the target user;

according to the operation feedback result corresponding to the current article preference, verifying the accuracy of the current article preference;

and if the verification is passed, recommending the target object to the target user based on the current object preference.

In one embodiment, determining, according to a preference determining manner corresponding to a user type to which the target user belongs, a current item preference of the target user and an operation feedback result corresponding to the current item preference includes:

If the preference determining mode corresponding to the user type of the target user is the first-round determining mode, determining a reference user corresponding to the target user based on the user information of the target user, wherein the similarity between the user information of the reference user and the user information of the target user is larger than a similarity threshold;

taking the historical item preference of the reference user as the current item preference of the target user; wherein the historical item preference is the item preference last determined for the reference user;

and taking the operation feedback result corresponding to the historical article preference of the reference user as the operation feedback result corresponding to the current article preference.

In one embodiment, determining, according to a preference determining manner corresponding to a user type to which the target user belongs, a current item preference of the target user and an operation feedback result corresponding to the current item preference, further includes:

if the preference determining mode corresponding to the user type of the target user is a non-first-round determining mode, determining the historical article preference of the target user and the historical operation feedback corresponding to the historical article preference;

updating the historical item preference of the target user based on the historical operation feedback corresponding to the historical item preference to obtain the current item preference of the target user;

And predicting an operation feedback result corresponding to the current item preference based on the historical operation feedback corresponding to the historical item preference and the current item preference of the target user.

In one embodiment, the method further comprises:

if the verification is not passed, updating the current item preference according to the operation feedback result corresponding to the current item preference to obtain the updated current item preference;

and recommending the target object to the target user based on the updated current object preference.

In one embodiment, updating the current item preference according to the operation feedback result corresponding to the current item preference to obtain the updated current item preference, including:

according to the operation feedback result corresponding to the current article preference, determining the interested articles and the uninteresting articles of the target user;

determining article attributes of the article of interest and the article not of interest;

updating the current item preference based on the item attributes of the interested items and the uninteresting items to obtain the updated current item preference.

In one embodiment, making a target item recommendation to a target user based on current item preferences includes:

determining item attributes of at least two candidate items;

Based on the current item preference, carrying out attribute screening on the item attribute of each candidate item to obtain a screening result, wherein the screening result comprises at least one target item matched with the current item preference in each candidate item and the matching degree of each target item and the current item preference;

and recommending the target object to the target user based on the screening result.

In a second aspect, the present application further provides an article recommendation device. The device comprises:

the determining module is used for determining the current article preference of the target user and an operation feedback result corresponding to the current article preference according to a preference determining mode corresponding to the user type of the target user;

the verification module is used for verifying the accuracy of the preference of the current article according to the operation feedback result corresponding to the preference of the current article;

and the recommending module is used for recommending the target object to the target user based on the current object preference if the verification is passed.

In a third aspect, the present application also provides a computer device. The computer device comprises a memory storing a computer program and a processor implementing the item recommendation method according to any of the embodiments of the first aspect described above when the processor executes the computer program.

In a fourth aspect, the present application also provides a computer-readable storage medium. A computer readable storage medium having stored thereon a computer program which when executed by a processor implements an item recommendation method as in any of the embodiments of the first aspect described above.

In a fifth aspect, the present application also provides a computer program product. A computer program product comprising a computer program which when executed by a processor implements a product recommendation method as in any of the embodiments of the first aspect described above.

According to the item recommending method, the device, the computer equipment and the storage medium thereof, the current item preference of the target user and the operation feedback result corresponding to the current item preference are determined based on the preference determining mode of the user type of the target user; and recommending the target object to the target user according to the accuracy verification result of the current object preference. In the process, the target object is recommended to the target user according to the accuracy verification result of the current object preference, so that the user does not need to find the object of interest through a traversal searching method, the object of interest can be obtained only according to the recommended content of the target object, the time wasted when the user searches the object is reduced, and the purchasing experience of the user is optimized.

Drawings

Fig. 1 is an application environment diagram of an article recommendation method provided in an embodiment of the present application;

FIG. 2 is a flowchart of an item recommendation method according to an embodiment of the present application;

FIG. 3 is a flowchart of determining a current item preference and an operation feedback result corresponding to the current item preference according to an embodiment of the present application;

FIG. 4 is a flowchart of another determination of current item preferences and operational feedback results corresponding to the current item preferences provided by embodiments of the present application;

FIG. 5 is a flowchart of recommending a target item to a target user according to an embodiment of the present application;

FIG. 6 is a flowchart of another method for recommending a target item to a target user according to an embodiment of the present application;

FIG. 7 is a flowchart of another method for recommending items according to an embodiment of the present application;

FIG. 8 is a block diagram of a first article recommendation device according to an embodiment of the present disclosure;

FIG. 9 is a block diagram of a second article recommendation device according to an embodiment of the present disclosure;

FIG. 10 is a block diagram of a third article recommendation device according to an embodiment of the present disclosure;

FIG. 11 is a block diagram of a fourth article recommendation device according to an embodiment of the present disclosure;

FIG. 12 is a block diagram of a fifth article recommendation device according to an embodiment of the present disclosure;

FIG. 13 is a block diagram illustrating a sixth article recommendation device according to an embodiment of the present disclosure;

fig. 14 is an internal structural diagram of a computer device in one embodiment.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application will be further described in detail with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the present application.

It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the present application. In the description of the present application, a description referring to terms "one embodiment," "some embodiments," "examples," "specific examples," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present application. In this specification, schematic representations of the above terms are not necessarily directed to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, the different embodiments or examples described in this specification and the features of the different embodiments or examples may be combined and combined by those skilled in the art without contradiction.

Based on the above situation, the method for recommending items provided in the embodiment of the present application may be applied to an application environment as shown in fig. 1. In one embodiment, a computer device is provided, which may be a server, the internal structure of which may be as shown in FIG. 1. The computer device includes a processor, a memory, and a network interface connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the computer device is used for storing the acquired data of the item recommendation method. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement an item recommendation method.

The application discloses an article recommendation method, an article recommendation device, computer equipment and a storage medium thereof. The computer equipment realizes the accuracy verification of the current item preference by determining the current item preference of the target user and the operation feedback result corresponding to the current item preference; and recommending the target object to the target user according to the accuracy verification result of the current object preference.

In one embodiment, as shown in fig. 2, fig. 2 is a flowchart of an item recommendation method provided in an embodiment of the present application, and an item recommendation method performed by a computer device in fig. 1 may include the following steps:

step 201, determining the current item preference of the target user and the operation feedback result corresponding to the current item preference according to the preference determination mode corresponding to the user type of the target user.

It should be noted that, the user types to which the target user belongs include an added user and a history user; the newly added user refers to a user incapable of directly acquiring the historical article preference and the operation feedback result corresponding to the historical article preference, and the historical user refers to a user capable of acquiring the historical article preference and the operation feedback result corresponding to the historical article preference.

Wherein, the historical item preference refers to: predicting the obtained object preference of the target user according to the historical browsing record of the target user; the operation feedback result corresponding to the historical article preference refers to: after recommending the articles to the target user according to the historical article preference, the click rate, browsing duration, purchase quantity and the like of the target user aiming at the recommended articles.

For example, if the user type to which the target user belongs is a newly added user, the target user may be a user who has just completed registration, or a user who has completed registration but has not generated an operation feedback result corresponding to the historical item preference. If the user type to which the target user belongs is a history user, the target user may be a user who has just completed registration but has generated an operation feedback result corresponding to the history item preference or a user who has completed registration and has generated an operation feedback result corresponding to the history item preference.

Further, in order to ensure that the target item recommendation can be performed to the target user subsequently, determining the current item preference of the target user and an operation feedback result corresponding to the current item preference are required; because the newly added user refers to a user incapable of directly acquiring the historical article preference and the operation feedback result corresponding to the historical article preference, a corresponding preference determination method is required to be determined according to the user types of different target users, so that the current article preference of the target user and the operation feedback result corresponding to the current article preference are determined.

As an implementation manner, since the newly added user refers to a user who cannot directly obtain the historical item preference and the operation feedback result corresponding to the historical item preference, when the user type to which the target user belongs is the newly added user, the corresponding preference determining manner is as follows: determining a reference user corresponding to the target user, taking the historical article preference of the reference user as the current article preference of the target user, and taking an operation feedback result corresponding to the historical article preference of the reference user as an operation feedback result corresponding to the current article preference of the target user.

When a reference user corresponding to a target user needs to be determined, user information of a plurality of candidate users can be screened based on user information of the target user, and candidate users with similarity between the user information and the user information of the target user being greater than a preset similarity threshold value are determined, wherein the candidate users are the reference users;

further, if the similarity between the user information and the user information of the target user is greater than a preset similarity threshold, the candidate user with the maximum similarity between the user information and the user information of the target user is taken as the reference user.

As another implementation manner, since the history user refers to a user that can obtain the history item preference and the operation feedback result corresponding to the history item preference, when the user type to which the target user belongs is the history user, the corresponding preference determining manner is: determining historical article preference of a target user and an operation feedback result corresponding to the historical article preference; updating the historical item preference of the target user through the operation feedback result corresponding to the historical item preference to obtain the current item preference of the target user, and predicting the operation feedback result corresponding to the current item preference based on the historical operation feedback corresponding to the historical item preference and the current item preference of the target user.

When the operation feedback result corresponding to the current article preference needs to be predicted, an operation prediction model can be trained in advance, and the operation feedback result corresponding to the current article preference output by the operation prediction model is obtained by inputting the historical operation feedback corresponding to the historical article preference and the current article preference of the target user into the operation prediction model.

Step 202, verifying the accuracy of the preference of the current item according to the operation feedback result corresponding to the preference of the current item.

It should be noted that, the operation feedback result corresponding to the current article preference refers to: after recommending the item to the target user based on the current item preference, the browsing duration of the target user for recommending the item and the purchase quantity of the target user for recommending the item.

The browsing time of the target user for the recommended items refers to the display time of the recommended items contained in the display interface of the computer equipment of the target user after the recommended items are recommended to the target user based on the current item preferences.

Wherein the purchase amount of the recommended items by the target user refers to the amount of the recommended items purchased by the target user after recommending the items to the target user based on the current item preference.

Further illustratively, in verifying the accuracy of the current item preference, the following are specifically included: after recommending the current article to the target user, setting an operation feedback threshold corresponding to the current article preference based on historical data and working experience of staff; and verifying the accuracy of the operation feedback result corresponding to the current article preference according to the operation feedback threshold corresponding to the current article preference, and further determining the accuracy verification result.

Wherein, the operation feedback threshold value corresponding to the current article preference refers to: a browsing duration threshold and a purchase quantity threshold of the target user for the recommended item, which are set based on the history data and the working experience of the staff member.

In one embodiment of the present application, if the preset time period is longer than the browsing time period threshold only when the target user browses the recommended items, and the purchase quantity of the target user for the recommended items is greater than the purchase quantity threshold, the current item preference verification is considered to pass after the current item preference is verified; when any one of the browsing time length and the purchase quantity of the target user aiming at the recommended article is smaller than the corresponding threshold value, the current article preference is considered to be not passed after the accuracy verification is carried out on the current article preference.

For example, based on historical data and working experience of staff, setting an operation feedback threshold corresponding to the current article preference, wherein the browsing duration threshold in the operation feedback threshold is ten minutes, and the purchase quantity threshold is one; if the operation feedback result corresponding to the current article preference is that the browsing time of the target user for the recommended article is fifteen minutes and the purchase quantity is two, the current article preference is considered to pass after accuracy verification is carried out on the current article preference; if the operation feedback result corresponding to the current item preference is that the browsing duration of the target user for the recommended item is twenty minutes and the purchase quantity is zero, so that the purchase quantity of the target user for the recommended item is smaller than the purchase quantity threshold, the current item preference is considered to be not passed after accuracy verification is performed on the current item preference.

Step 203, if the verification is passed, recommending the target item to the target user based on the current item preference.

When the target item recommendation needs to be performed to the target user, the attribute screening can be performed on the item attributes of at least two candidate items according to the current item preference, so as to determine at least one target item matched with the current item preference in each candidate item and the matching degree of each target item and the current item preference; and recommending the target item to the target user according to at least one target item matched with the current item preference in the candidate items and the matching degree of the target items and the current item preference.

In one embodiment of the application, after determining at least one target item matched with the preference of the current item in each candidate item and the matching degree of each target item and the preference of the current item, sorting each target item according to the matching degree of each target item and the preference of the current item to obtain a recommendation list; and recommending the target object to the target user according to the recommendation list.

For example, if the current item preference indicates that the target user prefers a red item, and the item needs to be related to the mobile phone, the "red item" and the "item related to the mobile phone" are screened in the item attributes of at least two candidate items based on the current item preference, so as to obtain two target items, wherein the two target items are respectively: the target article 1, the target article 2, the target article 3, the target article 4 and the target article 5, wherein the matching degree of each target article and the current article preference is respectively as follows: the matching degree of the target article 1 and the current article preference is 50%, the matching degree of the target article 2 and the current article preference is 60%, the matching degree of the target article 3 and the current article preference is 40%, the matching degree of the target article 4 and the current article preference is 70%, and the matching degree of the target article 5 and the current article preference is 80%, so that the target articles are ranked according to the matching degree of the target articles and the current article preference, and a recommendation list of the target article 5-target article 4-target article 2-target article 1-target article 3 is obtained.

According to the item recommending method, based on the preference determining mode of the user type of the target user, the current item preference of the target user and the operation feedback result corresponding to the current item preference are determined; and recommending the target object to the target user according to the accuracy verification result of the current object preference. In the process, the target object is recommended to the target user according to the accuracy verification result of the current object preference, so that the user does not need to find the object of interest through a traversal searching method, the object of interest can be obtained only according to the recommended content of the target object, the time wasted when the user searches the object is reduced, and the purchasing experience of the user is optimized.

With the increasing number of selectable articles on the internet and the various article attributes of each article, the user wants to find the article of interest, and needs to perform traversal search on the article attribute of each article by a traversal search method, so that the time consumed by the user to find the article of interest is also increased gradually; in order to prevent the poor purchasing experience of the user caused by the above problem, the computer device of the embodiment may determine, according to a preference determining manner corresponding to the user type to which the target user belongs, a current item preference of the target user and an operation feedback result corresponding to the current item preference in a manner as shown in fig. 3, and specifically includes the following steps:

Step 301, if the preference determining mode corresponding to the user type to which the target user belongs is a first round determining mode, determining a reference user corresponding to the target user based on user information of the target user, wherein a similarity between the user information of the reference user and the user information of the target user is greater than a similarity threshold.

It should be noted that, if the user type to which the target user belongs is the newly added user, the preference determination mode corresponding to the user type to which the target user belongs is the first round determination mode; it can be understood that: the first round of determining mode is a preference determining mode aiming at a target user with a user type being an added user.

The user information of the target user may include, but is not limited to: the login location of the target user, the age of the target user, the sex of the target user, etc.

When the reference user corresponding to the target user needs to be determined, the user information of the candidate user may be filtered based on the user information of the target user, and the candidate user whose user information similarity with the user information of the target user is greater than the similarity threshold may be used as the reference user.

The candidate users are users with known user information, historical article preferences and operation feedback results corresponding to the historical article preferences except the target users. The similarity threshold may be set according to the history experience of the worker in combination with the use scenario, and the method for setting the similarity threshold and the value of the similarity threshold are not limited.

In one embodiment of the present application, if there are a plurality of candidate users whose user information similarity with the target user is greater than a similarity threshold, the plurality of candidate users are ranked according to the similarity between the user information of the candidate users and the user information of the target user, and the candidate user with the highest similarity between the user information and the user information of the target user is determined, and the candidate user is used as a reference user.

For example, based on the user information of the target user, the user information of the candidate users is screened, and it is determined that the similarity between the user information of three candidate users and the user information of the target user is greater than a similarity threshold, where the three candidate users are respectively candidate user 1, candidate user 2 and candidate user 3, and the similarity between the user information of the three candidate users and the user information of the target user is respectively: and (3) determining candidate user 3 with highest similarity between the user information and the user information of the target user by sequencing the three candidate users according to the similarity between the user information of the three candidate users and the user information of the target user, wherein the candidate user 1 is 80%, the candidate user 2 is 85% and the candidate user 3 is 90%.

Step 302, taking historical item preference of a reference user as current item preference of a target user; wherein the historical item preference is the item preference last determined for the reference user.

Wherein, the historical item preference of the reference user refers to predicting the item preference of the reference user according to the historical purchase record and the historical browsing record of the reference user.

It should be noted that there are many methods when it is necessary to determine the historic item preference of the reference user, for example: the preference prediction model can be trained in advance, and the historical article preference of the reference user is determined according to the preference prediction model, or the historical article preference of the reference user is predicted according to the historical purchase record and the historical browsing record of the reference user; in summary, there are many methods for determining the preference of the historical item of the reference user, which will not be described in detail herein, and the two methods for determining the preference of the historical item of the reference user will be described in detail below:

as one implementation, when it is desired to determine the historical item preferences of the reference user, a preference prediction model may be trained based on the sample item preferences, sample purchase records, and sample browse records of the sample user; and inputting the historical purchase record and the historical browsing record of the reference user into the preference prediction model to obtain an output result of the preference prediction model, wherein the output result is the historical article preference of the reference user.

As another implementation manner, when the historical item preference of the reference user needs to be determined, the historical purchasing record and the historical browsing record of the reference user are obtained, the items purchased by the reference user and the items with more browsing times are screened out according to the historical purchasing record and the historical browsing record of the reference user, and further, the historical item preference of the reference user is determined according to the purchased items and the items with more browsing times.

And 303, taking an operation feedback result corresponding to the historical article preference of the reference user as an operation feedback result corresponding to the current article preference.

Wherein, the operation feedback result corresponding to the historical article preference of the reference user is as follows: after recommending the item to the reference user based on the historical item preference, the browsing duration of the reference user for recommending the item and the purchase quantity of the target user for recommending the item.

In one implementation of the present application, at least one recommended item that matches the historical item preferences of the reference user may be determined from among the candidate items in advance based on the historical item preferences; recommending the recommended articles to the reference user, and acquiring the browsing time length of the reference user for the recommended articles and the purchase quantity of the target user for the recommended articles, so as to determine the operation feedback result corresponding to the historical article preference of the reference user.

According to the item recommending method, when the preference determining mode corresponding to the user type of the target user is the first-round determining mode, the operation feedback result corresponding to the current item preference and the current item preference of the target user is determined according to the historical item preference and the operation feedback result corresponding to the historical item preference of the reference user by determining the reference user; a judgment basis is provided for the follow-up recommendation of the target object to the target user, and the accuracy of the recommendation of the target object is ensured.

In one embodiment, if the preference determining mode corresponding to the user type to which the target user belongs is a non-first round determining mode, determining, according to the preference determining mode corresponding to the user type to which the target user belongs, a current item preference of the target user and an operation feedback result corresponding to the current item preference, as shown in fig. 4, the method includes:

step 401, if the preference determining mode corresponding to the user type to which the target user belongs is a non-first round determining mode, determining a historical article preference of the target user and a historical operation feedback corresponding to the historical article preference.

It should be noted that, if the user type to which the target user belongs is a history user, the preference determination mode corresponding to the user type to which the target user belongs is a non-first-round determination mode; it can be understood that: the non-first round determination mode is a preference determination mode for a target user whose user type is a history user.

Further, since the history user refers to a user who can acquire the history item preference and the operation feedback result corresponding to the history item preference, a user who acquires the operation feedback result corresponding to at least one history item preference and each history item preference of the target user can be used. Further, the user who has recently performed the operation feedback result corresponding to the history item preference and the history item preference is used as the desired history item preference of the target user and the history operation feedback corresponding to the history item preference.

In one implementation of the present application, when at least one historical item preference of a target user and an operation feedback result corresponding to each historical item preference need to be determined, a record of a recommended item of each historical recommendation of the target user can be determined, and because the recommended item of each historical recommendation is determined according to the corresponding historical item preference, item attribute statistics can be performed on the recommended item of each historical recommendation, and the obtained statistical result is the historical item preference corresponding to each historical recommendation; and acquiring the browsing time length and the purchase quantity of the recommended articles of each historical recommendation of the target user, and further determining historical operation feedback corresponding to each historical article preference of the target user according to the browsing time length and the purchase quantity of the recommended articles of each historical recommendation of the target user.

The browsing time length and the purchase quantity of the recommended articles which are recommended by the target user for each history can be obtained by obtaining the article browsing record and the article purchase record of the target user.

Step 402, updating the historical item preference of the target user based on the historical operation feedback corresponding to the historical item preference, so as to obtain the current item preference of the target user.

It should be noted that, because the historical operation feedback corresponding to the historical article preference is the operation feedback of the target user for recommending articles after recommending articles to the target user based on the historical article preference, whether the historical article preference accords with the actual preference of the user can be verified through the historical operation feedback corresponding to the historical article preference.

In one implementation of the application, if the browsing time of the target user for the recommended article in the historical operation feedback corresponding to the historical article preference is longer than a preset browsing time threshold, and the purchase quantity of the target user for the recommended article is greater than the purchase quantity threshold, the historical article preference is considered to be in accordance with the actual preference of the user; if the browsing duration of the target user for the recommended articles in the historical operation feedback corresponding to the historical article preferences is smaller than or equal to a preset browsing duration threshold value, and/or the purchase quantity of the target user for the recommended articles is smaller than or equal to a purchase quantity threshold value, the historical article preferences are considered to be not in accordance with the actual preferences of the users.

Further, when the historical item preference does not accord with the actual preference of the user, updating the historical item preference of the target user based on the historical operation feedback corresponding to the historical item preference to obtain the current item preference of the target user. Specifically, according to historical operation feedback corresponding to historical article preference, an article of interest and an article of no interest of a target user are determined, and article attributes of the article of interest and article attributes of no interest are determined; if the historical item preference does not contain the item attribute of the item of interest, adding the item attribute of the item of interest to the historical item preference; if the historical item preference contains the item attribute of the uninteresting item, removing the item attribute of the uninteresting item from the historical item preference; and according to the content, finishing the updating operation of the historical article preference of the target user to obtain the current article preference of the target user.

Step 403, predicting an operation feedback result corresponding to the current item preference based on the historical operation feedback corresponding to the historical item preference and the current item preference of the target user.

It should be noted that, the operation prediction model may be trained in advance, so that the historical operation feedback corresponding to the historical article preference and the current article preference of the target user are input into the operation prediction model, and the prediction result output by the operation prediction model is obtained, and the result is the operation feedback result corresponding to the current article preference.

Further, the operation prediction model can be trained in advance according to a plurality of historical article preferences of the target user and historical operation feedback corresponding to each historical article recommendation, and when the prediction accuracy of the operation prediction model reaches the accuracy requirement of a worker, the operation of training the operation prediction model is completed.

According to the item recommending method, if the preference determining mode corresponding to the user type of the target user is a non-first-round determining mode, according to the historical item preference of the target user and the historical operation feedback corresponding to the historical item preference, determining the current item preference of the target user and the operation feedback result corresponding to the current item preference; a judgment basis is provided for the follow-up recommendation of the target object to the target user, and the accuracy of the recommendation of the target object is ensured.

In one embodiment, when the current item preference is verified for accuracy and verification is not passed, then the method includes, as shown in FIG. 5:

step 501, if the verification is not passed, updating the current item preference according to the operation feedback result corresponding to the current item preference, so as to obtain the updated current item preference.

It should be noted that, when the current item preference needs to be updated, the following may be specifically included: according to the operation feedback result corresponding to the current article preference, determining the interested articles and the uninteresting articles of the target user; determining article attributes of the article of interest and the article not of interest; updating the current item preference based on the item attributes of the interested items and the uninteresting items to obtain the updated current item preference.

Wherein the item attributes may include, but are not limited to: production time of the article, color of the article, kind of the article, applicable population of the article, etc.

Further, when the current item preference is updated based on the item attributes of the item of interest and the item of no interest, whether the item attributes of the item of interest are contained in the current item preference or not can be judged, if so, the processing is not performed, and if not, the item attributes of the item of interest are added into the current item preference; and judging whether the current item preference contains the item attribute of the uninteresting item, if not, not processing, and if so, removing the item attribute of the uninteresting item from the current item preference.

Step 502, recommending the target item to the target user based on the updated current item preference.

When recommending the target item to the target user, the target item with the item attribute belonging to the updated current item preference can be screened out according to the updated current item preference, and the target item is recommended to the target user.

Further, after screening out the target articles with the article attribute belonging to the updated current article preference, sequencing the target articles according to the loading time of the target articles, and recommending the target articles to the curtain user according to the sequencing result, wherein the closer the loading time is to the current moment, the earlier the article recommendation sequence corresponding to the loading time is.

For example, if three target items are included, the three target items are the target item 1, the target item 2, and the target item 3, and the time to put on the shelf of the three target items is: the target object 1 is put on shelf on one month, the target object 2 is put on shelf on fifteen months, and the target object 3 is put on shelf on twenty months, because the closer the time of putting on shelf is to the current moment, the more the object recommendation sequence corresponding to the time of putting on shelf is, therefore, the recommendation sequence of the three target objects is as follows: target object 3-target object 2-target object 1.

According to the item recommending method, when the accuracy verification of the current item preference is not passed, the updated current item preference is enabled to be more accordant with the item preference of the target user through updating the current item preference, so that the user does not need to find the item of interest through a traversing searching method, the item of interest can be obtained only according to the recommended content of the target item, the time wasted when the user searches the item is reduced, and the purchasing experience of the user is optimized.

In one embodiment, when making a target item recommendation to a target user based on current item preferences, the method includes, as shown in FIG. 6:

At step 601, item attributes of at least two candidate items are determined.

It should be noted that, since the item attribute is used to represent the item attribute of the candidate item, when the item attribute of the candidate item needs to be determined, the item attribute of at least two candidate items may be determined by acquiring the item attribute of the candidate item from various channels.

As an implementation way, the processing raw materials, the production date and other relevant information of the candidate users can be obtained through the production instruction book of the candidate articles; for example, the raw materials and date of production of the food are obtained.

Step 602, screening the attributes of the candidate items based on the current item preferences to obtain screening results, wherein the screening results comprise at least one target item matched with the current item preferences in the candidate items and the matching degree of the target items and the current item preferences.

When the attribute screening needs to be performed on the item attribute of each candidate item, whether the current item preference contains the item attribute of each candidate item or not can be determined, if the current item preference contains the item attribute of a certain candidate item, the candidate item is taken as a target item; and, the greater the number of item attributes of the target item contained in the current item preference, the higher the degree of matching of the target item with the current item preference.

For example, if there are two target objects, the target object a and the target object b, respectively, wherein the object attributes of the target object a are: attribute a, attribute B, and attribute C, the object attributes of the target object B are: attribute D, attribute E, and attribute F, the current item preferences are known to include: attribute a, attribute D, and attribute B; therefore, if the number of item attributes including the target item a in the current item preference is greater than the number of item attributes including the target item b in the current item preference, the matching degree of the target item a and the current item preference is greater than the matching degree of the target item b and the current item preference.

In one embodiment of the present application, the calculated percentage may be used as the matching degree of each target item and the current item preference by calculating the percentage of the portion of the current item preference containing the item attribute of each target item to the whole of the current item preference.

And 603, recommending the target object to the target user based on the screening result.

When the target item recommendation is required to be performed to the target user, the target items can be sequenced according to the matching degree of the target items and the preference of the current item in the screening result, and then the target item recommendation is performed to the target user according to the sequencing result, so that the target item with higher matching degree with the preference of the current item is arranged in front, and then the recommendation is performed to the target user.

For example, if there are three target items, namely, target item c, target item d, and target item e, the matching degree of the three target items and the current item preference is: the matching degree of the target article c is fifty percent, the matching degree of the target article d is sixty percent and the matching degree of the target article e is seventy percent, so that the three target articles are sequenced according to the matching degree of the three target articles and the preference of the current article, and the obtained result is that: target item e-target item d-target item c; accordingly, the target item recommendation is performed to the target user in the order of the target item e-target item d-target item c.

According to the article recommending method, the screening result is obtained by screening the attributes in the machine boxes of the candidate articles, so that the follow-up object article recommending can be carried out on the target users according to the screening result; the accuracy of recommending the target object to the target user is guaranteed, the time wasted when the user searches the object is reduced, and the purchase experience of the user is optimized.

In one embodiment, when the target item recommendation needs to be made to the target user, as shown in fig. 7, the following may be specifically included:

Step 701, beginning. If the preference determination mode corresponding to the user type to which the target user belongs is the first round determination mode, executing step 702; if the preference determination mode corresponding to the user type to which the target user belongs is a non-first round determination mode, step 705 is executed.

Step 702, determining a reference user corresponding to the target user based on the user information of the target user, wherein the similarity between the user information of the reference user and the user information of the target user is greater than a similarity threshold.

Step 703, taking the historical item preference of the reference user as the current item preference of the target user; wherein the historical item preference is the item preference last determined for the reference user.

And step 704, taking the operation feedback result corresponding to the historical article preference of the reference user as the operation feedback result corresponding to the current article preference.

Step 705, determining historical item preference of the target user and historical operation feedback corresponding to the historical item preference.

Step 706, updating the historical item preference of the target user based on the historical operation feedback corresponding to the historical item preference, so as to obtain the current item preference of the target user.

Step 707, based on the historical operation feedback corresponding to the historical item preference and the current item preference of the target user, predicting an operation feedback result corresponding to the current item preference.

Step 708, verifying the accuracy of the current item preference according to the operation feedback result corresponding to the current item preference. If the verification is passed, step 709 is performed; if the verification is not passed, step 712 is performed.

In step 709, item attributes of at least two candidate items are determined.

And step 710, carrying out attribute screening on the item attributes of each candidate item based on the current item preference to obtain screening results.

Step 711, recommending the target object to the target user based on the screening result.

And step 712, determining the interested articles and the uninteresting articles of the target user according to the operation feedback result corresponding to the current article preference.

In step 713, item attributes for items of interest and items of no interest are determined.

Step 714, updating the current item preference based on the item attributes of the item of interest and the item of no interest, resulting in an updated current item preference.

Step 715, recommending the target item to the target user based on the updated current item preference.

It should be understood that, although the steps in the flowcharts related to the above embodiments are sequentially shown as indicated by arrows, these steps are not necessarily sequentially performed in the order indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in the flowcharts described in the above embodiments may include a plurality of steps or a plurality of stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of the steps or stages is not necessarily performed sequentially, but may be performed alternately or alternately with at least some of the other steps or stages.

Based on the same inventive concept, the embodiment of the application also provides an article recommending device for realizing the article recommending method. The implementation of the solution provided by the device is similar to the implementation described in the above method, so the specific limitation of the embodiment of the one or more article recommendation devices provided below may refer to the limitation of the article recommendation method hereinabove, and will not be repeated herein.

In one embodiment, as shown in fig. 8, there is provided an item recommendation device, comprising: a determination module 10, a verification module 20 and a first recommendation module 30, wherein:

the determining module 10 is configured to determine, according to a preference determining manner corresponding to a user type to which the target user belongs, a current item preference of the target user and an operation feedback result corresponding to the current item preference.

And the verification module 20 is used for verifying the accuracy of the current article preference according to the operation feedback result corresponding to the current article preference.

The first recommendation module 30 is configured to, if the verification is passed, recommend the target item to the target user based on the current item preference.

In one embodiment, as shown in fig. 9, there is provided an item recommendation device in which the determining module 10 includes: a first determination unit 11, a second determination unit 12, and a third determination unit 13, wherein:

the first determining unit 11 is configured to determine, based on user information of the target user, a reference user corresponding to the target user if the preference determining mode corresponding to the user type to which the target user belongs is a first round determining mode, where similarity between the user information of the reference user and the user information of the target user is greater than a similarity threshold.

A second determining unit 12 for taking the historical item preference of the reference user as the current item preference of the target user; wherein the historical item preference is the item preference last determined for the reference user.

And a third determining unit 13, configured to take an operation feedback result corresponding to the historical article preference of the reference user as an operation feedback result corresponding to the current article preference.

In one embodiment, as shown in fig. 10, there is provided an item recommendation device in which the determining module 10 includes: a fourth determination unit 14, a fifth determination unit 15, and a sixth determination unit 16, wherein:

and a fourth determining unit 14, configured to determine a historical item preference of the target user and a historical operation feedback corresponding to the historical item preference if the preference determining mode corresponding to the user type to which the target user belongs is a non-first round determining mode.

And a fifth determining unit 15, configured to update the historical item preference of the target user based on the historical operation feedback corresponding to the historical item preference, so as to obtain the current item preference of the target user.

A sixth determining unit 16, configured to predict an operation feedback result corresponding to the current item preference based on the historical operation feedback corresponding to the historical item preference and the current item preference of the target user.

In one embodiment, as shown in fig. 11, there is provided an item recommendation device, the item recommendation device further comprising: an update module 40 and a second recommendation module 50, wherein:

and the updating module 40 is configured to update the current item preference according to the operation feedback result corresponding to the current item preference if the verification is not passed, so as to obtain an updated current item preference.

The second recommendation module 50 is configured to make a target item recommendation to the target user based on the updated current item preference.

In one embodiment, as shown in fig. 12, there is provided an item recommendation device in which the update module 40 includes: a seventh determination unit 41, an eighth determination unit 42, and an update unit 43, wherein:

a seventh determining unit 41, configured to determine, according to the operation feedback result corresponding to the current item preference, an item of interest and an item of no interest of the target user.

An eighth determining unit 42 is configured to determine an item attribute of the item of interest and the item of non-interest.

The updating unit 43 is configured to update the current item preference based on the item attributes of the item of interest and the item of no interest, and obtain the updated current item preference.

In one embodiment, as shown in fig. 13, there is provided an item recommending apparatus in which a first recommending module 30 includes: a seventh determination unit 41, an eighth determination unit 42, and an update unit 43, wherein:

a ninth determining unit 31 is configured to determine item attributes of at least two candidate items.

And a screening unit 32, configured to perform attribute screening on the item attribute of each candidate item based on the current item preference, to obtain a screening result, where the screening result includes at least one target item matching the current item preference in each candidate item, and a matching degree between each target item and the current item preference.

And a recommending unit 33, configured to recommend the target item to the target user based on the screening result.

The respective modules in the above-described item recommendation apparatus may be implemented in whole or in part by software, hardware, and combinations thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.

In one embodiment, a computer device is provided, which may be a terminal, and the internal structure thereof may be as shown in fig. 14. The computer device includes a processor, a memory, an input/output interface, a communication interface, a display unit, and an input means. The processor, the memory and the input/output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input/output interface. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The input/output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used for carrying out wired or wireless communication with an external terminal, and the wireless mode can be realized through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. The computer program is executed by a processor to implement an item recommendation method. The display unit of the computer device is used for forming a visual picture, and can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, can also be a key, a track ball or a touch pad arranged on the shell of the computer equipment, and can also be an external keyboard, a touch pad or a mouse and the like.

It will be appreciated by those skilled in the art that the structure shown in fig. 14 is merely a block diagram of a portion of the structure associated with the present application and is not limiting of the computer device to which the present application applies, and that a particular computer device may include more or fewer components than shown, or may combine certain components, or have a different arrangement of components.

In one embodiment, a computer device is provided comprising a memory and a processor, the memory having stored therein a computer program, the processor when executing the computer program performing the steps of:

In one embodiment, the processor when executing the computer program further performs the steps of:

determining item attributes of at least two candidate items;

In one embodiment, a computer readable storage medium is provided having a computer program stored thereon, which when executed by a processor, performs the steps of:

In one embodiment, the computer program when executed by the processor further performs the steps of:

determining item attributes of at least two candidate items;

In one embodiment, a computer program product is provided comprising a computer program which, when executed by a processor, performs the steps of:

determining item attributes of at least two candidate items;

It should be noted that, the user information (including, but not limited to, user equipment information, user personal information, etc.) and the data (including, but not limited to, data for analysis, stored data, presented data, etc.) referred to in the present application are information and data authorized by the user or sufficiently authorized by each party, and the collection, use and processing of the related data are required to comply with the related laws and regulations and standards of the related countries and regions.

Those skilled in the art will appreciate that implementing all or part of the above-described methods in accordance with the embodiments may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed may comprise the steps of the embodiments of the methods described above. Any reference to memory, database, or other medium used in the various embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, high density embedded nonvolatile Memory, resistive random access Memory (ReRAM), magnetic random access Memory (Magnetoresistive Random Access Memory, MRAM), ferroelectric Memory (Ferroelectric Random Access Memory, FRAM), phase change Memory (Phase Change Memory, PCM), graphene Memory, and the like. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory, and the like. By way of illustration, and not limitation, RAM can be in the form of a variety of forms, such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), and the like. The databases referred to in the various embodiments provided herein may include at least one of relational databases and non-relational databases. The non-relational database may include, but is not limited to, a blockchain-based distributed database, and the like. The processors referred to in the embodiments provided herein may be general purpose processors, central processing units, graphics processors, digital signal processors, programmable logic units, quantum computing-based data processing logic units, etc., without being limited thereto.

The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.

The foregoing examples represent only a few embodiments of the present application, which are described in more detail and are not thereby to be construed as limiting the scope of the present application. It should be noted that it would be apparent to those skilled in the art that various modifications and improvements could be made without departing from the spirit of the present application, which would be within the scope of the present application. Accordingly, the scope of protection of the present application shall be subject to the appended claims.

Claims

1. A method of recommending items, the method comprising:

determining the current article preference of a target user and an operation feedback result corresponding to the current article preference according to a preference determination mode corresponding to the user type of the target user;

2. The method according to claim 1, wherein the determining, according to a preference determining manner corresponding to a user type to which the target user belongs, a current item preference of the target user and an operation feedback result corresponding to the current item preference includes:

if the preference determining mode corresponding to the user type of the target user is a first-round determining mode, determining a reference user corresponding to the target user based on the user information of the target user, wherein the similarity between the user information of the reference user and the user information of the target user is larger than a similarity threshold;

taking the historical item preference of the reference user as the current item preference of the target user; wherein the historical item preference is an item preference that was last determined for a reference user;

and taking an operation feedback result corresponding to the historical article preference of the reference user as an operation feedback result corresponding to the current article preference.

3. The method according to claim 1, wherein the determining, according to a preference determining manner corresponding to a user type to which the target user belongs, a current item preference of the target user and an operation feedback result corresponding to the current item preference further includes:

4. The method according to claim 1, wherein the method further comprises:

if the verification is not passed, updating the current article preference according to the operation feedback result corresponding to the current article preference to obtain the updated current article preference;

5. The method of claim 4, wherein updating the current item preference according to the operation feedback result corresponding to the current item preference to obtain an updated current item preference, comprises:

6. The method of claim 1, wherein the making a target item recommendation to the target user based on the current item preference comprises:

determining item attributes of at least two candidate items;

7. An item recommendation device, the device comprising:

The verification module is used for verifying the accuracy of the current article preference according to the operation feedback result corresponding to the current article preference;

8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any of claims 1 to 6 when the computer program is executed.

9. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 6.

10. A computer program product comprising a computer program, characterized in that the computer program, when being executed by a processor, realizes the steps of the method according to any one of claims 1 to 6.