CN109711875B

CN109711875B - Content recommendation method and device

Info

Publication number: CN109711875B
Application number: CN201811552763.1A
Authority: CN
Inventors: 李江
Original assignee: Koubei Shanghai Information Technology Co Ltd
Current assignee: Koubei Shanghai Information Technology Co Ltd
Priority date: 2018-12-19
Filing date: 2018-12-19
Publication date: 2021-04-20
Anticipated expiration: 2038-12-19
Also published as: CN109711875A

Abstract

The invention discloses a content recommendation method and device. The content recommendation method comprises the following steps: determining a target user, and inquiring a user requirement grade of the target user prestored in a database; determining target content of at least one store to be recommended, and inquiring the store level of the at least one store prestored in the database; matching the user requirement grade of the target user with the store grade of at least one store, and selecting target content to be recommended from the target content of the at least one store according to a matching result; and pushing the target content to be recommended to the target user. Based on the scheme provided by the invention, the appropriate target content to be recommended is recommended to the user according to the user requirement level, the interest degree of the target content to be recommended of the user can be improved, and the possibility of further knowing or purchasing by the user is greatly improved, so that the marketing flow of stores is further improved, the waste proportion of the marketing flow is reduced, and the pushing effect is improved.

Description

Content recommendation method and device

Technical Field

The invention relates to the technical field of internet, in particular to a content recommendation method and device.

Background

With the continuous development of internet technology, various platforms, such as shopping platforms, takeout platforms, car rental platforms, etc., have come to be used. More and more merchants choose to set up stores on the platform and promote correspondingly on the platform, for example, advertisements or commodities are presented on corresponding pages of the platform, and more users choose to select resources required by themselves, such as commodities or services, on the platforms.

Taking the example of presenting advertisements on a page, the traditional advertisement delivery to a page is a broadcast network type, and the advertisement delivered to a certain page or a certain advertisement position of a certain page within a certain time period is fixed, such as: the kentucky 5-fold advertisement can be seen by all users when they enter a certain page for a certain period of time, but is not interested by all users, and therefore much marketing traffic is wasted.

Disclosure of Invention

In view of the above, the present invention has been made to provide a content recommendation method and apparatus that overcomes or at least partially solves the above problems.

According to an aspect of the present invention, there is provided a content recommendation method including:

determining a target user, and inquiring a user requirement grade of the target user prestored in a database;

determining target content of at least one store to be recommended, and inquiring the store level of the at least one store prestored in the database;

matching the user requirement grade of the target user with the store grade of at least one store, and selecting target content to be recommended from the target content of the at least one store according to a matching result;

and pushing the target content to be recommended to the target user.

Optionally, matching the user requirement level of the target user with the store level of at least one store, and selecting the target content to be recommended from the target content of the at least one store according to the matching result further includes: judging whether the user requirement grade of the target user is matched with the store grade of any store; if so, selecting the target content of the matched store as the target content to be recommended; and if not, selecting the target content of the store with the store grade higher than the grade required by the user as the target content to be recommended.

Optionally, before determining the target user, the method further comprises: and acquiring evaluation information, determining the store grade and the user requirement grade according to the evaluation information, and storing the store grade and the user requirement grade in a database.

Optionally, the determining the store level and the user requirement level according to the evaluation information further comprises:

determining a cluster of stores to be evaluated, and calculating the grade of the 1 st round of stores of the stores according to the evaluation information of the stores aiming at each store in the cluster of stores to be evaluated;

determining a user cluster to be evaluated, and calculating the 1 st round user requirement level of each user in the user cluster to be evaluated according to the evaluation information of the user;

taking the 1 st round of store level and the 1 st round of user requirement level as initial values, performing iterative adjustment on the initial values according to evaluation information of each store in the to-be-rated store cluster of each user in the to-be-rated user cluster, and finally determining the store level of each store in the to-be-rated store cluster and the user requirement level of each user in the to-be-rated user cluster.

Optionally, performing a round of iterative adjustment on the initial value according to evaluation information of each store in the to-be-rated store cluster of each user in the to-be-rated user cluster, and finally determining the store level of each store in the to-be-rated store cluster and the user requirement level of each user in the to-be-rated user cluster further includes: starting from the point where i is 2,

s1, aiming at each store in the store cluster to be evaluated, calculating the i-th round of store grade of the store according to the i-1 th round of user requirement grade and the evaluation information of the store, and aiming at each user in the user cluster to be evaluated, calculating the i-th round of user requirement grade of the user according to the i-1 th round of store grade and the evaluation information of the user;

s2, judging whether the ith round of store grade of each store is the same as the ith-1 round of store grade and whether the ith round of user requirement grade of each user is the same as the ith-1 round of user requirement grade;

s3, if yes, the ith round of store grade of each store is finally determined as the store grade of each store, the ith round of user requirement grade of each user is finally determined as the user requirement grade of each user, and the iterative adjustment process is ended;

s4, if not, assigning i as i +1, and jumping to execute step S1.

Optionally, for each store in the cluster of stores to be rated, calculating the ith store rating of the store according to the (i-1) th round of user requirement rating and the evaluation information of the store further includes:

determining an evaluation level of each piece of evaluation information of the store; classifying each piece of evaluation information according to a first classification rule according to the evaluation level of each piece of evaluation information and the i-1 th round user requirement level of the user corresponding to each piece of evaluation information; and calculating the ith round of store level of the store by using a first preset rule based on the classification result of each piece of evaluation information of the store.

Optionally, calculating the ith round of store level of the store by using the first preset rule based on the classification result of each piece of evaluation information of the store further comprises: and calculating the number ratio of the evaluation information in at least one preset category obtained after classification according to the first classification rule, and calculating the ith round of store level of the stores according to the number ratio.

Optionally, for each user in the user cluster to be evaluated, calculating the ith round of user requirement level of the user according to the ith-1 round of store level and the evaluation information of the user further includes:

determining the evaluation level of each piece of evaluation information of the user; classifying each piece of evaluation information according to a second classification rule according to the evaluation level of each piece of evaluation information and the grade of the (i-1) th round of store corresponding to each piece of evaluation information; and calculating the requirement grade of the ith round of the user by utilizing a second preset rule based on the classification result of each piece of evaluation information of the user.

Optionally, calculating an ith round of user requirement rating of the user by using a second preset rule based on the classification result of each piece of evaluation information of the user further includes: and calculating the number ratio of the evaluation information in at least one preset category obtained after classification according to the second classification rule, and calculating the ith round of user requirement grade of the user according to the number ratio.

Optionally, the determining the group of stores to be rated further comprises: judging whether the number of the evaluation information of the store is greater than or equal to a first preset threshold value or not; if yes, adding the store into a store cluster to be evaluated; and if not, allocating a default store grade for the store.

Optionally, the determining the user cluster to be evaluated further comprises: judging whether the number of the evaluation information of the user is greater than or equal to a second preset threshold value or not; if so, adding the user into a user cluster to be evaluated; and if the evaluation information of any user in the user cluster to be evaluated does not belong to the evaluation information aiming at each store in the store cluster to be evaluated, allocating a default user requirement level for the user.

According to another aspect of the present invention, there is provided a content recommendation apparatus including:

a first determination module adapted to determine a target user;

the first query module is suitable for querying the user requirement level of the target user prestored in the database;

the second determination module is suitable for determining target content of at least one store to be recommended;

the second query module is suitable for querying the store grade of at least one store prestored in the database;

the processing module is suitable for matching the user requirement grade of the target user with the store grade of at least one store, and selecting target content to be recommended from the target content of the at least one store according to a matching result;

and the pushing module is suitable for pushing the target content to be recommended to the target user.

Optionally, the processing module is further adapted to: judging whether the user requirement grade of the target user is matched with the store grade of any store; if so, selecting the target content of the matched store as the target content to be recommended; and if not, selecting the target content of the store with the store grade higher than the grade required by the user as the target content to be recommended.

Optionally, the apparatus further comprises: the third determining module is suitable for acquiring evaluation information and determining the store grade and the user requirement grade according to the evaluation information;

and the database is suitable for storing the store grades and the user requirement grades.

Optionally, the third determining module is further adapted to: determining a cluster of stores to be evaluated, and calculating the grade of the 1 st round of stores of the stores according to the evaluation information of the stores aiming at each store in the cluster of stores to be evaluated;

Optionally, the third determination module is further adapted to perform iterative adjustment according to the following procedure: starting from the point where i is 2,

s4, if not, assigning i as i +1, and jumping to execute S1.

Optionally, the third determining module is further adapted to: determining an evaluation level of each piece of evaluation information of the store; classifying each piece of evaluation information according to a first classification rule according to the evaluation level of each piece of evaluation information and the i-1 th round user requirement level of the user corresponding to each piece of evaluation information; and calculating the ith round of store level of the store by using a first preset rule based on the classification result of each piece of evaluation information of the store.

Optionally, the third determining module is further adapted to: and calculating the number ratio of the evaluation information in at least one preset category obtained after classification according to the first classification rule, and calculating the ith round of store level of the stores according to the number ratio.

Optionally, the third determining module is further adapted to: determining the evaluation level of each piece of evaluation information of the user; classifying each piece of evaluation information according to a second classification rule according to the evaluation level of each piece of evaluation information and the grade of the (i-1) th round of store corresponding to each piece of evaluation information; and calculating the requirement grade of the ith round of the user by utilizing a second preset rule based on the classification result of each piece of evaluation information of the user.

Optionally, the third determining module is further adapted to: and calculating the number ratio of the evaluation information in at least one preset category obtained after classification according to the second classification rule, and calculating the ith round of user requirement grade of the user according to the number ratio.

Optionally, the third determining module is further adapted to: judging whether the number of the evaluation information of the store is greater than or equal to a first preset threshold value or not; if yes, adding the store into a store cluster to be evaluated; and if not, allocating a default store grade for the store.

Optionally, the third determining module is further adapted to: judging whether the number of the evaluation information of the user is greater than or equal to a second preset threshold value or not; if so, adding the user into a user cluster to be evaluated; and if the evaluation information of any user in the user cluster to be evaluated does not belong to the evaluation information aiming at each store in the store cluster to be evaluated, allocating a default user requirement level for the user.

According to yet another aspect of the present invention, there is provided a computing device comprising: the processor, the memory and the communication interface complete mutual communication through the communication bus;

the memory is used for storing at least one executable instruction, and the executable instruction enables the processor to execute the operation corresponding to the content recommendation method.

According to still another aspect of the present invention, there is provided a computer storage medium having at least one executable instruction stored therein, the executable instruction causing a processor to perform operations corresponding to the content recommendation method.

According to the scheme provided by the invention, a target user is determined, and the user requirement grade of the target user prestored in a database is inquired; determining target content of at least one store to be recommended, and inquiring the store level of the at least one store prestored in the database; matching the user requirement grade of the target user with the store grade of at least one store, and selecting target content to be recommended from the target content of the at least one store according to a matching result; and pushing the target content to be recommended to the target user. Based on the scheme provided by the invention, the appropriate target content to be recommended is recommended to the user according to the user requirement level, the interest degree of the target content to be recommended of the user can be improved, and the possibility of further knowing or purchasing by the user is greatly improved, so that the marketing flow of stores is further improved, the waste proportion of the marketing flow is reduced, and the pushing effect is improved.

The foregoing description is only an overview of the technical solutions of the present invention, and the embodiments of the present invention are described below in order to make the technical means of the present invention more clearly understood and to make the above and other objects, features, and advantages of the present invention more clearly understandable.

Drawings

Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to refer to like parts throughout the drawings. In the drawings:

FIG. 1 shows a flow diagram of a content recommendation method according to one embodiment of the invention;

FIG. 2 is a flow diagram illustrating a store level and user demand level calculation method according to one embodiment of the present invention;

FIG. 3 is a schematic flow chart diagram illustrating a method of calculating an ith round of store ratings in an embodiment of the present invention;

FIG. 4 is a flow chart illustrating a method for calculating the ith round of user requirement rating in an embodiment of the present invention;

fig. 5 is a schematic structural diagram showing a content recommendation apparatus according to an embodiment of the present invention;

FIG. 6 shows a schematic structural diagram of a computing device according to one embodiment of the invention.

Detailed Description

Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

Fig. 1 shows a flow diagram of a content recommendation method according to an embodiment of the invention. As shown in fig. 1, the method comprises the steps of:

and S100, determining a target user, and inquiring the user requirement level of the target user prestored in the database.

The target user is a user who uses the corresponding application, for example, the use here may be to start the application or have entered into an application browsing page, the user requirement level is a division of the store service requirement by the user, for example, the user requirement level may include: the user requirement levels of different users may not be the same, and are only for illustration and have no limiting effect, and other forms of divided user requirement levels all belong to the protection scope of the present invention.

Specifically, which users use the application is monitored in real time, the users using the application are target users, and after the target users are determined, the database is queried to determine the user requirement levels of the target users. Optionally, the database stores the corresponding relationship between the user requirement level and the user identifier in advance, when it is monitored that the user uses the application, the user identifier of the user is obtained, the database is queried according to the user identifier, and the user requirement level corresponding to the user identifier is determined, so that the user requirement level of the target user is determined. The present embodiment is not limited to determining the user requirement level of the target user by querying the database according to the user identifier, and other information that can be used as the unique identity of the user may be used to query the database, where the database stores in advance the corresponding relationship between the user requirement level and other information that is used as the unique identity of the user.

Step S101, determining target content of at least one store to be recommended, and inquiring the store level of the at least one store prestored in the database.

The store level is divided according to the service quality of the store, and for example, the store level may include: the high-quality, medium-quality and low-quality store grades may be different from one store to another, and are only for illustration and have no limiting effect, and other types of store grades are within the protection scope of the present invention.

In real life, an application as an operation platform maintains a large number of stores, however, not all stores may have content recommendation needs, but some stores have content recommendation needs, and therefore, it is necessary to determine target content of at least one store to be recommended, where the target content may be an advertisement (e.g., a banner advertisement), a commodity, and the like, and after determining the target content of at least one store to be recommended, query a database to determine a store rank of the at least one store.

Optionally, the database stores the corresponding relationship between the store level and the store identifier in advance, after the target content of the at least one store to be recommended is determined, the store identifier of the at least one store is extracted, the database is queried according to the store identifier, the store level corresponding to the store identifier is determined, and thus the store level of the at least one store is determined. The embodiment is not limited to determining the store level of at least one store according to the store identification query database, and other information which can be used as the unique identity of the store can be used for querying the database, and at this time, the database stores the corresponding relationship between the store level and other information which is used as the unique identity of the store in advance.

And step S102, matching the user requirement level of the target user with the store level of at least one store, and selecting the target content to be recommended from the target content of the at least one store according to the matching result.

In order to avoid the problem that the content viewed by all users on a certain page in a certain time period is the same, which results in that some users do not have interest in the content and marketing traffic is wasted, the step determines target content to be recommended according to a user requirement level and an store level, specifically, matches the user requirement level of the target user with the store level of at least one store to obtain a matching result, where the matching is to determine whether there is a store level equivalent to the user requirement level of the target user, where the equivalent is specifically: the user request level is a medium request, and the store level corresponding to the user request level is medium quality. And then, selecting target content to be recommended from the target content of at least one store according to the matching result.

The selected target content to be recommended can more easily generate interest for the user, can obtain more attention of the user, and enables the user to be interested in knowing or purchasing and the like, so that the marketing flow of stores is further improved.

And step S103, pushing the target content to be recommended to the target user.

After the target content to be recommended is selected, the selected target content to be recommended can be pushed to the user, so that the user can see the pushed target content to be recommended.

In an optional implementation manner of the present invention, matching the user requirement level of the target user with the store level of at least one store, and selecting the target content to be recommended from the target content of the at least one store according to the matching result further includes:

judging whether the user requirement grade of the target user is matched with the store grade of any store; if so, selecting the target content of the matched store as the target content to be recommended; and if not, selecting the target content of the store with the store grade higher than the grade required by the user as the target content to be recommended.

After the user requirement level of the target user and the store level of at least one store are determined, comparing the user requirement level of the target user with the store level of at least one store to judge whether the user requirement level of the target user is matched with the store level of any store, if so, selecting the target content of the matched store as the target content to be recommended, in the optional implementation mode, the situation that the user requirement level of the target user is matched with the store levels of a plurality of stores may exist, and at the moment, randomly selecting the target content of one store from the matched plurality of stores as the target content to be recommended; or, according to a preset priority, selecting a target content of one store from a plurality of matched stores as a target content to be recommended, which is not specifically limited herein; if not, selecting the target content of the store with the store level higher than the user requirement level as the target content to be recommended, for example, if the user requirement level is a low requirement, the user requirement level of the target user does not match the store level of any store, if the store level is a medium requirement store, selecting the target content of the store with the store level medium requirement as the target content to be recommended, and if only the store level is a high requirement store, selecting the target content of the store with the store level high requirement as the target content to be recommended.

For example, the user requirement levels of the target user a and the target user B are medium requirement and low requirement, the target contents of three stores to be recommended, for example, the target contents of store 1, store 2 and store 3, are determined, and the store levels of store 1, store 2 and store 3 are obtained by querying: the method comprises the steps of determining that the user requirement level of a target user A is matched with the store level of the store 2 through judgment, selecting the target content of the store 2 matched with the user requirement level as the target content to be recommended, and pushing the target content to be recommended to the target user A, so that the user A can see the content to be recommended of the store 2; by determining that the user requirement level of the target user B is matched with the store level of the store 3, the target content of the store 3 matched with the user requirement level is selected as the target content to be recommended, and the target content to be recommended is pushed to the target user B, so that the user B can see the content to be recommended of the store 3.

For another example, the user requirement level of the target user a is a medium requirement, the target contents of three stores to be recommended, for example, the target contents of store 1, store 2, and store 3, are determined, and the store levels of store 1, store 2, and store 3 are found by querying: the method comprises the steps of high quality, low quality and low quality, if the user requirement level of a target user A is determined to be not matched with the store level of any store, the target content of the store 1 with the store level higher than the user requirement level is selected as the target content to be recommended, the target content to be recommended is pushed to the target user A, and therefore the user A can see the content to be recommended of the store 1.

The content recommendation method in this embodiment may be applied to recommendation of content to be presented by an open screen page of an application, where the open screen page is a page displayed before presenting a home page of the application in the starting process of an application program, and may also be used to enter a specific page in the application after starting the application, and the recommendation of the content to be presented by the page is, for example, the home page and other detailed pages.

According to the method provided by the embodiment of the invention, the target user is determined, and the user requirement level of the target user prestored in the database is inquired; determining target content of at least one store to be recommended, and inquiring the store level of the at least one store prestored in the database; matching the user requirement grade of the target user with the store grade of at least one store, and selecting target content to be recommended from the target content of the at least one store according to a matching result; and pushing the target content to be recommended to the target user. Based on the scheme provided by the invention, the appropriate target content to be recommended is recommended to the user according to the user requirement level, the interest degree of the target content to be recommended of the user can be improved, and the possibility of further knowing or purchasing by the user is greatly improved, so that the marketing flow of stores is further improved, the waste proportion of the marketing flow is reduced, and the pushing effect is improved.

Fig. 2 is a flowchart illustrating a store level and user demand level calculation method according to an embodiment of the present invention. The method includes the steps of determining store levels and user requirement levels according to acquired evaluation information, storing the finally determined store levels and user requirement levels in a database, determining the user requirement level of a target user and the store levels of at least one store by querying the database when content is recommended, matching the user requirement level of the target user with the store levels of the at least one store, selecting target content to be recommended from the target content of the at least one store according to matching results, and pushing the target content to be recommended to the target user. The calculation process of the store level and the user requirement level is described in detail below, and as shown in fig. 2, the method includes the following steps:

s200, determining a cluster of stores to be evaluated, and calculating the 1 st round of store level of each store in the cluster of stores to be evaluated according to the evaluation information of the store.

The store cluster to be rated is a set of all stores participating in store rating, and in order to improve the accuracy of the determined store rating and ensure the accuracy of subsequent content recommendation, optionally, the store cluster to be rated may be determined by: counting the number of the evaluation information of the stores, and determining whether the number of the evaluation information of the stores is greater than or equal to a first preset threshold, wherein a person skilled in the art can flexibly set the first preset threshold according to actual needs, for example, set the first preset threshold to 50 or 100, which is not specifically limited herein; if the number of the evaluation information of the store is greater than or equal to a first preset threshold value, adding the store into a store cluster to be evaluated; and if the number of the evaluation information of the stores is less than a first preset threshold value, assigning a default store grade to the store. The method comprises the steps that a first preset threshold value is used for measuring the number of evaluation information of stores, the number of the evaluation information of the stores is small when the first preset threshold value is smaller than the first preset threshold value, and the grade of the stores cannot be accurately determined according to the evaluation information of the stores, so that a default store grade can be allocated to the stores, and the method is ended after the default store grade is allocated to the stores.

The evaluation information of the stores reflects the service conditions of the stores, so that after the cluster of the stores to be evaluated is determined, for each store in the cluster of the stores to be evaluated, the 1 st round of store level of the store is calculated according to the evaluation information of the store, specifically, the evaluation information of the stores can be analyzed, for example, semantic analysis and/or score analysis, and the evaluation level of each piece of evaluation information of the stores can be determined, wherein the evaluation levels include: the evaluation level of the store is determined according to the evaluation level ratio.

Evaluation grade: the ratio of good comment, medium comment and bad comment can be called as: a store goodness rating, a store high rating, a store poor rating, wherein,

the number of good evaluations in the evaluation information of stores/the total number of evaluation information of stores

The number of evaluations in the evaluation information of stores/the total number of evaluation information of stores

The number of bad evaluations in the evaluation information of stores/the total number of evaluation information of stores

In the present embodiment, the following can be specified: when the good rating of the store is > 90% and the bad rating of the store is < 1%, the 1 st round of store rating of the store is: high quality.

When the store goodness score is < 70%, or the store badness score is > 30%, the 1 st round of store ranking of the store is: the quality is low.

In other cases, the 1 st round of stores is ranked as: medium quality.

The occupation ratio according to which the skilled person can flexibly set the rating of each store according to the actual need, which is not specifically limited herein, and for example, the following may be specified:

when the store goodness score is > 85% and the store badness score is < 5%, the 1 st round of store ranking of the stores is: high quality.

When the store goodness score is < 65%, or the store badness score is > 35%, the 1 st round of store ranking of the store is: the quality is low.

In other cases, the 1 st round of stores is ranked as: medium quality.

S201, determining a user cluster to be evaluated, and calculating the 1 st round user requirement level of each user in the user cluster to be evaluated according to the evaluation information of the user.

In order to improve the accuracy of the determined user requirement level and ensure the accuracy of subsequent content recommendation, optionally, the user cluster to be rated may be determined by the following method: counting the number of the evaluation information of the user, and judging whether the number of the evaluation information of the user is greater than or equal to a second preset threshold value; a person skilled in the art may flexibly set the second preset threshold according to actual needs, for example, the second preset threshold is set to 50 or 100, which is not specifically limited herein; if the number of the evaluation information of the user is larger than or equal to a second preset threshold value, adding the user into a user cluster to be evaluated; if the number of the evaluation information of the user is smaller than a second preset threshold value, distributing a default user requirement grade for the user; if the evaluation information of any user in the to-be-evaluated user cluster does not belong to the evaluation information for each store in the to-be-evaluated store cluster, the method in the embodiment cannot be used for determining the user requirement level, and therefore, a default user requirement level can be allocated to the user.

The evaluation information of the user reflects the service requirement condition of the user on the store, so that after the user cluster to be evaluated is determined, for each user in the user cluster to be evaluated, the 1 st round user requirement level of the user is calculated according to the evaluation information of the user, specifically, the evaluation information of the user may be analyzed, for example, semantic analysis and/or score analysis, and the evaluation level of each piece of evaluation information of the user is determined, where the evaluation level includes: and (4) carrying out good evaluation, medium evaluation and poor evaluation, calculating the ratio of various evaluation levels (the number of pieces of evaluation information of various evaluation levels/the total number of pieces of evaluation information of the user), and determining the 1 st round user requirement level of the user according to the ratio of the evaluation levels.

Evaluation grade: the ratio of good comment, medium comment and bad comment can be called as: the user rating, the user medium rating, and the user poor rating, wherein,

the user rating is the number of good comments in the user's rating information/the total number of the user's rating information

User rating is the number of ratings in the user's rating information/total number of user's rating information

The user bad rating is the number of bad ratings in the user's rating information/the total number of user's rating information

In the present embodiment, the following can be specified:

when the user rating is greater than 90% and the user rating is less than 1%, the user requirement rating of the 1 st round of the user is: and the requirement is low.

When the user rating is < 70%, or the user rating is > 30%, the user requirement rating of the 1 st round of the user is: high requirements are imposed.

In other cases, the 1 st round of user requirement levels of the users are as follows: the method is as required in (1).

The occupation ratio according to which each user requires level division can be flexibly set by those skilled in the art according to actual needs, which is not specifically limited herein, and for example, the following may be specified:

when the user rating is greater than 85% and the user rating is less than 5%, the user requirement rating of the 1 st round of the user is: and the requirement is low.

When the user rating is < 65% or the user rating is > 35%, the user requirement rating of the 1 st round of the user is: high requirements are imposed.

After the 1 st round of store level and the 1 st round of user requirement level are determined, taking the 1 st round of store level and the 1 st round of user requirement level as initial values, performing iterative adjustment on the initial values according to evaluation information of each store in the to-be-rated store cluster of each user in the to-be-rated user cluster, and finally determining the store level of each store in the to-be-rated store cluster and the user requirement level of each user in the to-be-rated user cluster.

Specifically, the store level of each store in the store cluster to be evaluated and the user requirement level of each user in the user cluster to be evaluated can be determined by the following methods in steps S202 to S205:

starting from the point where i is 2,

s202, aiming at each store in the store cluster to be evaluated, calculating the i-th store grade of the store according to the i-1 st user requirement grade and the evaluation information of the store, and aiming at each user in the user cluster to be evaluated, calculating the i-th user requirement grade of the user according to the i-1 st store grade and the evaluation information of the user.

The 1 st round of store grades determined according to the evaluation information of the stores can reflect the service conditions of the stores to a certain extent, but the 1 st round of store grades can not accurately reflect the quality of the stores, and in order to determine the store grades more accurately, the embodiment calculates the ith round of store grades of the stores by combining the i-1 st round of user requirement grades and the evaluation information of the stores, wherein the user requirement grades reflect different requirements of the users for the store services, and the evaluation information of the stores reflects the service conditions of the stores, so the ith round of store grades calculated based on the i-1 st round of user requirement grades can reflect the service conditions of the stores more objectively.

The 1 st round of user requirement level of each user determined according to the evaluation information of the user can reflect the service requirement of the user to the store to a certain extent, but the 1 st round of user requirement level may not accurately reflect the service requirement of the user to the store, in order to determine the user requirement level more accurately, the embodiment calculates the ith round of user requirement level of the user by combining the i-1 th round of store level and the evaluation information of the user, wherein the store level reflects the store service condition, and the evaluation information of the user reflects the service requirement of the user to the store, therefore, the ith round of user requirement level calculated based on the i-1 th round of store level can more objectively reflect the service requirement of the user to the store.

Fig. 3 is a schematic flow chart illustrating a method for calculating the ith round of store ranking in the embodiment of the present invention, that is: a detailed flow diagram of a part of the method in step S202 is shown in fig. 3, and the method includes the following steps:

in step S300, the evaluation level of each piece of evaluation information of the store is determined.

Specifically, the evaluation information of the store may be analyzed, for example, semantic analysis and/or score analysis, and an evaluation level of each piece of evaluation information of the store may be determined, where the evaluation level includes: good, medium and bad. For example, the evaluation information includes: when words such as good, excellent and extraordinary are used, the evaluation level of the evaluation information can be considered as good evaluation; the evaluation information includes: when words are general, can also be used, the evaluation level of the evaluation information can be regarded as a middle evaluation; the evaluation information includes: when words such as bad words and bad words are too bad words, the evaluation level of the evaluation information can be considered as bad evaluation; for another example, the evaluation information has a corresponding score of 5 stars, and the evaluation level of the evaluation information can be considered as good; the evaluation information is correspondingly scored to be 3 stars or 4 stars, and the evaluation level of the evaluation information can be considered as a middle score; the evaluation information is rated as 1 star or 2 stars, and the evaluation level of the evaluation information is considered as poor evaluation, which is only an example and has no limitation.

Step S301, classifying each piece of evaluation information according to the evaluation level of each piece of evaluation information and the i-1 th round user requirement level of the user corresponding to each piece of evaluation information and a first classification rule.

The first classification rule specifically defines how to classify the evaluation information of the store according to the user requirement level, and the rule definition of the first classification rule is simply listed as follows:

the high-requirement user is good comment and is 1 type evaluation information; the high-requirement user is given a medium comment which is 2 types of evaluation information; the poor rating of the high-demand user is 3 types of rating information.

The high comment of the user is required to be 2 types of evaluation information; the user is requested to have a comment, namely 3 types of evaluation information; the user's bad comment is required to be 4 types of evaluation information.

The favorable comment of the user with low requirement is 3 types of evaluation information; the low-requirement user is given a medium rating of 4 types of rating information; the poor rating of the low-requirement user is 5 types of rating information.

After the evaluation level (e.g., good evaluation, medium evaluation, or bad evaluation) of each piece of evaluation information is determined, each piece of evaluation information is classified according to the above classification rule according to the evaluation level of each piece of evaluation information and the i-1 st round of user requirement level of the user corresponding to each piece of evaluation information, for example, when the 2 nd round of store level is calculated, the evaluation level of a certain piece of evaluation information is good evaluation, the 1 st round of store level of the user corresponding to the piece of evaluation information is medium requirement, and the category corresponding to the piece of evaluation information is 2 categories according to the above classification rule, which is only for example and does not have any limiting effect.

Step S302, based on the classification result of each item of evaluation information of the store, the ith round of store level of the store is calculated by using a first preset rule.

The first preset rule specifically defines conditions corresponding to each store level, each item of evaluation information is classified according to step S301, a classification result of the item of evaluation information is obtained, and the ith store level of the store is calculated by using the first preset rule based on the classification result of each item of evaluation information of the store, specifically, the ith store level of the store can be calculated by using the following method: and calculating the number ratio of the evaluation information in at least one preset category obtained after classification according to the first classification rule, and calculating the ith round of store level of the stores according to the number ratio.

In the present embodiment, the following can be specified:

when the number of 1-2 type evaluation information accounts for > 90% and the number of 4-5 type evaluation information accounts for < 1%, the store level is high quality;

when the number of 1 or 2 types of evaluation information accounts for < 70%, or the number of 4 or 5 types of evaluation information accounts for > 30%, the store level is of low quality;

in other cases, the store level is medium quality.

Of course, the skilled person can flexibly set the corresponding values according to the actual needs, for example, the following can also be specified:

when the number of 1-2 type evaluation information is more than 85% and the number of 4-5 type evaluation information is less than 5%, the store level is high quality;

when the number of 1 or 2 types of evaluation information accounts for < 65%, or the number of 4 or 5 types of evaluation information accounts for > 35%, the store level is of low quality;

in other cases, the store level is medium quality.

Fig. 4 is a flowchart illustrating a method for calculating the ith round of user requirement level in the embodiment of the present invention, that is: a detailed flow diagram of a part of the method in step S202 is shown in fig. 4, and the method includes the following steps:

in step S400, the rating level of each piece of rating information of the user is determined.

Specifically, the evaluation information of the user may be analyzed, for example, semantic analysis and/or score analysis, and an evaluation level of each piece of evaluation information of the user may be determined, where the evaluation level includes: good, medium and bad. For example, the evaluation information includes: when words such as good, excellent and extraordinary are used, the evaluation level of the evaluation information can be considered as good evaluation; the evaluation information includes: when words are general, can also be used, the evaluation level of the evaluation information can be regarded as a middle evaluation; the evaluation information includes: when words such as bad words and bad words are too bad words, the evaluation level of the evaluation information can be considered as bad evaluation; for another example, the evaluation information has a corresponding score of 5 stars, and the evaluation level of the evaluation information can be considered as good; the evaluation information is correspondingly scored to be 3 stars or 4 stars, and the evaluation level of the evaluation information can be considered as a middle score; the evaluation information is rated as 1 star or 2 stars, and the evaluation level of the evaluation information is considered as poor evaluation, which is only an example and has no limitation.

Step S401, classifying each piece of evaluation information according to a second classification rule according to the evaluation level of each piece of evaluation information and the (i-1) th round of store level of the store corresponding to each piece of evaluation information.

The second classification rule specifically defines how to classify the user's evaluation information according to store level, and the rule definition of the second classification rule is simply listed below:

poor evaluation of high-quality stores, which is 1-type evaluation information; the high-quality store is subjected to the evaluation, namely 2 types of evaluation information; the favorable comment of high-quality stores is 3 types of evaluation information.

Poor evaluation of the medium-quality store is 2 types of evaluation information; the evaluation of the medium-quality store is 3 types of evaluation information; favorable comment of medium-quality stores, 4 types of evaluation information.

Poor evaluation of low-quality stores, which is 3 types of evaluation information; the evaluation of low-quality stores is 4 types of evaluation information; the favorable comment of low-quality stores is 5 types of evaluation information.

After the evaluation level (e.g., good rating, medium rating, or bad rating) of each piece of evaluation information is determined, each piece of evaluation information is classified according to the classification rule according to the evaluation level of each piece of evaluation information and the i-1 st round store level of the store corresponding to each piece of evaluation information, for example, when the 2 nd round user request level is calculated, the evaluation level of a certain piece of evaluation information is good rating, the 1 st round store level of the store corresponding to the piece of evaluation information is medium quality, and the category corresponding to the piece of evaluation information is 4 categories according to the classification rule, which is only for example and does not have any limiting effect.

Step S402, based on the classification result of each piece of evaluation information of the user, calculating the ith round of user requirement level of the user by using a second preset rule.

The second preset rule specifically defines conditions corresponding to each user requirement level, after each piece of evaluation information is classified according to step S401, a classification result of the piece of evaluation information is obtained, and based on the classification result of each piece of evaluation information of the user, the ith round of user requirement level of the user is calculated by using the second preset rule, specifically, the ith round of user requirement level of the store can be calculated by using the following method: and calculating the number ratio of the evaluation information in at least one preset category obtained after classification according to the second classification rule, and calculating the ith round of user requirement grade of the user according to the number ratio.

In the present embodiment, the following can be specified:

when the number of 4-5 types of evaluation information accounts for > 90% and the number of 1-2 types of evaluation information accounts for < 1%, the user requirement level is a low requirement;

when the number of 4-5 types of evaluation information accounts for < 70%, or the number of 1-2 types of evaluation information accounts for > 30%, the user requirement level is a high requirement;

in other cases, the user request level is medium.

when the number of 4-5 types of evaluation information accounts for > 85% and the number of 1-2 types of evaluation information accounts for < 5%, the user requirement level is a low requirement;

when the number of 4-5 types of evaluation information accounts for < 65%, or the number of 1-2 types of evaluation information accounts for > 35%, the user requirement level is a high requirement;

in other cases, the user request level is medium.

S203, judging whether the ith round of store grade of each store is the same as the ith-1 round of store grade and whether the ith round of user requirement grade of each user is the same as the ith-1 round of user requirement grade; if yes, go to step S204; if not, go to step S205.

In order to determine whether to end the iterative adjustment process, after the ith round of store level of each store and the ith round of user requirement level of each user are obtained through calculation, whether the ith round of store level of each store is the same as the ith-1 round of store level and whether the ith round of user requirement level of each user is the same as the ith-1 round of user requirement level are required to be judged, and when the ith round of store level of each store is judged to be the same as the ith-1 round of store level and the ith round of user requirement level of each user is the same as the ith-1 round of user requirement level, the store level of each store is indicated to be constant and the user requirement level of each user is also constant, so that the iterative adjustment process can be ended; and under the condition that the ith round of store grade of a certain store is judged to be different from the (i-1) th round of store grade or the ith round of user requirement grade of a certain user is judged to be different from the (i-1) th round of user requirement grade, continuing the iterative adjustment process.

And S204, finally determining the ith store grade of each store as the store grade of each store, finally determining the ith user requirement grade of each user as the user requirement grade of each user, and ending the iterative adjustment process.

When it is determined that the ith round of store level of each store is the same as the (i-1) th round of store level and the ith round of user requirement level of each user is the same as the (i-1) th round of user requirement level, it indicates that the store level of each store is constant and the user requirement level of each user is also constant, at this time, the ith round of store level of each store may be finally determined as the store level of each store, the ith round of user requirement level of each user may be finally determined as the user requirement level of each user, and the iterative adjustment process is ended.

S205, assigning i to i +1, and jumping to execute the step S202.

And assigning i to be i +1 under the condition that the ith round of store grade of a certain store is judged to be different from the (i-1) th round of store grade or the ith round of user requirement grade of a certain user is judged to be different from the (i-1) th round of user requirement grade, and skipping to execute the step S202.

In an optional implementation manner of the present invention, in order to save resources and avoid the calculation time of the store levels and the user requirement levels being too long, the iterative adjustment execution round number may be controlled, when i is greater than a third preset threshold, the iterative adjustment process is ended, the ith store level of each store is finally determined as the store level of each store, and the ith user requirement level of each user is finally determined as the user requirement level of each user.

Step S206, store the store level and the user request level in the database.

After the store grade of each store in the store cluster to be evaluated and the user requirement grade of each user in the user cluster to be evaluated are finally determined according to the steps of the method, the store grade and the user requirement grade are stored in the database. For example, store identifications may be stored in association with store ratings and user identifications may be stored in association with user requirement ratings. The store level and the user request level stored in the database can be used for content recommendation.

The detailed process of calculating the store level and the user request level is described below with reference to specific examples:

assuming that there are 5 stores to be rated, which are respectively S1-S5, and there are 5 users to be rated, which are respectively C1-C5, the evaluation information is shown in table 1:

table 1:

	C1	C2	C3	C4	C5
						S1	good 10, medium 15, and difference 5 equal to 30	20 or 5 or 25	Preferably 10 to 10	Medium 20 to 20	Good 15 ═ 15
S2	Good 20, medium 5, poor 5 equal to 30	Good 15, poor 5-20	Good 5, poor 5-10	Good 5, poor 5-10	Good 15, medium 5-20
						S3	Good 10, medium 5-15	Good 15 ═ 15	Preferably 40 to 40	Preferably 10 to 10	Preferably 20 to 20
S4	Good 5, medium 5, difference 10 to 20	Preferably 20 to 20	Good 10, poor 20-30	Good 10, poor 5-15	Good 15 ═ 15
						S5	Good 5 is 5	10 in, 20 in difference	10 in	10 in, 5 in 15	Preferably 30 to 30

For the stores S1 to S5, the 1 st round store level of each store is determined, specifically as follows:

s1: the good rating of store is 55% (10+20+10+15)/100, and the bad rating of store is 5% (low quality) 5/100

S2: the store good rating is (20+15+5+5+15)/90 is 66%, the store poor rating is (5+5+ 5)/90 is 22% (low quality)

S3: the good rating of store is (10+15+40+10+20)/100 is 95%, and the bad rating of store is 0% (high quality)

S4: the good rating of store is 60% (5+20+10+10+15)/100, and the bad rating of store is 35% (low quality)

S5: the good rating of the store is (5+30)/80 is 43%, and the bad rating of the store is (10+5)/80 is 18% (low quality)

For users C1-C5, the 1 st round user requirement level of each user is respectively determined, which is specifically as follows:

c1: the user rating of good is 50% (10+20+10 +5)/100, and the user rating of bad is 20% (high requirement)

C2: the user rating of good is 70% (20+15+15+20)/100, and the user rating of bad is 15% (medium requirement)

C3: the user rating of good is 65% (10+5+40+10)/100, and the user rating of bad is 25% (high requirement)

C4: the user rating of good is 35% (5+10+10)/70, and the user rating of bad is 21% (high requirement)

C5: the user rating of good is 95% (15+15+20+15+30)/100, and the user rating of bad is 0% (low requirement)

Determining the classification result of each piece of evaluation information according to the evaluation level of each piece of evaluation information and the user requirement level of the 1 st round corresponding to each piece of evaluation information and according to a first classification rule, wherein the classification result is shown in table 2:

table 2:

	c1 (high)	C2 (middle)	C3 (high)	C4 (high)	C5 (Low)
						S1 Low	Class 1, class 10, class 2, class 15, class 3, class 5 ═ 30	Class 2 20, class 3, 5-25	Class 1 10 ═ 10	Class 2 20-20	Class 3 15-15
S2 Low	Class 1 20, class 25, class 35 ═ 30	Class 2 15, class 4, class 5 ═ 20	Class 1, 5, 3, 5 ═ 10	Class 1, 5, 3, 5 ═ 10	Class 3 15, class 4, class 5 ═ 20
						S3 high	Class 1, class 10, class 2, class 5 ═ 15	Class 2 15-15	Class 1 40-40	Class 1 10 ═ 10	Class 3 20-20
S4 Low	Class 15, class 25, class 3, class 10 ═ 20	Class 2 20-20	Class 1, class 10, class 3, class 20-30	Class 1, class 10, class 3, class 5 ═ 15	Class 3 15-15
						S5 Low	Class 15 ═ 5	Class 3, class 10, class 4, class 10-20	Class 2, 10 ═ 10	Class 2, class 10, class 3, class 5 ═ 15	Class 3 30-30

Calculating the level of the 2 nd round store according to the classification result, which is concretely as follows:

s1: 1. the ratio of the number of pieces of evaluation information of 2 types to (25+20+10+20)/100 to 75%, and the ratio of the number of pieces of evaluation information of 4 and 5 types to 0% (high-quality shop)

S2: 1. the ratio of the number of pieces of the 2-type evaluation information (25+15+ 5)/90-55%, and the ratio of the number of pieces of the 4-type and 5-type evaluation information (5+ 5)/90-11% (low-quality store)

S3: 1. the ratio of the number of 2-type evaluation information items to (15+15+40+10)/100 to 80%, and the ratio of the number of 4-type and 5-type evaluation information items to 0% (high-quality shop)

S4: 1. the ratio of the number of 2-type evaluation information pieces to (10+20+10+10)/100 to 50%, and the ratio of the number of 4-type and 5-type evaluation information pieces to 0% (low-quality store)

S5: 1. the ratio of the number of pieces of the 2-type evaluation information (5+10+ 10)/80-31%, and the ratio of the number of pieces of the 4-type and 5-type evaluation information (10/80-12.5% (low-quality store)

According to the evaluation level of each piece of evaluation information and the 1 st round of store level of the store corresponding to each piece of evaluation information, determining the classification result of each piece of evaluation information according to a second classification rule, wherein the classification result is shown in table 3:

table 3:

	c1 (high)	C2 (middle)	C3 (high)	C4 (high)	C5 (Low)
						S1 Low	Class 5, class 10, class 4, class 15, class 3, class 5 ═ 30	In the 5-class 20 state of the art,class 4 5-25	Class 5 with 10 ═ 10	Class 4 20-20	Class 5 15-15
S2 Low	Class 520, class 4 class 5, class 3 class 5 ═ 30	Class 5 15, class 3, class 5 ═ 20	5 class 5, 3 class 5 ═ 10	5 class 5, 3 class 5 ═ 10	Class 5 15, class 4, class 5 ═ 20
						S3 high	Class 3, class 10, class 2, class 5 ═ 15	Class 3 15-15	Class 3 40-40	Class 3, 10 ═ 10	Class 3 20-20
S4 Low	Class 5, class 4, class 5, class 3, class 10 ═ 20	Class 5 20-20	Class 5, class 10, class 3, class 20-30	Class 5, class 10, class 3, class 5 ═ 15	Class 5 15-15
						S5 Low	Class 5-5 ═ 5	Class 4, class 3, class 10-20	Class 4, 10 ═ 10	Class 4, class 10, class 3, class 5 ═ 15	Class 5 30-30

Calculating the requirement grade of the 2 nd round of users according to the classification result, which is concretely as follows:

c1: 4. the ratio of the number of 5-type evaluation information pieces to (25+25+10+5)/100 to 65%, and the ratio of the number of 1-type and 2-type evaluation information pieces to 5/100 to 5% (high requirement)

C2: 4. the ratio of the number of 5-type evaluation information pieces to (25+15+20+10)/100 to 70%, and the ratio of the number of 1-type and 2-type evaluation information pieces to 0% (requirements in the specification)

C3: 4. the ratio of the number of 5-type evaluation information pieces to (10+5+10+10)/100 to 35%, and the ratio of the number of 1-type and 2-type evaluation information pieces to 0% (high requirement)

C4: 4. the ratio of the number of 5-type evaluation information pieces to (20+5+10+ 10)/70-64%, and the ratio of the number of 1-type and 2-type evaluation information pieces to 0% (high requirement)

C5: 4. the ratio of the number of 5-type evaluation information pieces to (15+20+15+30)/100 to 80%, and the ratio of the number of 1-type and 2-type evaluation information pieces to 0% (requirements in the specification)

Through comparison, if the 2 nd round store grades of the stores S1 and S3 are different from the 1 st round store grade, and the 2 nd round user requirement grade of the user C5 is different from the 1 st round user requirement grade, the 3 rd round store grade calculation and the 3 rd round user requirement grade calculation need to be carried out, after several rounds of iterative adjustment, the store grade of each store in the to-be-rated store cluster and the user requirement grade of each user in the to-be-rated user cluster are finally determined, and the store grade and the user requirement grade are stored in a database and used in content recommendation.

According to the method provided by the embodiment of the invention, the store grade of each store in the store cluster to be evaluated and the user requirement grade of each user in the user cluster to be evaluated are more accurate through multiple adjustments of the mutual restriction of the store grade and the user requirement grade, and the requirements of the users and the service conditions of the stores can be objectively and accurately reflected, so that when content recommendation is carried out according to the determined store grade and the user requirement grade, the recommended target content to be recommended is more easy to generate interest for the users, more attention can be given to the users, the users can be interested in knowing or purchasing, and the like, and the marketing flow of the stores is further improved.

Fig. 5 is a schematic structural diagram of a content recommendation apparatus according to an embodiment of the present invention. As shown in fig. 5, the apparatus includes: a first determining module 500, a first querying module 510, a second determining module 520, a second querying module 530, a processing module 540, and a pushing module 550.

A first determination module 500 adapted to determine a target user.

The first query module 510 is adapted to query the pre-stored user requirement levels of the target users in the database.

A second determining module 520 adapted to determine the target content of the at least one store to be recommended.

A second query module 530 adapted to query the database for a pre-stored store rank of at least one store.

And the processing module 540 is adapted to match the user requirement level of the target user with the store level of at least one store, and select the target content to be recommended from the target content of at least one store according to the matching result.

And the pushing module 550 is adapted to push the target content to be recommended to the target user.

Optionally, the processing module 540 is further adapted to: judging whether the user requirement grade of the target user is matched with the store grade of any store; if so, selecting the target content of the matched store as the target content to be recommended; and if not, selecting the target content of the store with the store grade higher than the grade required by the user as the target content to be recommended.

Optionally, the apparatus further comprises: a third determining module 560, adapted to obtain the evaluation information, and determine a store level and a user requirement level according to the evaluation information;

a database 570 adapted to store ratings and user demand ratings.

Optionally, the third determining module 560 is further adapted to: determining a cluster of stores to be evaluated, and calculating the grade of the 1 st round of stores of the stores according to the evaluation information of the stores aiming at each store in the cluster of stores to be evaluated;

Optionally, the third determining module 560 is further adapted to perform iterative adjustment according to the following procedure: starting from the point where i is 2,

s4, if not, assigning i as i +1, and jumping to execute S1.

Optionally, the third determining module 560 is further adapted to: determining an evaluation level of each piece of evaluation information of the store; classifying each piece of evaluation information according to a first classification rule according to the evaluation level of each piece of evaluation information and the i-1 th round user requirement level of the user corresponding to each piece of evaluation information; and calculating the ith round of store level of the store by using a first preset rule based on the classification result of each piece of evaluation information of the store.

Optionally, the third determining module 560 is further adapted to: and calculating the number ratio of the evaluation information in at least one preset category obtained after classification according to the first classification rule, and calculating the ith round of store level of the stores according to the number ratio.

Optionally, the third determining module 560 is further adapted to: determining the evaluation level of each piece of evaluation information of the user; classifying each piece of evaluation information according to a second classification rule according to the evaluation level of each piece of evaluation information and the grade of the (i-1) th round of store corresponding to each piece of evaluation information; and calculating the requirement grade of the ith round of the user by utilizing a second preset rule based on the classification result of each piece of evaluation information of the user.

Optionally, the third determining module 560 is further adapted to: and calculating the number ratio of the evaluation information in at least one preset category obtained after classification according to the second classification rule, and calculating the ith round of user requirement grade of the user according to the number ratio.

Optionally, the third determining module 560 is further adapted to: judging whether the number of the evaluation information of the store is greater than or equal to a first preset threshold value or not; if yes, adding the store into a store cluster to be evaluated; and if not, allocating a default store grade for the store.

Optionally, the third determining module 560 is further adapted to: judging whether the number of the evaluation information of the user is greater than or equal to a second preset threshold value or not; if so, adding the user into a user cluster to be evaluated; and if the evaluation information of any user in the user cluster to be evaluated does not belong to the evaluation information aiming at each store in the store cluster to be evaluated, allocating a default user requirement level for the user.

According to the device provided by the embodiment of the invention, the target user is determined, and the user requirement level of the target user prestored in the database is inquired; determining target content of at least one store to be recommended, and inquiring the store level of the at least one store prestored in the database; matching the user requirement grade of the target user with the store grade of at least one store, and selecting target content to be recommended from the target content of the at least one store according to a matching result; and pushing the target content to be recommended to the target user. Based on the scheme provided by the invention, the appropriate target content to be recommended is recommended to the user according to the user requirement level, the interest degree of the target content to be recommended of the user can be improved, and the possibility of further knowing or purchasing by the user is greatly improved, so that the marketing flow of stores is further improved, the waste proportion of the marketing flow is reduced, and the pushing effect is improved.

An embodiment of the present application further provides a non-volatile computer storage medium, where at least one executable instruction is stored in the computer storage medium, and the computer executable instruction may execute the method in any of the above method embodiments.

Fig. 6 is a schematic structural diagram of a computing device according to an embodiment of the present invention, and the specific embodiment of the present invention does not limit the specific implementation of the computing device.

As shown in fig. 6, the computing device may include: a processor (processor)602, a communication Interface 604, a memory 606, and a communication bus 608.

Wherein:

the processor 602, communication interface 604, and memory 606 communicate with one another via a communication bus 608.

A communication interface 604 for communicating with network elements of other devices, such as clients or other servers.

The processor 602 is configured to execute the program 610, and may specifically perform relevant steps in the foregoing method embodiments.

In particular, program 610 may include program code comprising computer operating instructions.

The processor 602 may be a central processing unit CPU or an application Specific Integrated circuit asic or one or more Integrated circuits configured to implement embodiments of the present invention. The computing device includes one or more processors, which may be the same type of processor, such as one or more CPUs; or may be different types of processors such as one or more CPUs and one or more ASICs.

And a memory 606 for storing a program 610. Memory 606 may comprise high-speed RAM memory, and may also include non-volatile memory (non-volatile memory), such as at least one disk memory.

The program 610 may specifically be adapted to cause the processor 602 to perform a method in any of the method embodiments described above. For specific implementation of each step in the program 610, reference may be made to corresponding steps and corresponding descriptions in units in the foregoing embodiments, which are not described herein again. It can be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described devices and modules may refer to the corresponding process descriptions in the foregoing method embodiments, and are not described herein again.

The algorithms and displays presented herein are not inherently related to any particular computer, virtual machine, or other apparatus. Various general purpose systems may also be used with the teachings herein. The required structure for constructing such a system will be apparent from the description above. Moreover, the present invention is not directed to any particular programming language. It is appreciated that a variety of programming languages may be used to implement the teachings of the present invention as described herein, and any descriptions of specific languages are provided above to disclose the best mode of the invention.

In the description provided herein, numerous specific details are set forth. It is understood, however, that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure an understanding of this description.

Similarly, it should be appreciated that in the foregoing description of exemplary embodiments of the invention, various features of the invention are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure and aiding in the understanding of one or more of the various inventive aspects. However, the disclosed method should not be interpreted as reflecting an intention that: that the invention as claimed requires more features than are expressly recited in each claim. Rather, as the following claims reflect, inventive aspects lie in less than all features of a single foregoing disclosed embodiment. Thus, the claims following the detailed description are hereby expressly incorporated into this detailed description, with each claim standing on its own as a separate embodiment of this invention.

Those skilled in the art will appreciate that the modules in the device in an embodiment may be adaptively changed and disposed in one or more devices different from the embodiment. The modules or units or components of the embodiments may be combined into one module or unit or component, and furthermore they may be divided into a plurality of sub-modules or sub-units or sub-components. All of the features disclosed in this specification (including any accompanying claims, abstract and drawings), and all of the processes or elements of any method or apparatus so disclosed, may be combined in any combination, except combinations where at least some of such features and/or processes or elements are mutually exclusive. Each feature disclosed in this specification (including any accompanying claims, abstract and drawings) may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise.

Furthermore, those skilled in the art will appreciate that while some embodiments described herein include some features included in other embodiments, rather than other features, combinations of features of different embodiments are meant to be within the scope of the invention and form different embodiments. For example, in the following claims, any of the claimed embodiments may be used in any combination.

The various component embodiments of the invention may be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. It will be appreciated by those skilled in the art that a microprocessor or Digital Signal Processor (DSP) may be used in practice to implement some or all of the functionality of some or all of the components of a content recommendation device according to embodiments of the present invention. The present invention may also be embodied as apparatus or device programs (e.g., computer programs and computer program products) for performing a portion or all of the methods described herein. Such programs implementing the present invention may be stored on computer-readable media or may be in the form of one or more signals. Such a signal may be downloaded from an internet website or provided on a carrier signal or in any other form.

It should be noted that the above-mentioned embodiments illustrate rather than limit the invention, and that those skilled in the art will be able to design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention may be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by one and the same item of hardware. The usage of the words first, second and third, etcetera do not indicate any ordering. These words may be interpreted as names.

Claims

1. A content recommendation method, comprising:

determining target content of at least one store to be recommended, and inquiring the store level of the at least one store prestored in a database;

matching the user requirement grade of the target user with the store grade of the at least one store, and selecting target content to be recommended from the target content of the at least one store according to a matching result;

pushing the target content to be recommended to the target user;

the method further comprises the following steps: obtaining evaluation information, aiming at each store in the store cluster to be evaluated, calculating the i-th round of store grade of the store according to the i-1 th round of user requirement grade and the evaluation information of the store, aiming at each user in the user cluster to be evaluated, calculating the i-th round of user requirement grade of the user according to the i-1 th round of store grade and the evaluation information of the user, performing iterative adjustment for several rounds, finally determining the store grade of each store in the store cluster to be evaluated and the user requirement grade of each user in the user cluster to be evaluated, and storing the store grade and the user requirement grade in a database.

2. The method of claim 1, wherein the matching the user requirement rating of the target user with the store rating of the at least one store, and the selecting the target content to be recommended from the target content of the at least one store according to the matching result further comprises:

judging whether the user requirement level of the target user is matched with the store level of any store;

if so, selecting the target content of the matched store as the target content to be recommended;

and if not, selecting the target content of the store with the store grade higher than the grade required by the user as the target content to be recommended.

3. The method according to claim 1 or 2, wherein the method further comprises:

and determining a user cluster to be evaluated, and calculating the 1 st round user requirement level of each user in the user cluster to be evaluated according to the evaluation information of the user.

4. The method of claim 3, wherein the calculating, for each store in the cluster of stores to be rated, an ith store level of the store according to the (i-1) th user requirement level and the evaluation information of the store, and for each user in the cluster of users to be rated, calculating an ith user requirement level of the user according to the (i-1) th store level and the evaluation information of the user, and performing iterative adjustment for several rounds to finally determine the store level of each store in the cluster of stores to be rated and the user requirement level of each user in the cluster of users to be rated further comprises: starting from the point where i is 2,

s4, if not, assigning i as i +1, and jumping to execute step S1.

5. The method of claim 4, wherein the calculating, for each store in the cluster of stores to be rated, an ith store level for the store from the (i-1) th round user requirement level and the evaluation information for the store further comprises:

determining an evaluation level of each piece of evaluation information of the store;

classifying each piece of evaluation information according to a first classification rule according to the evaluation level of each piece of evaluation information and the i-1 th round user requirement level of the user corresponding to each piece of evaluation information;

and calculating the ith round of store level of the store by using a first preset rule based on the classification result of each piece of evaluation information of the store.

6. The method of claim 5, wherein the calculating the ith round of store ranking of the store by using the first preset rule based on the classification result of each piece of evaluation information of the store further comprises:

and calculating the number ratio of the evaluation information in at least one preset category obtained after classification according to the first classification rule, and calculating the ith round of store level of the stores according to the number ratio.

7. The method of claim 4, wherein the calculating, for each user in the cluster of users to be rated, an ith round of user requirement rating for the user based on the ith-1 round of store ratings and the user's ratings information further comprises:

determining the evaluation level of each piece of evaluation information of the user;

classifying each piece of evaluation information according to a second classification rule according to the evaluation level of each piece of evaluation information and the grade of the (i-1) th round of store corresponding to each piece of evaluation information;

and calculating the requirement grade of the ith round of the user by utilizing a second preset rule based on the classification result of each piece of evaluation information of the user.

8. The method of claim 7, wherein the calculating an ith round of user requirement rating of the user by using a second preset rule based on the classification result of each piece of evaluation information of the user further comprises:

and calculating the number ratio of the evaluation information in at least one preset category obtained after classification according to a second classification rule, and calculating the ith round of user requirement grade of the user according to the number ratio.

9. The method of claim 3, wherein the determining a cluster of stores to be rated further comprises:

judging whether the number of the evaluation information of the store is greater than or equal to a first preset threshold value or not; if yes, adding the store into a store cluster to be evaluated;

and if not, allocating a default store grade for the store.

10. The method of claim 3, wherein the determining a cluster of users to be rated further comprises:

judging whether the number of the evaluation information of the user is greater than or equal to a second preset threshold value or not;

if so, adding the user into a user cluster to be evaluated;

and if the evaluation information of any user in the to-be-evaluated user cluster does not belong to the evaluation information aiming at each store in the to-be-evaluated store cluster, allocating a default user requirement level for the user.

11. A content recommendation apparatus comprising:

a first determination module adapted to determine a target user;

the second query module is suitable for querying the store grade of the at least one store prestored in the database;

the processing module is suitable for matching the user requirement grade of the target user with the store grade of the at least one store, and selecting target content to be recommended from the target content of the at least one store according to a matching result;

the pushing module is suitable for pushing the target content to be recommended to the target user;

the third determining module is suitable for acquiring evaluation information, calculating the ith round of store level of the store according to the (i-1) th round of user requirement level and the evaluation information of the store for each store in the to-be-evaluated store cluster, calculating the ith round of user requirement level of each user according to the (i-1) th round of store level and the evaluation information of the user for each user in the to-be-evaluated user cluster, and finally determining the store level of each store in the to-be-evaluated store cluster and the user requirement level of each user in the to-be-evaluated user cluster through several rounds of iterative adjustment;

12. The apparatus of claim 11, wherein the processing module is further adapted to: judging whether the user requirement level of the target user is matched with the store level of any store; if so, selecting the target content of the matched store as the target content to be recommended; and if not, selecting the target content of the store with the store grade higher than the grade required by the user as the target content to be recommended.

13. The apparatus of claim 11 or 12, wherein the third determining means is further adapted to:

14. The apparatus of claim 13, wherein the third determination module is further adapted to iteratively adjust according to the following procedure: starting from the point where i is 2,

s4, if not, assigning i as i +1, and jumping to execute S1.

15. The apparatus of claim 14, wherein the third determination module is further adapted to:

16. The apparatus of claim 15, wherein the third determination module is further adapted to:

17. The apparatus of claim 14, wherein the third determination module is further adapted to:

18. The apparatus of claim 17, wherein the third determining module is further adapted to:

19. The apparatus of claim 13, wherein the third determination module is further adapted to: judging whether the number of the evaluation information of the store is greater than or equal to a first preset threshold value or not; if yes, adding the store into a store cluster to be evaluated; and if not, allocating a default store grade for the store.

20. The apparatus of claim 13, wherein the third determination module is further adapted to: judging whether the number of the evaluation information of the user is greater than or equal to a second preset threshold value or not; if so, adding the user into a user cluster to be evaluated; and if the evaluation information of any user in the to-be-evaluated user cluster does not belong to the evaluation information aiming at each store in the to-be-evaluated store cluster, allocating a default user requirement level for the user.

21. A computing device, comprising: the system comprises a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface complete mutual communication through the communication bus;

the memory is used for storing at least one executable instruction, and the executable instruction causes the processor to execute the operation corresponding to the content recommendation method according to any one of claims 1-10.

22. A computer storage medium having at least one executable instruction stored therein, the executable instruction causing a processor to perform operations corresponding to the content recommendation method according to any one of claims 1-10.