CN113569148A

CN113569148A - Target information recommendation method and device, electronic equipment and storage medium

Info

Publication number: CN113569148A
Application number: CN202110864027.5A
Authority: CN
Inventors: 王若宇
Original assignee: Beijing Shareit Information Technology Co Ltd
Current assignee: Beijing Shareit Information Technology Co Ltd
Priority date: 2021-07-29
Filing date: 2021-07-29
Publication date: 2021-10-29
Also published as: WO2023005419A1

Abstract

The disclosure relates to a recommendation method of target information, which comprises the following steps: dividing request information of a plurality of different users into M parts, wherein the Nth part comprises a plurality of Nth request information sets, the (N + 1) th part comprises at least one (N + 1) th request information set, and the quantity of the request information in the Nth request information set is smaller than that in the (N + 1) th request information set; n is more than or equal to 1 and less than or equal to M-1, and both N and M are positive integers; determining Nth target information matched with the request information in the Nth request information set from the information to be recommended according to the Nth weight of at least one dimension information of the information to be recommended; acquiring feedback information after the Nth target information is recommended; updating the (N + 1) th weight of at least one dimension information of the information to be recommended according to the feedback information; and determining the (N + 1) th target information matched with the request information in the (N + 1) th request information set from the information to be recommended based on the updated (N + 1) th weight.

Description

Target information recommendation method and device, electronic equipment and storage medium

Technical Field

The present disclosure relates to the field of information processing, and in particular, to a method and an apparatus for recommending target information, an electronic device, and a storage medium.

Background

With the development of scientific technology, a plurality of emerging application technologies and application scenarios appear, and the application technologies can be applied to relevant application scenarios. The information corresponding to the application scene can be generated in different application scenes, and the information can be better applied to the relevant application scenes through processing the information, so that a better application effect is achieved.

For example, when information is recommended, after the relevant information is processed by the information processing technology, more matched information can be recommended to the user.

Disclosure of Invention

The disclosure provides a recommendation method and device of target information, electronic equipment and a storage medium.

In a first aspect of the embodiments of the present disclosure, a method for recommending target information is provided, including: dividing request information of a plurality of different users into M parts, wherein the Nth part comprises a plurality of Nth request information sets, the N +1 th part comprises at least one (N + 1) th request information set, and the quantity of the request information in the Nth request information set is smaller than that in the (N + 1) th request information set; wherein N is greater than or equal to 1 and less than or equal to M-1, and both N and M are positive integers; determining Nth target information matched with the request information in the Nth request information set from the information to be recommended according to the Nth weight of at least one dimension information of the information to be recommended; acquiring feedback information after the Nth target information is recommended; updating the (N + 1) th weight of at least one dimension information of the information to be recommended according to the feedback information; and determining the (N + 1) th target information matched with the request information in the (N + 1) th request information set from the information to be recommended based on the updated (N + 1) th weight.

In one embodiment, the mth portion comprises: one said mth request information set; the method further comprises the following steps: after the Mth weight is updated, updating a first weight corresponding to a plurality of first request information sets included in the first part according to the updated Mth weight; and according to the updated first weight, determining first target information matched with the request information in the first request information set from the information to be recommended again.

In one embodiment, after updating the mth weight, updating, according to the updated mth weight, first weights corresponding to a plurality of first request information sets included in the first part includes: determining the target weights with the same quantity as the first request information set included in the first part according to the updated Mth weight; updating the first weight according to the target weight.

In one embodiment, after updating the first weights corresponding to the plurality of first request information sets included in the first part, the method further includes: updating a second weight to the Mth weight according to the updated first weight; and updating the (N + 1) th weight according to the updated (N) th weight.

In one embodiment, a plurality of request messages in the nth set of request messages have the same nth weight; the obtaining of the feedback information after the nth target information is recommended includes: determining feedback dimension information of the Nth target information and parameters of each piece of feedback dimension information; and determining a feedback value according to the feedback dimension information and the parameters of each piece of feedback dimension information.

In an embodiment, the updating the (N + 1) th weight of the at least one dimension information of the information to be recommended according to the feedback information includes: determining the feedback values corresponding to the request information in the plurality of Nth request information sets respectively; determining an nth weight corresponding to the request information in the nth request information set with the maximum feedback value as the nth target weight; updating the N +1 th weight according to the Nth target weight.

In one embodiment, the feedback dimension information includes at least one of: the method comprises the following steps of downloading amount, collection amount, praise amount, the number of concerned users of an author of information to be recommended, average browsing duration, comment amount and forwarding amount.

In one embodiment, the dividing the request information of the plurality of different users into M parts includes: dividing the request information with the receiving time in the same time period into the same part according to the receiving time for receiving the request information; and/or obtaining a ranking value according to the priority of the user sending the request information and a ranking factor corresponding to the receiving time of the request information; and dividing the request information with the sorting values in the same interval into the same part.

In one embodiment, the information to be recommended includes at least one of: video; a public number; an article; advertising; the dimension information includes at least one of: the method comprises the following steps of downloading amount, collection amount, praise amount, the number of concerned users of an author of information to be recommended, average browsing duration, comment amount and forwarding amount.

In a second aspect of the embodiments of the present disclosure, there is provided a recommendation apparatus for target information, including: the device comprises a dividing module, a sending module and a receiving module, wherein the dividing module is used for dividing request information of a plurality of different users into M parts, the N part comprises a plurality of N request information sets, the N +1 part comprises at least one N +1 request information set, and the quantity of the request information in the N request information set is smaller than that in the N +1 request information set; wherein N is greater than or equal to 1 and less than or equal to M-1, and both N and M are positive integers; the Nth target information determining module is used for determining Nth target information matched with the request information in the Nth request information set from the information to be recommended according to the Nth weight of at least one dimension information of the information to be recommended; the feedback information acquisition module is used for acquiring feedback information after the Nth target information is recommended; the updating module is used for updating the (N + 1) th weight of at least one dimension information of the information to be recommended according to the feedback information; and the (N + 1) th target information determining module is used for determining (N + 1) th target information matched with the request information in the (N + 1) th request information set from the information to be recommended based on the updated (N + 1) th weight.

In a third aspect of the embodiments of the present disclosure, there is provided an electronic device, including:

a processor and a memory for storing executable instructions operable on the processor, wherein:

when the processor is used for executing the executable instructions, the executable instructions execute the method of any one of the above embodiments.

In a fourth aspect of the embodiments of the present disclosure, a non-transitory computer-readable storage medium is provided, in which computer-executable instructions are stored, and when executed by a processor, the computer-executable instructions implement the method according to any of the embodiments described above.

The technical scheme provided by the embodiment of the disclosure can have the following beneficial effects:

the embodiment of the disclosure divides the request information of a plurality of different users into M parts, the Nth part comprises a plurality of Nth request information sets, the (N + 1) th part comprises at least one (N + 1) th request information set, and the quantity of the request information included in the Nth request information set is smaller than that of the request information included in the (N + 1) th request information set. N is greater than or equal to 1 and less than or equal to M-1, and both N and M are positive integers. This enables the classification of the requested information into sets comprising different requested information.

When determining the target information of the request information included in each request information set, each request information set has a weight of at least one dimension information of the corresponding information to be recommended, for example, the nth weight is used to determine the nth target information of the request information in the nth request information set. And determining the Nth target information matched with the request information in the Nth request information set from the information to be recommended according to the Nth weight of the at least one dimension information of the information to be recommended. This determines the target information for the requested information in each nth set of requested information in the nth portion.

And then obtaining feedback information after the Nth target information is recommended, and updating the (N + 1) th weight of at least one dimension information of the information to be recommended according to the feedback information. And determining the (N + 1) th target information matched with the request information in the (N + 1) th request information set from the information to be recommended based on the updated (N + 1) th weight. Therefore, the (N + 1) th weight is updated according to the Nth weight, and the (N + 1) th target information is determined according to the Nth weight, so that the second weight is updated according to the first weight, the third weight is updated according to the second weight, and the like, and the next weight is automatically updated according to the previous weight.

Because the quantity of the request information included in the Nth request information set is less than that of the request information included in the (N + 1) th request information set, the weight of the (N + 1) th request information set with larger quantity of the request information is updated according to the feedback information corresponding to the Nth request information set with smaller quantity of the request information, and the determined (N + 1) th target information is more accurate. Namely, the recommendation of the large flow is updated according to the feedback effect of the recommendation of the small flow, and the dynamic update recommendation of the small data volume is realized. After the corresponding weight is updated, the corresponding target information can be determined according to the updated weight, so that the matching degree of the determined target information and the request information of the user is higher, the recommended target information is more accurate, and more accurate recommendation is realized.

It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and together with the description, serve to explain the principles of the disclosure.

FIG. 1 is a flow diagram illustrating a method for recommending target information in accordance with an exemplary embodiment;

FIG. 2 is a schematic diagram illustrating a process for obtaining feedback information according to an example embodiment;

FIG. 3 is a schematic flow diagram illustrating an update of an N +1 th weight according to an exemplary embodiment;

FIG. 4 illustrates another method of recommending target information, according to an example embodiment;

FIG. 5 is a block diagram illustrating an apparatus for recommending target information according to an exemplary embodiment;

FIG. 6 is a schematic diagram illustrating another recommendation method in accordance with an exemplary embodiment;

fig. 7 is a schematic diagram of a terminal device shown according to an example embodiment.

Detailed Description

Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the exemplary embodiments below are not intended to represent all implementations consistent with the present disclosure. Rather, they are merely examples of devices consistent with certain aspects of the present disclosure, as detailed in the appended claims.

In general, when information is recommended, because there are many reference factors, that is, there are many factors affecting the information recommendation result, when information is recommended, the final information is recommended by combining a plurality of reference factors. However, in general, the weights corresponding to the reference factors may be different, and the final information to be recommended is determined according to the weights occupied by the different reference factors when the final information is recommended.

As the reference factors increase and the weights occupied by different reference factors are different for different users, the weights corresponding to the reference factors may also be changed under the influence of factors such as user preference, and the adjustment of the weights corresponding to the reference factors is very important. In general, the weight is adjusted in an offline adjustment mode such as manual adjustment, and after the offline adjustment is completed, the weight is updated to the online state, so that manual intervention cannot be completely removed, and online automatic parameter adjustment is realized. In addition, the manual parameter adjustment and the other online adjustment methods need to depend on human experience, and under the condition of more reference factors, the parameter adjustment by the human experience is more and more difficult.

Under the circumstances, because the adjustment of the weight is more and more difficult, the situations that the weight is not updated timely and the like may occur, the accuracy of the finally recommended information is reduced, and the most suitable or matched recommended information cannot be recommended, so that the recommendation effect and the use experience of the user are reduced.

Referring to fig. 1, a schematic flowchart of a method for recommending target information according to an embodiment of the present disclosure is provided, where the data processing method includes the following steps:

step S100, dividing the request information of a plurality of different users into M parts, wherein the Nth part comprises a plurality of Nth request information sets, the (N + 1) th part comprises at least one (N + 1) th request information set, and the quantity of the request information included in the Nth request information set is smaller than that of the request information included in the (N + 1) th request information set. N is greater than or equal to 1 and less than or equal to M-1, and both N and M are positive integers.

Step S200, according to the Nth weight of at least one dimension information of the information to be recommended, the Nth target information matched with the request information in the Nth request information set is determined from the information to be recommended.

Step S300, feedback information after the Nth target information is recommended is obtained.

And step S400, updating the (N + 1) th weight of at least one dimension information of the information to be recommended according to the feedback information.

Step S500, based on the updated (N + 1) th weight, determining (N + 1) th target information matched with the request information in the (N + 1) th request information set from the information to be recommended.

The method can be executed in at least a mobile terminal, i.e. the execution body of the method can at least comprise the mobile terminal. The mobile terminal can include a mobile phone, a tablet computer, a vehicle-mounted central control device, a wearable device, an intelligent device and the like, and the intelligent device can include an intelligent office device, an intelligent home device and the like.

The method of the embodiment can be applied to various recommendation scenes, can be a recommendation scene in which a recommendation result is determined by one kind of dimension information, and can also be a recommendation scene in which a recommendation result is determined by a plurality of kinds of dimension information. In the recommendation scenes, the candidate information to be recommended has a plurality of pieces of dimension information which can be referred to, and final recommendation information can be determined according to the dimension information. Different pieces of dimension information may have different proportions in determining the final target information, so that different pieces of dimension information may have different reference values for determining the final target information, that is, different pieces of dimension information may have different weights.

Of course, any scenario in which the target information to be recommended is determined from the information to be recommended according to the different pieces of dimension information is within the scope of the embodiment.

In this embodiment, the information to be recommended includes at least one of: videos, public numbers, advertisements, articles, and the like. Of course, other types of content may also be included, which are not listed here, and it is within the scope of the embodiment that the content including multiple pieces of dimension information may be used as the information to be recommended.

In one embodiment, the dimension information of the information to be recommended at least comprises one of the following information: the method comprises the steps of downloading amount, collection amount, praise amount, the number of concerned users of the author of the alternative information, average browsing duration, comment amount, forwarding amount and the like. Of course, there may be at least one dimension information for the different alternative information that matches the alternative information. In this embodiment, the dimension information of the information to be recommended may be: the information recommendation method is generated according to the historical operation of the user who has received the information on the information to be recommended.

In another embodiment, the dimension information of the information to be recommended at least includes one of the following: the information content of the information to be recommended and the information content preferred by the recommended user are different, the similarity between the information content of the information to be recommended and the preferred content of the recommended user, the recommendation degree of the information to be recommended and the like.

In a specific application scenario, for example, when watching or browsing videos, public numbers, advertisements, articles, and the like, the videos, the public numbers, the advertisements, the articles, and the like may be downloaded, collected, complied with, commented on, and/or forwarded, authors of the videos, the public numbers, the advertisements, or the articles may be concerned, and the like. The different dimension information such as the downloading amount, the collection amount, the praise amount, the number of concerned users of the author of the alternative information, the average browsing duration, the comment amount, the forwarding amount and the like can reflect the preference degree of the user on the contents such as videos, public numbers, advertisements, articles and the like which are watched or browsed.

Different pieces of dimension information occupy different weights in the information to be recommended, and the weights of the pieces of dimension information in the information to be recommended can determine which pieces of information to be recommended are to be determined as target information, that is, the target information can be determined from the information to be recommended according to the weights of the pieces of dimension information in the information to be recommended. When a subsequent video, a public number, an advertisement or an article is recommended to a user, target information needing to be recommended can be determined from the information to be recommended according to information such as downloading, collection, approval, comment and/or forwarding of the video, the public number, the article or other content to be recommended.

Therefore, the weights of the information with different dimensions of the information to be recommended play a key role in determining the target information, and the embodiment is to automatically adjust the weights of the information with different dimensions of the information to be recommended, so that the description of the target information is determined from the information to be recommended according to the adjusted weights.

The request information of the user may include: the information of the request of the user for requesting the recommendation target information may be request information matched with an actual application program, and a specific format may be determined according to the actual application program, which is not limited herein.

For example, when an application for viewing a video is applied, when switching from a currently viewed video to a next video, request information for viewing the next video is generated and transmitted to a terminal or a server. Each switching action corresponds to a request message, the request message is used for indicating that the terminal or the server requests the next video which needs to be watched by the current user, and the terminal or the server recommends the next video to the current user according to the request message.

Of course, when information is to be recommended, other types of information may be generated and transmitted to the terminal or the server when the current content is switched to the next content.

With respect to step S100, when request information of a plurality of different users is addressed, the embodiment processes the plurality of request information and then recommends information for the plurality of request information.

When there are request information of a plurality of different users, the request information of the plurality of different users is divided into a plurality of parts, for example, into M parts. Then, the request information in each part is divided again, for example, the first part includes ten request information sets, and each request information set includes a plurality of request information.

The request information set included in the nth part of the M parts is used as the nth request information set, and the plurality of nth request information sets may be the same or different. For example, the ten request information sets in the first part are all used as the first request information sets, and may be specifically divided into a first request information set 1, a first request information set 2, a first request information set 3, and a … first request information set 10, where the number of request sets included in different first request sets may be the same or different. The 5 request information sets in the second part are all used as the second request information set, and may be specifically divided into a second request information set 1, a second request information set 2, a second request information set 3, and …, and a second request information set 5. The request information set included in the other part is also divided in the above manner.

The principle of the division may be that the number of request information in the request information set included in different parts is different, and the request information is divided in an increasing or decreasing order.

For example, the request information is divided into 3 parts, M equals to 3, of the first part, the first part includes 5 first request information sets, namely a first request information set 1, a first request information set 2, a first request information set 3, a first request information set 4 and a first request information set 5, and each first request information set includes the same amount of request information. When N is equal to 1, any one of the first request set 1 to the first request set 5 may be used as the first request information. The second part comprises 2 second request information sets including a second request information set 1 and a second request information set 2, and the quantity of the request information included in each second request information set is the same. When N is equal to 2, any one of the second request set 1 and the second request set 2 may be used as the second request information. The third portion includes a third set of request information.

The quantity of the request information included in the first request information set is smaller than that of the request information included in the second request information set; the number of request information included in the second request information set is smaller than the number of request information included in the third request information set. The number of request information included in the first request information set is sequentially increased to the number of request information included in the third request information set, or the number of request information included in the third request information set is sequentially decreased to the number of request information included in the first request information set.

That is, for different portions of request information sets, the request information may be sorted in a sequentially increasing or sequentially decreasing manner according to different amounts of request information included in the request information sets, for example, the amount of request information included in the first request information set to the amount of request information included in the mth request information set is sequentially increased, or the amount of request information included in the mth request information set to the amount of request information included in the first request information set is sequentially decreased. The number of request information included in the plurality of request information sets in the unified portion may be the same.

For example, the number of request information included in the 5 first request information sets is the same, and the number of request information included in the 2 second request information sets is the same.

The value of N may be any value of M-1, N is 1 or more and M-1 or less, and both N and M are positive integers. For example, when N is equal to one, the first request information sets are the first request information set 1 to the first request information set 5. The (N + 1) th request information set is a second request information set in the second part, such as the second request information set 1 and the second request information set 2. For any N, the number of request messages included in the nth request message set is smaller than the number of request messages included in the (N + 1) th request message set.

The number of M may be determined according to the number of request messages, and the number of request message sets included in each part may also be determined according to actual demand.

In one embodiment, the nth part includes a greater number of request information sets than the N +1 st part, i.e., the N-th request information sets are greater than the N +1 th request information sets.

In another embodiment, N is greater than or equal to 2 and less than or equal to M.

For step S200, after the request information is divided, each request information set in each part has a weight of at least one dimension information of the information to be recommended, and a plurality of request information in the same request information set share the same weight. The weights possessed by different sets of requested information may be the same or different. The weights possessed by different request information sets of the same part may be the same or different; the weights possessed by different parts of the requested information set may be the same or different.

In this embodiment, the weight possessed by each request information set in each part may be a weight set corresponding to at least one piece of dimension information, where the weight set includes the weight occupied by each piece of dimension information when determining the target information from the information to be recommended.

For example, the dimension information of the information to be recommended includes seven dimension information, namely a download amount, a collection amount, a praise amount, a number of users interested by an author of the candidate information, an average browsing duration, a number of comments, and a forwarding number, a weight possessed by a certain request information set may be represented as X ═ { a1, b1, c1, d1, e1, f1, g1}, and a1 to g1 may be different in X possessed by different request information sets. a1, b1, c1, d4, e1, f1 and g1 respectively represent the downloading amount, the collection amount, the praise amount, the number of the users concerned by the author of the alternative information, the average browsing duration, the number of comments and the forwarding number in determining the weight of the target information from the information to be recommended.

For another example, the weight possessed by the request information set in the nth part is taken as the nth weight. The first weight X1 possessed by the first request information set 1 in the first section is { a1, b1, c1, d1, e1, f1, g1}, the first weight X2 possessed by the first request information set 2 is { a2, b2, c2, d2, e2, f2, g2}, and the first weight X10 possessed by the first request information set 10 of … is { a10, b10, c10, d10, e10, f10, g10 }. The second weight X11 possessed by the second request information set 1 in the second part is { a11, b11, c11, d11, e11, f11, g11}, and the second weight X12 possessed by the second request information set 2 is { a12, b12, c12, d12, e12, f12, g12 }. The third set of request information in the third portion has a third weight X13 ═ a13, b13, c13, d13, e13, f13, g13 }.

The weight possessed by each request information set in each part is defaulted to an initial value in an initial state, and the initial value can be determined according to historical weight, can be randomly distributed, can be determined according to registration information of the user, such as gender, occupation, hobbies and the like, can be determined according to the geographical position of the user, and the like.

And then, according to the Nth weight owned by the Nth request information set, determining the Nth target matched with the request information in the Nth request information set from the information to be recommended.

For example, the first weight X1 possessed by the first request information set 1 in the first part, the target information matching the first request information set 1 may be determined from the information to be recommended according to the first weight X1, the target information matching the first request information set 2 may be determined from the information to be recommended according to the first weight X2, and so on. Therefore, the corresponding target information can be determined according to the weight possessed by each request information set, the determination of the target information matched with each request information set in each part is realized, and the target information is recommended to the request information in each request information set in each part.

For step S300, after determining the target information matched with the nth request information set in the nth part, recommending the corresponding target information to the user corresponding to the request information in the nth request information set. Then the corresponding user can obtain the corresponding target information and obtain the feedback information of the user to the recommended target information. The feedback information may include a degree of interest of the user in the target information, thereby facilitating updating of the corresponding weights according to the feedback information. The feedback information is determined through operations such as browsing time, approval or disapproval, collection or retransmission or not, the feedback information can be directly acquired through the terminal, and the feedback information can be presented in a numerical value information form.

The target information matched with each request information set in each part has corresponding feedback information.

For step S400, after the feedback information is obtained, the (N + 1) th weight of at least one dimension information of the information to be recommended is updated according to the feedback information.

And recording feedback information of the Nth target information matched with the Nth request information set in the Nth part as Nth feedback information, and updating the (N + 1) th weight according to the Nth feedback information.

The main objective of the step is to determine the accuracy of the nth weight to the recommended nth target information through the feedback information of the nth target determined by the nth weight, so as to update the (N + 1) th weight according to the nth weight, and further realize the update of the (N + 1) th weight.

Through the steps, the second weight owned by the second request information set in the second part is updated according to the first weight owned by the first request information set in the first part, the weight owned by the third request information set in the third part is updated according to the second weight owned by the second request information set in the second part, and the like, the weight is updated to the Mth weight owned by the Mth request information set in the Mth part, so that the second weight owned by the second request information set in the second part to the Mth weight owned by the Mth request information set in the Mth part are all updated.

The updating of the second weight to the Mth weight can be automatically realized without human participation, and can be performed on line, so that the adjustment of the weight is realized on line. More accurate target information can be determined according to the updated weight, so that the recommendation accuracy of the recommended target information is improved, more matched target information is recommended to the user, and the use experience of the user is improved.

For step S500, after the (N + 1) th weight is updated, based on the updated (N + 1) th weight, the (N + 1) th target information matched with the request information in the (N + 1) th request information set is determined from the information to be recommended, so as to recommend the (N + 1) th target information to the request information in the (N + 1) th request information set.

After the weight of each request information combination in each part is updated, corresponding target information is recommended to the request information included in each request information set in each corresponding part according to the updated weight owned by the request information set in each part.

According to the method, the weight corresponding to the request information set with the small quantity of the request information is updated, the target information matched with the request information set with the small quantity of the request information is determined according to the weight corresponding to the request information set with the small quantity of the request information, and the update of the weight corresponding to the request information set with the large quantity of the request information is determined according to the feedback information. The method saves the participation of manpower and reduces the difficulty of updating the weight manually.

In another aspect, when the target information matched with the request information set with the small number of request information is determined to have a problem according to the weight corresponding to the request information set with the small number of request information, large-scale users are not affected, and the influence on the users is reduced.

In another embodiment, a plurality of request messages in the nth set of request messages have the same nth weight. Each request information set in each part comprises a plurality of request information, and the plurality of request information in the same request information set have the same weight.

For example, if the first weight 1 is owned by the first request information set 1 in ten first request information sets included in the second portion, the multiple pieces of request information in the first request information set 1 share the first weight 1, that is, the multiple pieces of request information in the first request information set 1 have the same weight and are all the first weight 1. The first weight 2 is possessed by the first request information set 2, and the plurality of request information in the first request information set 2 share the first weight 2, that is, the plurality of request information in the first request information set 2 possess the same weight, which is the first weight 2. The second request information set 1 in the second part has the second weight 1, and then the plurality of request information in the second request information set 1 share the second weight 1, that is, the plurality of request information in the second request information set 1 have the same weight and are all the second weight 1.

For the same request information set, determining target information recommended to each request information in the request information set from the information to be recommended according to the same weight.

Referring to fig. 2, a schematic flow chart of obtaining feedback information is shown. Step S300, acquiring feedback information after the nth target information is recommended, including:

step S301, determining feedback dimension information of the nth target information and parameters of each feedback dimension information.

Step S302, determining a feedback value according to the feedback dimension information and the parameters of each feedback dimension information.

The main objectives of this embodiment are: after target information recommended to the request information in the corresponding request information set is determined according to the weight possessed by each request information set in each part, feedback information of the recommended target information is determined, and therefore whether the weight used in determining the target information is accurate or not is determined.

For step S301, the specific feedback dimension information may be determined according to at least one dimension information of the information to be recommended, and may be information that is the same as the at least one dimension information of the information to be recommended, that is, the at least one dimension information of the information to be recommended is also used as the feedback dimension information.

For example, the feedback dimension information may include at least one of:

the method comprises the following steps of downloading amount, collection amount, praise amount, the number of concerned users of an author of information to be recommended, average browsing duration, comment amount and forwarding amount. Of course, the corresponding feedback dimension information is different according to different information to be recommended, and the feedback dimension information is exemplified here as the feedback dimension information of the information to be recommended, such as videos, public numbers, advertisements, articles, and the like.

For another example, the feedback dimension information may also be: content duration, per-person play times, and so on.

Of course, the feedback dimension information may also be determined according to actual needs, for example, the feedback dimension information is preset according to the information to be recommended, that is, the feedback dimension information is preset according to the information to be recommended.

The feedback dimension information may be known in one embodiment.

After the inverse dimension information is determined, different feedbacks of the user to the recommended target information can be reflected by different feedback dimension information, and then parameters of the feedback dimension information are determined. The parameter of the feedback dimension information can represent the proportion or weight of the feedback dimension information, and the final feedback information result can be conveniently determined according to the parameter of the feedback dimension information.

For step S302, after determining the feedback information and the corresponding parameters, a feedback value is determined according to the feedback dimension information and the parameters of each feedback dimension information.

The step aims to quantize the feedback information and display the feedback information in a numerical value mode, so that the feedback information of the target information corresponding to different request information sets can be better compared.

The feedback value may be determined according to the sum of each feedback dimension information and the parameter of each feedback dimension information. For example, the feedback dimension information includes: the content playing time length is x, the per-person playing time length is y, and the per-person playing time number is z. For example, in the first feedback information 1 of the target information matched with the first request information set 1, the parameters of the feedback dimension information are (x1, y1, z1), in the first feedback information 2 of the target information matched with the first request information set 2, the parameters of the feedback dimension information are (x2, y2, z2), in the second feedback information 1 of the target information matched with the second request information set 1, the parameters of the feedback dimension information are (x3, y3, z3), and the like.

Each feedback information may have a quantized value, which may be the value of each feedback dimension information, or may be a converted value, etc. For example, the content duration, the per-person playing duration, and the per-person playing frequency may be specific values, and a feedback value may be determined according to the specific values of the content duration, the per-person playing duration, and the per-person playing frequency and the corresponding parameters, where the feedback value is a final result of the feedback information.

And recording the feedback value as R, wherein R can be determined according to the numerical value of each piece of feedback dimension information and the parameter corresponding to the numerical value of each piece of feedback dimension information. The expression for R may be R ═ x content duration + y average playout duration + z average playout times.

A feedback value of each target information may be determined so that the request information set may correspond to the feedback information, and further, a weight possessed by the request information set may correspond to the feedback information. After the feedback value of each target information is determined, updating of the weights may be facilitated.

In another embodiment, referring to fig. 3, a flow chart of updating the (N + 1) th weight is shown. Step S400, updating the (N + 1) th weight of at least one dimension information of the information to be recommended according to the feedback information, and comprising the following steps:

step S401, determining feedback values corresponding to the request information in the plurality of nth request information sets.

Step S402, determining the Nth weight corresponding to the request information in the Nth request information set with the maximum feedback value as the Nth target weight.

In step S403, the (N + 1) th weight is updated according to the nth target weight.

Since the feedback information after the nth target information is recommended has already been determined in step S300, the feedback information corresponding to the target information matched with each request information set may be determined according to step S300. After determining the feedback information of the target information matched with the respective request information sets in each part, the weights possessed by the plurality of N +1 th request information sets in the N +1 th part may be updated according to the feedback information of the target information respectively matched with the plurality of N th request information sets in the N th part and the weights possessed by the plurality of N th request information sets respectively.

Specifically, the weight of the nth request information set with the largest corresponding feedback value is determined from the weights corresponding to the at least two nth request information sets, and the weight is determined as the nth target weight. The nth target weight is used to update the (N + 1) th weight that the (N + 1) th request information set in the (N + 1) th part has. The larger the feedback value is, the better the recommendation effect of the target information determined according to the corresponding weight is, that is, the nth weight corresponding to the target information with the best recommendation effect is updated to the (N + 1) th weight. Therefore, the N +1 th target information determined according to the N +1 th weight is more accurate and is more matched with the requirements of the user.

For example, the second request information set 1 in the second part is updated according to the feedback information of the target information respectively matched with 5 first request information sets, namely, the first request information set 1, the first request information set 2, the first request information set 3, the first request information set 4 and the first request information set 5 in the first part. And according to feedback values corresponding to the 5 first request information sets of the woman, determining the first request information set with the maximum feedback value, and then updating a first weight owned by the first request information set with the maximum feedback value to a weight owned by a second request information set 1 in the second part.

Assemble the other 5 first request information in the first part: the first request information set 6, the first request information set 7, the first request information set 8, the first request information set 9 and the first request information set 10 respectively match the feedback information of the target information, and the other second request information set in the second part is updated: the second set of request information 2. According to the feedback values respectively corresponding to the first request information set 6 to the first request information set 10, the first request information set with the maximum feedback value is determined, and then the first weight possessed by the first request information set with the maximum feedback value is updated to the weight possessed by the second request information set 2 in the second part. This enables updating the second weight in dependence of the first weight.

For another example, according to feedback values corresponding to two second request information sets, namely, a second request information set 1 and a second request information set 2, the second request information set corresponding to the maximum feedback value is determined, and the weight possessed by the second request information set is used for determining the parameter possessed by the third request information set in the third part, so that the third weight is updated according to the second weight in the second part.

When the request information is divided into M parts, according to the principle, the weight owned by the second request information set in the adjacent second part is updated according to the weight owned by the first request information set in the first part, and then the weight owned by the third request information set in the third part is updated according to the updated weight owned by the second request information set, and so on until the weight owned by the mth request information set updated to the mth part completes the updating of the second weight to the mth weight.

In another embodiment, the mth section includes: an mth request information set.

After the request information of a plurality of different users is divided into a plurality of parts, the part with the largest amount of request information in the contained request information sets only has one request information set.

For example, the request information is divided into M parts, i.e., a first part to an mth part, if M is equal to 10, the 10 th part includes only 1 10 th request information sets, and the number of request information included in the 10 th request information set is the largest among the prime request information sets.

Referring to fig. 4, another method for recommending target information further includes:

step S600, after updating the mth weight, updating the first weights corresponding to the plurality of first request information sets included in the first part according to the updated mth weight.

Step S700, according to the updated first weight, determining the first target information matched with the request information in the first request information set from the information to be recommended again.

For step S600, after updating the weight owned by the mth request information set in the mth part, i.e., the mth weight, the weight owned by the first request information set in the first part is updated according to the mth weight. Therefore, the second weight owned by the second request information set in the second part is updated according to the first weight owned by the first request information set in the first part, the third weight owned by the third request information set in the third part is updated according to the updated second weight, and the like, until the first weight is updated according to the updated M-1 weight owned by the M-1 request information set in the M-1 part and the updated M-th weight owned by the M-th request information set in the M-1 part, and the first weight is updated according to the updated M-th weight, so that a weight updating cycle is formed.

The update procedure is not limited to this, and may be any procedure as long as the weights owned by the plurality of first request information sets in the first part are generated based on the mth weight. The number of weights generated from the mth weight is the same as the number of first request information sets in the first portion. The weight generated by the mth weight may be randomly assigned to the first request information sets in the first portion, and each pair of first request information sets in the first portion may be matched with a new weight to update the existing first weight, thereby implementing the update of the first weight.

For example, according to a third weight possessed by a third request information set of the third part, the first weights respectively corresponding to the 10 first request information sets included in the first part are updated. Specifically, 10 first weights may be generated according to the third weight, and the first weights owned by the 10 first request information sets in the first portion are updated according to the 10 newly generated first weights.

For step S700, according to the updated first weight, the first target information matching the request information in the first request information set is determined again from the information to be recommended. Due to the fact that the first weight is updated again, the target information which is more matched with the request information in the first request information set, namely the first target information, can be determined from the information to be recommended again according to the updated weight. And marking the target information matched with the request information in the Nth request information set as the Nth target information.

After each weight is updated circularly, each weight can be updated automatically, full automatic updating is realized, and the process of updating each weight under the artificial line is omitted, so that the difficulty in updating each weight under the artificial line can be reduced, the update of each weight is not timely, and the inaccurate condition of the recommended target information is further influenced. By the method, each weight can be automatically and timely updated, more matched target information can be recommended to the request information in the corresponding request information set according to the updated weight, the accuracy of the recommended target information is further improved, and user experience is improved.

In another embodiment, in step S600, after updating the mth weight, updating the first weights corresponding to the plurality of first request information sets included in the first part according to the updated mth weight, including:

and determining the same number of target weights as the first request information set included in the first part according to the updated Mth weight, and then updating the first weight according to the target weights.

Specifically, a new first weight, that is, a target weight, may be determined according to the updated mth weight through gaussian distribution, and a new value may be randomly generated through gaussian distribution in the specific process and determined through a mean and a variance. The current Mth weight is selected as a mean value through Gaussian distribution, the variance can be set according to actual requirements, and the target weight can be determined through the Gaussian distribution according to the mean value and the variance. The smaller the variance, the closer the value of the target weight randomly determined by the gaussian distribution is to the mean, i.e., fluctuates around the mean, and the larger the variance, the farther away the target weight randomly determined by the gaussian distribution is from the mean, i.e., is distributed at a place farther from the mean. The target weights may be randomly determined around the mean value, which may be selected in combination with the actual traffic demand, with a small variance. The target weight may also be determined from more locations, if desired, with a greater variance.

For example, 10 target weights are determined by the above method, and the first weights possessed by 10 first request information sets in the first part are updated by the 10 target weights. The 10 target weights and the 10 existing first weights may be randomly assigned, with each target weight updating one of the existing first weights.

In another embodiment, after updating the first weights corresponding to the plurality of first request information sets included in the first part, the method further includes:

updating the second weight to the Mth weight according to the updated first weight; and updating the (N + 1) th weight according to the updated (N) th weight.

And after the first weight is updated, updating the existing second weight according to the updated weight, updating the existing third weight according to the updated second weight, and so on, updating the existing Mth weight according to the updated M-1 th weight, and repeating the steps to realize the automatic updating of the weight owned by each request information set in each part.

In another embodiment, the step S100 of dividing the request information of a plurality of different users into M parts includes:

dividing the request information with the receiving time in the same time period into the same part according to the receiving time of the request information; and/or obtaining a ranking value according to the priority of the user sending the request information and a ranking factor corresponding to the receiving time of the request information; and dividing the request information with the sequencing values in the same interval into the same part.

A method divides a reception time of a reception request message. When request information of a plurality of different users is divided, the request information can be divided by referring to the time of receiving the request information, and the request information with the receiving time in the same time period is divided into the same part, so that the recommended target information can be determined according to the time information used by the users.

For example, the request messages received at eight to ten am are divided into the same part, e.g., the first part. The request information received from twelve am to one pm is divided into the same part, e.g., the second part. The request information received from eight to eleven nights is divided into the same part, for example, the third part. The total number of the request information included in each request information set of the first part is smaller than the total number of the request information included in each request information set of the second part; the total number of pieces of request information included in each of the second included request information sets is smaller than the total number of pieces of request information included in each of the third included request information sets. And the number of request information included in each request information set included in the first part is smaller than the number of request information included in each request information set included in the second part, and the number of request information included in each request information set included in the second part is smaller than the number of request information included in each request information set included in the third part.

This can facilitate updating the weight of the set including the greater number of pieces of request information by the weight of the set including the lesser number of pieces of request information, and can reduce the influence on the weight possessed by the set including the greater number of pieces of request information when it is determined that abnormality occurs in the target information recommended to the request information of the set including the lesser number of pieces of request information based on the weight used for the set including the lesser number of pieces of request information, thereby reducing the influence on the accuracy of determining that the target information is recommended to the request information of the set including the greater number of pieces of request information.

Another method divides according to the time of reception and user priority. The priority of the user who sends the request message may be determined according to a preset priority, for example, the priority of the user who opens a member is higher than that of the general user.

The sorting factors can be set according to needs, the priority of a user sending the request information and the receiving time of the request information can be respectively corresponding to the sorting factors, different sorting factors correspond to different numerical values, the final sorting value can be determined according to the sorting factors of the two sorting factors, the division of the request information is determined according to the sorting values, and the request information with the sorting values in the same interval is divided into the same part.

Illustratively, according to the receiving time sequence of the request information, the earlier the receiving time is, the larger the value of the first calculation factor is, the earlier the receiving time is; according to the user priority, the higher the priority, the larger the value of the second calculation factor.

The receiving time corresponds to a first weight factor, and the priority corresponds to a second priority;

the ranking factor may be obtained by performing a weighted average based on the first calculation factor, the first weighting factor, the second priority, and the second.

In another example, the number of request messages of each priority user is determined according to the priority, the priority divides the request messages of the same priority user into a request message set, and if the number of request messages included in the request message set is not enough, divides part or all of the request messages of the next priority user into the request message set. When the partial request information of the next priority is divided into the request information set, the partial request information with the previous receiving time is divided into the current request information set according to the receiving time of the request information of the user with the next priority.

In the embodiment of the present disclosure, the request information sets responded at different time points comprehensively consider the user priority and the receiving time of the request information, so that the average response delay control can be realized, and meanwhile, different requirements of users with different priorities on the response delay can be considered.

The method determines how to divide the request information by taking the priority of a user sending the request information and the receiving time of the request information as reference factors.

In another embodiment, referring to fig. 5, a schematic structural diagram of a recommendation apparatus for target information is shown, the recommendation apparatus includes:

a dividing module 1, configured to divide request information of multiple different users into M parts, where the nth part includes multiple nth request information sets, and the N +1 th part includes at least one (N + 1) th request information set, where a quantity of request information included in the nth request information set is smaller than a quantity of request information included in the (N + 1) th request information set; wherein N is greater than or equal to 1 and less than or equal to M-1, and both N and M are positive integers;

the Nth target information determining module 2 is configured to determine, according to an Nth weight of at least one piece of dimensional information of information to be recommended, Nth target information that matches request information in the Nth request information set from the information to be recommended;

the feedback information acquisition module 3 is used for acquiring feedback information after the Nth target information is recommended;

the updating module 4 is configured to update the (N + 1) th weight of the at least one dimension information of the information to be recommended according to the feedback information;

and the (N + 1) th target information determining module 5 is configured to determine, based on the updated (N + 1) th weight, N +1 th target information that is matched with the request information in the (N + 1) th request information set from the information to be recommended.

In another embodiment, the mth portion includes: one said mth request information set;

the device further comprises:

the first weight updating module is used for updating first weights corresponding to a plurality of first request information sets included in the first part according to the updated Mth weight after the Mth weight is updated;

and the first target information re-determination module is used for re-determining the first target information matched with the request information in the first request information set from the information to be recommended according to the updated first weight.

In another embodiment, the first weight updating module is specifically configured to: determining the target weights with the same quantity as the first request information set included in the first part according to the updated Mth weight through a cross entropy algorithm; updating the first weight according to the target weight.

In another embodiment, the apparatus further includes a second updating module, configured to update a second weight to the mth weight according to the updated first weight after updating the first weight corresponding to the plurality of first request information sets included in the first portion; and updating the (N + 1) th weight according to the updated (N) th weight.

In another embodiment, a plurality of request messages in the nth set of request messages have the same nth weight;

the feedback information obtaining module 3 includes:

and the first determining module is used for determining the feedback dimension information of the Nth target information and the parameters of each piece of feedback dimension information.

And the second determining module is used for determining a feedback value according to the feedback dimension information and the parameters of each piece of feedback dimension information.

In another embodiment, the update module 4 includes:

a feedback value determining module, configured to determine the feedback values corresponding to the request information in the nth request information sets respectively;

an nth target weight determining module, configured to determine an nth weight corresponding to request information in the nth request information set with a largest feedback value as the nth target weight;

and the (N + 1) th weight updating module is used for updating the (N + 1) th weight according to the (N) th target weight.

In another embodiment, the feedback dimension information includes at least one of:

the method comprises the following steps of downloading amount, collection amount, praise amount, the number of concerned users of an author of information to be recommended, average browsing duration, comment amount and forwarding amount.

In another embodiment, the partitioning module 1 includes:

the first dividing module is used for dividing the request information with the receiving time in the same time period into the same part according to the receiving time for receiving the request information;

and/or the presence of a gas in the gas,

the second division module is used for obtaining a ranking value according to the priority of a user sending the request information and a ranking factor corresponding to the receiving time of the request information; and dividing the request information with the sorting values in the same interval into the same part.

In another embodiment, the information to be recommended includes at least one of: video; a public number; an article; advertising;

the dimension information includes at least one of: the method comprises the following steps of downloading amount, collection amount, praise amount, the number of concerned users of an author of information to be recommended, average browsing duration, comment amount and forwarding amount.

In another embodiment, there is also provided an electronic device including:

In another embodiment, a non-transitory computer-readable storage medium is also provided, having stored therein computer-executable instructions that, when executed by a processor, implement the method of any of the above embodiments.

In another embodiment, another method for recommending target information is also provided.

In the information age of today, the amount of data is increasing day by day, and in mass data, it is often difficult to determine the required information in the face of various choices. Therefore, the recommendation system is developed in order to solve the problem of making reasonable selection in mass data so as to determine needed information. Taking video recommendation as an example, the recommendation system determines information which is recommended to the user and is possibly interested by the user by using a corresponding recommendation algorithm according to historical behavior data of the user, portrait data of the user and statistical data characteristics of the user side or the video side. The recommendation accuracy rate is higher and higher with the update of the recommendation system, but the corresponding problem is the requirement of more complex recommendation system and update experience.

Particularly, in the current personalized recommendation scene of the information flow, recommendation algorithms are changing day by day, such as a multi-target recommendation model in the middle of the day is widely applied. There is less display feedback in the recommendation system because most user feedback is not directly scored, and recommendation systems mostly recommend based on implicit feedback, such as user's attention, sharing, likes, watching duration, and full play. There may be some cognitive bias in evaluating user satisfaction, including mainly target bias, item bias, user bias, etc. In order to solve the deviation of the target, the deviation of the article, the deviation of the user and the like, a multi-model score fusion mode is adopted, each optimization target predicts a score by an independent model, and finally the scores are fused in a weighting mode. Each model has a score, and a final score is designed according to the weight of the model to complete the multi-objective optimization.

Generally, methods adopted for a multi-target parameter adjusting technology are various and include manual parameter adjustment and automatic parameter adjustment, but the parameter adjustment modes are all offline adjustment and then updated online no matter whether the parameters are manually or automatically adjusted, and iteration of online automatic parameter adjustment cannot be realized without manual intervention. In addition, although the manual parameter adjustment is light-weight and well-interpretable, the automatic parameter adjustment requires much experience for the analysis of the association between a plurality of targets depending on the experience of an engineer. Especially when the parameters are expanded to 10, 20, or even more, the adjustment under the wire becomes more difficult. So if this set of parameters could themselves receive rewards based on changes in their parameters and then interact with the environment, new information could be obtained and then the user could change his own rewards. Therefore, the parameters can change themselves and move to a better direction by obtaining the reward, and the human intervention is avoided.

In the recommendation system, the request information includes traffic, which refers to access requests of the client. Taking a video recommendation system as an example, each time a user requests information, the system can obtain results through a plurality of layers and models according to the information in the request and offline data, and return the results to the user for watching. The split flow refers to that the access request information of the client is divided into M parts, and each part has a plurality of request information sets. For example, each section may include several buckets, each of which may have its own system black box for traffic and may have different model parameters. The final selection can be decided according to the feedback in different black boxes. The scheme selects a mode of feeding back small flow to large flow, and mainly has two aspects of consideration, on one hand, the experiment of small flow only affects a small part of users, and once problems occur, large-scale problems cannot be caused. On the other hand, each flow bucket is randomly sampled, and the overall effect is fed back by the effect of small flow assuming that the flow buckets meet the normal distribution.

Referring to fig. 6, a schematic diagram of another method is shown.

The method comprises the following steps: 100% of flow is divided into a plurality of small flows, and then the selection of large flow is determined according to the feedback of small flow, so as to form a feedback type parameter updating mode. For example, 100% of the flow is divided into three portions, including a third portion, a second portion, and a first portion. The third part comprises a request information set (also called 30% flow) comprising 30% request information, the second part comprises two request information sets (also called 20% flow) respectively comprising 20% request information, the first part comprises 10 request information sets (also called 3% flow) respectively comprising 3% request information, and the total sum of the request information included by all the request information sets is 100%. The feedback effect of the target information determined by the parameters in 3% of the request information sets can determine how to determine and update the parameters in 20% of the request information sets, and then the parameters in 30% of the request information sets are determined by the parameters in 20% of the request information sets.

Step two: the 10 request information sets comprising 3% are divided into two groups of 5 request information sets comprising 3%. Each 3% of the requested information sets of each group will have its own set of parameters (a, b, c, etc.), such as an + praise + bn + cn + share + etc. (n-1, 2,3,4, 5). From this parameter, target information recommended to each of the request information sets including 3% can be determined. And then, determining feedback information according to feedback information of the target information, for example, by using a return equation of reward ═ p × content duration + t × average playing duration + e × average playing times +, and the like, and taking the content duration, the average playing duration, and the average playing times as feedback dimension information. The set of parameters an, bn, cn, etc. that yields the best feedback results comprises 3% of the requested information set.

Step three: the parameters comprising 20% of the requested information set are updated according to the set of parameters comprising 3% of the requested information set with the best feedback effect. Since the request information sets including 3% are divided into 2 groups, two sets of parameters including 3% of the request information sets with the best feedback effect are obtained, and then the 2 sets of parameters including 20% of the request information sets are updated respectively.

Step four: according to the feedback information comprising 2 groups of 20% request information sets, a group of parameters comprising 20% request information sets with the best feedback effect is selected, and the parameters comprising 30% request information sets are updated.

Step five: from the feedback information of the target information comprising 30% of the requested information set, for example, a reward equation reward is derived, and if it works better than the feedback of the previous set, the new parameters are used subsequently. Then through a cross entropy algorithm, 10 groups of parameters are selected by using the current parameters, and the parameters are respectively updated to 10 request information sets comprising 3% of request information.

Step six: and circularly executing the steps, continuously iterating the parameters of each request information set, and finishing the continuous iteration of the parameters.

It should be noted that "first" and "second" in the embodiments of the present disclosure are merely for convenience of description and distinction, and have no other specific meaning.

Fig. 7 is a block diagram illustrating a terminal device according to an example embodiment. For example, the terminal device may be a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, a fitness device, a personal digital assistant, and the like.

Referring to fig. 7, the terminal device may include one or more of the following components: a processing component 802, a memory 804, a power component 806, a multimedia component 808, an audio component 810, an input/output (I/O) interface 812, a sensor component 814, and a communication component 816.

The processing component 802 generally controls overall operation of the terminal device, such as operations associated with presentation, telephone calls, data communications, camera operations, and recording operations. The processing components 802 may include one or more processors 820 to execute instructions to perform all or a portion of the steps of the methods described above. Further, the processing component 802 can include one or more modules that facilitate interaction between the processing component 802 and other components. For example, the processing component 802 can include a multimedia module to facilitate interaction between the multimedia component 808 and the processing component 802.

The memory 804 is configured to store various types of data to support operations at the terminal device. Examples of such data include instructions for any application or method operating on the terminal device, contact data, phonebook data, messages, pictures, videos, etc. The memory 804 may be implemented by any type or combination of volatile or non-volatile memory devices such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disks.

The power component 806 provides power to various components of the terminal device. The power components 806 may include a power management system, one or more power sources, and other components associated with generating, managing, and distributing power for the terminal device.

The multimedia component 808 includes a screen that provides an output interface between the terminal device and the user. In some embodiments, the screen may include a Liquid Crystal Display (LCD) and a Touch Panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive an input signal from a user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor may not only sense the boundary of a touch or slide action, but also detect the duration and pressure associated with the touch or slide operation. In some embodiments, the multimedia component 808 includes a front facing camera and/or a rear facing camera. When the terminal device is in an operation mode, such as a shooting mode or a video mode, the front camera and/or the rear camera can receive external multimedia data. Each front camera and rear camera may be a fixed optical lens system or have a focal length and optical zoom capability.

The audio component 810 is configured to output and/or input audio signals. For example, the audio component 810 includes a Microphone (MIC) configured to receive external audio signals when the terminal device is in an operational mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signals may further be stored in the memory 804 or transmitted via the communication component 816. In some embodiments, audio component 810 also includes a speaker for outputting audio signals.

The I/O interface 812 provides an interface between the processing component 802 and peripheral interface modules, which may be keyboards, click wheels, buttons, etc. These buttons may include, but are not limited to: a home button, a volume button, a start button, and a lock button.

The sensor component 814 includes one or more sensors for providing various aspects of state assessment for the terminal device. For example, sensor assembly 814 may detect the open/closed status of the terminal device, the relative positioning of components, such as the display and keypad of the terminal device, the change in position of the terminal device or a component of the terminal device, the presence or absence of user contact with the terminal device, the orientation or acceleration/deceleration of the terminal device, and the change in temperature of the terminal device. Sensor assembly 814 may include a proximity sensor configured to detect the presence of a nearby object without any physical contact. The sensor assembly 814 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor assembly 814 may also include an acceleration sensor, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

The communication component 816 is configured to facilitate communication between the terminal device and other devices in a wired or wireless manner. The terminal device may access a wireless network based on a communication standard, such as WiFi, 2G or 3G, or a combination thereof. In an exemplary embodiment, the communication component 816 receives a broadcast signal or broadcast related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, communications component 816 further includes a Near Field Communications (NFC) module to facilitate short-range communications. For example, the NFC module may be implemented based on Radio Frequency Identification (RFID) technology, infrared data association (IrDA) technology, Ultra Wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

In an exemplary embodiment, the terminal device may be implemented by one or more Application Specific Integrated Circuits (ASICs), Digital Signal Processors (DSPs), Digital Signal Processing Devices (DSPDs), Programmable Logic Devices (PLDs), Field Programmable Gate Arrays (FPGAs), controllers, micro-controllers, microprocessors or other electronic components for performing the above-described methods.

Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of the disclosure following, in general, the principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.

It will be understood that the present disclosure is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A method for recommending target information, comprising:

dividing request information of a plurality of different users into M parts, wherein the Nth part comprises a plurality of Nth request information sets, the N +1 th part comprises at least one (N + 1) th request information set, and the quantity of the request information in the Nth request information set is smaller than that in the (N + 1) th request information set; wherein N is greater than or equal to 1 and less than or equal to M-1, and both N and M are positive integers;

determining Nth target information matched with the request information in the Nth request information set from the information to be recommended according to the Nth weight of at least one dimension information of the information to be recommended;

acquiring feedback information after the Nth target information is recommended;

updating the (N + 1) th weight of at least one dimension information of the information to be recommended according to the feedback information;

and determining the (N + 1) th target information matched with the request information in the (N + 1) th request information set from the information to be recommended based on the updated (N + 1) th weight.

2. The method of claim 1, wherein the mth portion comprises: one said mth request information set;

the method further comprises the following steps:

after the Mth weight is updated, updating a first weight corresponding to a plurality of first request information sets included in the first part according to the updated Mth weight;

and according to the updated first weight, determining first target information matched with the request information in the first request information set from the information to be recommended again.

3. The method according to claim 2, wherein after updating the mth weight, updating the first weights corresponding to the plurality of first request information sets included in the first part according to the updated mth weight, and including:

determining the target weights with the same quantity as the first request information set included in the first part according to the updated Mth weight;

updating the first weight according to the target weight.

4. The method of claim 2, further comprising, after updating the first weights corresponding to the plurality of first request information sets included in the first portion:

updating a second weight to the Mth weight according to the updated first weight; and updating the (N + 1) th weight according to the updated (N) th weight.

5. The method of claim 1, wherein a plurality of request messages in the Nth set of request messages have the same Nth weight;

the obtaining of the feedback information after the nth target information is recommended includes:

determining feedback dimension information of the Nth target information and parameters of each piece of feedback dimension information;

and determining a feedback value according to the feedback dimension information and the parameters of each piece of feedback dimension information.

6. The method according to claim 5, wherein the updating the (N + 1) th weight of the at least one dimension information of the information to be recommended according to the feedback information comprises:

determining the feedback values corresponding to the request information in the plurality of Nth request information sets respectively;

determining an nth weight corresponding to the request information in the nth request information set with the maximum feedback value as the nth target weight;

updating the N +1 th weight according to the Nth target weight.

7. The method of claim 5, wherein the feedback dimension information comprises at least one of:

8. The method of claim 1, wherein the dividing the request information of the plurality of different users into M parts comprises:

dividing the request information with the receiving time in the same time period into the same part according to the receiving time for receiving the request information;

and/or the presence of a gas in the gas,

obtaining a ranking value according to the priority of the user sending the request information and a ranking factor corresponding to the receiving time of the request information; and dividing the request information with the sequencing values in the same interval into the same part.

9. The method according to any one of claims 1 to 8, wherein the information to be recommended comprises at least one of:

video;

a public number;

an article;

advertising;

the dimension information includes at least one of:

10. An apparatus for recommending object information, comprising:

the device comprises a dividing module, a sending module and a receiving module, wherein the dividing module is used for dividing request information of a plurality of different users into M parts, the N part comprises a plurality of N request information sets, the N +1 part comprises at least one N +1 request information set, and the quantity of the request information in the N request information set is smaller than that in the N +1 request information set; wherein N is greater than or equal to 1 and less than or equal to M-1, and both N and M are positive integers;

the Nth target information determining module is used for determining Nth target information matched with the request information in the Nth request information set from the information to be recommended according to the Nth weight of at least one dimension information of the information to be recommended;

the feedback information acquisition module is used for acquiring feedback information after the Nth target information is recommended;

the updating module is used for updating the (N + 1) th weight of at least one dimension information of the information to be recommended according to the feedback information;

and the (N + 1) th target information determining module is used for determining (N + 1) th target information matched with the request information in the (N + 1) th request information set from the information to be recommended based on the updated (N + 1) th weight.

11. An electronic device, comprising:

the processor is configured to execute the executable instructions, when the executable instructions are executed, to perform the method of any of the preceding claims 1 to 9.

12. A non-transitory computer-readable storage medium having stored thereon computer-executable instructions that, when executed by a processor, perform the method of any of claims 1 to 9.