CN107229622B

CN107229622B - Information processing method and server

Info

Publication number: CN107229622B
Application number: CN201610170054.1A
Authority: CN
Inventors: 陈蓉
Original assignee: Tencent Technology Beijing Co Ltd
Current assignee: Tencent Technology Beijing Co Ltd
Priority date: 2016-03-23
Filing date: 2016-03-23
Publication date: 2021-02-05
Anticipated expiration: 2036-03-23
Also published as: CN107229622A

Abstract

The embodiment of the invention discloses an information processing method and a server, wherein the method can comprise the following steps: when receiving an association request generated based on user behavior, acquiring M pieces of media information according to media information types specified by the association request; preprocessing the playing data of the M media information to obtain M effective data, wherein the M effective data comprise first data corresponding to one piece of media information to be associated in the M media information and second data corresponding to other M-1 pieces of media information; performing operation between at least two pieces of media information on the M effective data according to a first preset strategy based on the appointed media information category to obtain a characteristic value for representing the correlation between the first data and the second data; obtaining a first correlation value used for representing the correlation of the media information according to the characteristic value and a second preset strategy; and determining the media information associated with the media information to be associated from the M-1 media information according to the first relevance value.

Description

Information processing method and server

Technical Field

The present invention relates to internet technologies in the field of communications, and in particular, to an information processing method and a server.

Background

With the rapid development of communication technology, the functions and intelligence of the client are more and more abundant, when a user performs a service application through the client, for example, when the video application installed on the client plays a video, a service request such as a play request or a request for downloading the video is sent, the server collects the service requests, and issues associated information to the client according to the service request, and the client can intelligently recommend associated media information (such as a certain variety video or a tv series video) to the user through the associated information, so that the user can conveniently select more interested related media information.

For the server to implement intelligent recommendation of a video associated with user attention or request by judging the user, in the prior art, the server calculates an association value of two items (media information, service application, and the like) by using an association rule based on a behavior of the user (such as a service request of the user), and determines whether the two items are associated (that is, an associated condition is reached) by comparing the association value with an experimentally obtained threshold value, so as to issue the determined association information to the client, thereby implementing intelligent recommendation for the user logging in the client. The specific method for the server to calculate the association value between the two items X and Y by using the association rule may be:

(1) the server directly utilizes the association rule to calculate the support degree and the confidence degree between X and Y; the method specifically comprises the following steps: if the relationship between X and Y is shown in FIG. 1, where N is the range of the items preset in the client, and X and Y are two items in N, the server calculates the support degree (support)

Confidence (confidence):

then, the server judges whether X and Y are related according to whether the sum is larger than a set threshold value, if so, the X and Y are related, otherwise, the X and Y are not related.

(2) The server can calculate a lift degree lift value by using the association rule, and judge the association between the two items according to the lift value. The larger the lift value is, the more the two items are related, specifically: if the relationship between X and Y is as shown in FIG. 1, then:

where a and b are smoothing factors. Then, the server compares the lift value with the experimentally obtained threshold value to determine the correlation between X and Y.

However, no matter which method in the prior art is adopted, when the data distribution corresponding to X and Y changes, the setting of the threshold value obtained by experiment or the smoothing factor (such as a and b) needs to be adjusted, how to adjust the threshold value and the smoothing factor (such as a and b) also needs a lot of experiments and experience support, and most of the adjustments need to be manually set, so that the difficulty in implementation is high, and when the data distribution corresponding to X and Y changes, the stability and accuracy of the server in judging the relevance can be affected.

Disclosure of Invention

In order to solve the above technical problem, embodiments of the present invention desirably provide a data association method and a server, which can improve the stability of association determination on the basis of ensuring the determination accuracy of association.

The technical scheme of the invention is realized as follows:

the embodiment of the invention provides an information processing method, which comprises the following steps:

when receiving an association request generated based on user behavior, acquiring M pieces of media information according to media information types specified by the association request; m is a positive integer greater than 1;

preprocessing the playing data of the M pieces of media information to obtain M pieces of effective data, wherein the M pieces of effective data comprise first data corresponding to one piece of media information to be associated in the M pieces of media information and second data corresponding to other M-1 pieces of media information;

performing operation between at least two pieces of media information on the M effective data according to a first preset strategy based on the appointed media information category to obtain a characteristic value for representing the correlation between the first data and the second data;

obtaining a first correlation value for representing the correlation of the media information according to the characteristic value for representing the correlation between the first data and the second data and a second preset strategy;

and determining the media information associated with the media information to be associated from the M-1 media information according to a first relevance value for representing the relevance of the media information.

In the foregoing solution, after obtaining a first correlation value for characterizing a correlation of media information according to the feature value for characterizing a correlation between the first data and the second data and a second preset policy, the method further includes:

according to the first data and a third preset strategy, obtaining a second correlation value used for representing the correlation of the first data in the M effective data; and/or the presence of a gas in the gas,

according to the second data and a fourth preset strategy, obtaining a third correlation value of the correlation of each data in the second data in the M effective data;

and obtaining a total correlation value for representing the correlation of the media information according to the first correlation value, the second correlation value, the third correlation value and a fifth preset strategy.

In the foregoing solution, the determining, according to a first relevance value used for characterizing relevance of media information, media information associated with the media information to be associated from the M-1 media information includes:

sorting the first correlation values corresponding to the second data according to the numerical values;

and determining the media information associated with the media information to be associated from the M-1 media information according to the association request and the sequencing result.

In the foregoing solution, the associating request includes a request for data having a correlation with H pieces of the first data, where H is a positive integer greater than or equal to 1 and less than M, and determining, according to the associating request and a sorting result, media information associated with the media information to be associated from the M-1 pieces of media information includes:

determining H maximum first correlation values in the sequencing result according to the value of H and the sequencing result;

and determining the H pieces of media information corresponding to the H maximum correlation values as the media information associated with the media information to be associated.

In the foregoing solution, after determining, according to a first correlation value used for characterizing correlation of media information, media information associated with the media information to be associated from the M-1 media information, the method further includes:

and issuing the media information associated with the media information to be associated.

In the above scheme, the M pieces of media information include one piece of media information to be associated and other M-1 pieces of media information, and the piece of media information to be associated and the other M-1 pieces of media information belong to the same category;

and the M effective data are effective data based on user behaviors obtained after data preprocessing and filtering.

The embodiment of the invention provides a server, which comprises:

the acquiring unit is used for acquiring M pieces of media information according to the media information types specified by the association request when the receiving unit receives the association request generated based on the user behavior; m is a positive integer greater than 1;

the computing unit is used for preprocessing the playing data of the M pieces of media information acquired by the acquiring unit to obtain M pieces of effective data, wherein the M pieces of effective data comprise first data corresponding to one piece of media information to be associated in the M pieces of media information and second data corresponding to other M-1 pieces of media information; performing operation between at least two pieces of media information on the M effective data according to a first preset strategy based on the appointed media information type to obtain a characteristic value for representing the correlation between the first data and the second data; obtaining a first correlation value for representing the correlation of the media information according to the characteristic value for representing the correlation between the first data and the second data and a second preset strategy;

and the determining unit is used for determining the media information related to the media information to be related from the M-1 pieces of media information acquired by the acquiring unit according to the first relevance value which is calculated by the calculating unit and used for representing the relevance of the media information.

In the server, the calculating unit is further configured to obtain, after obtaining a first correlation value used for characterizing correlation of media information according to the feature value used for characterizing correlation between the first data and the second data and a second preset policy, a second correlation value used for characterizing correlation of the first data in the M valid data according to the first data and a third preset policy acquired by the acquiring unit; and/or obtaining a third correlation value of the correlation of each data in the second data in the M effective data according to the second data acquired by the acquisition unit and a fourth preset strategy; and obtaining a total correlation value for representing the correlation of the media information according to the first correlation value, the second correlation value, the third correlation value and a fifth preset strategy.

In the above server, the server further includes: a sorting unit;

the sorting unit is configured to sort the first correlation values corresponding to the second data calculated by the calculating unit according to the numerical values;

the determining unit is specifically configured to determine, according to the association request received by the receiving unit and the sorting result obtained by the sorting unit, media information associated with the media information to be associated from the M-1 media information acquired by the acquiring unit.

In the server, the association request includes a request for data having a correlation with H pieces of the first data, where H is a positive integer greater than or equal to 1 and less than M;

the determining unit is further configured to determine, according to the value of H received by the receiving unit and the sorting result obtained by the sorting unit, H maximum first correlation values calculated by the calculating unit in the sorting result;

the determining unit is further specifically configured to determine the H pieces of media information corresponding to the H largest correlation values as media information associated with the media information to be associated.

In the above server, the server further includes: a transmitting unit;

the sending unit is configured to, after the determining unit determines the media information associated with the media information to be associated from the M-1 media information according to the first correlation value used for representing the correlation of the media information, issue the media information associated with the media information to be associated determined by the determining unit.

In the server, the M pieces of media information acquired by the acquiring unit include one piece of media information to be associated and other M-1 pieces of media information, and the piece of media information to be associated and the other M-1 pieces of media information belong to the same category;

the M effective data calculated by the calculating unit are effective data based on user behaviors obtained after data preprocessing and filtering.

The embodiment of the invention provides a data association method and a server, wherein when an association request generated based on user behavior is received, M pieces of media information are obtained according to media information types specified by the association request; m is a positive integer greater than 1; preprocessing the playing data of the M pieces of media information to obtain M pieces of effective data, wherein the M pieces of effective data comprise first data corresponding to one piece of media information to be associated in the M pieces of media information and second data corresponding to other M-1 pieces of media information; performing operation between at least two pieces of media information on the M effective data according to a first preset strategy based on the appointed media information category to obtain a characteristic value for representing the correlation between the first data and the second data; obtaining a first correlation value for representing the correlation of the media information according to a characteristic value for representing the correlation between the first data and the second data and a second preset strategy; and determining the media information associated with the media information to be associated from the M-1 media information according to the first relevance value for representing the relevance of the media information. By adopting the technical implementation scheme, the first correlation value for representing the correlation of the media information obtained by the server is calculated by the characteristic value of the correlation between the first data and the second preset strategy, and the characteristic value is calculated according to the M effective data and the first preset strategy, so that the first correlation value obtained by the server is an exact numerical value, the server judges according to the size of the exact numerical value, the media information with high correlation with the media information to be correlated can be obtained, and the method can be realized without the assistance of uncertain parameters or experimental numerical values. In short, the server can improve the stability of the relevance determination on the basis of ensuring the accuracy of the relevance determination.

Drawings

Fig. 1 is a relationship diagram in the prior art of an information processing method according to an embodiment of the present invention;

fig. 2 is a system block diagram of an information processing method according to an embodiment of the present invention;

fig. 3 is a first flowchart of an information processing method according to an embodiment of the present invention;

fig. 4 is an exemplary eigenvalue graph of an information processing method provided by the embodiment of the present invention;

FIG. 5 is a diagram illustrating a client interacting with a server according to a first embodiment of the present invention;

fig. 6 is a second flowchart of an information processing method according to an embodiment of the present invention;

fig. 7 is a third flowchart of an information processing method according to an embodiment of the present invention;

fig. 8 is a fourth flowchart of an information processing method according to an embodiment of the present invention;

FIG. 9 is a diagram illustrating client-server interaction according to a fourth embodiment of the present invention;

fig. 10 is a first schematic structural diagram of a server according to an embodiment of the present invention;

fig. 11 is a schematic structural diagram of a server according to an embodiment of the present invention;

fig. 12 is a schematic structural diagram of a server according to an embodiment of the present invention;

fig. 13 is a schematic structural diagram of a server according to a fourth embodiment of the present invention.

Detailed Description

The technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

Fig. 2 is a schematic structural diagram of a system for implementing an embodiment of the present invention. Referring to fig. 2, a system of a client and a server for media information application on a terminal is shown. The system 2 comprises one or more client-equipped terminals 21 and one or more servers 22, the terminals 21 and servers 22 being connected by a network 23. The terminal 21 has installed therein a client capable of media information application, and the user plays and downloads media information and the like from the server 22 by using the client.

Example one

An embodiment of the present invention provides an information processing method, as shown in fig. 3, the method may include:

s101, when receiving a correlation request generated based on user behavior, acquiring M pieces of media information according to media information types specified by the correlation request; m is a positive integer greater than 1.

It should be noted that the information processing method provided in the embodiment of the present invention is suitable for a case where, when a user uses a client to perform a media information application, the client obtains, through a server, related media information based on a user behavior of the media application. The embodiment of the invention mainly provides a method for processing information based on a server side.

Optionally, the media information in the embodiment of the present invention may be video information, audio information, or picture information, and the like, and the embodiment of the present invention is not limited.

It should be noted that the user behavior-based operations in the embodiments of the present invention refer to the number of times that the user uses the media information or the number of times that the user uses the media information, for example, operations performed by the user, such as playing a video, downloading a video, collecting a certain video, or paying attention to a certain video, and the specific setting of the user behavior is not limited in the embodiments of the present invention.

It is understood that when a user uses one piece of media information through a client and wants media information related to the one piece of media information (media information to be associated), the client is implemented by sending an association request generated based on user behavior to a server.

It should be noted that the client may be installed on a terminal, an internet output terminal such as a network video medium is used as a terminal in the embodiment of the present invention, the terminal is used for outputting the form of the multimedia information, and various internet output terminals such as a mobile phone or a tablet computer in the internet output terminal receive a network data packet through an internet to output the multimedia information.

Optionally, the association request in the embodiment of the present invention may include: the name of the media information to be associated and the category of the media information to be associated. The category of the video information to be associated may be a movie, a television show, a comprehensive art, a sport, an animation, a video, a primary advertisement, and the like, and the embodiment of the present invention is not limited.

Specifically, when the server receives the association request, the server obtains M pieces of media information according to the type of the media information to be associated.

It should be noted that, because the information processing method provided in the embodiment of the present invention mainly needs to acquire the media information associated with the media information to be associated, the server must also acquire the media information to be associated in the process of performing the correlation, that is, M pieces of media information to be acquired by the server include one piece of media information to be associated, and the server needs to find out the media information with high correlation with the media information to be associated from other M-1 pieces of media information. In this way, at least 2M media information items can implement the information processing method in the embodiment of the present invention.

Optionally, the M pieces of media information include one piece of media information to be associated and other M-1 pieces of media information, and the piece of media information to be associated and the other M-1 pieces of media information belong to the same category.

It should be noted that the categories of the media information to be associated and the other M-1 media information in the embodiment of the present invention should be consistent. For example, in a video application, when a user wants to find a video with the same high click rate as an a video, or a user who has viewed an a video but also what video the server has viewed, it is necessary that the server determines a video with high correlation with the video to be correlated, not music. Therefore, the specified media information category in the embodiment of the present invention is the media information category to be associated.

Further, the triggering of the association request in the embodiment of the present invention may be that the user actively sends the to-be-associated media information to the server when the user triggers the to-be-associated media information to be played or used, or that the to-be-associated media information is generated and sent to the server through the user operating the associated function key on the play or use interface when the user triggers the to-be-associated media information to be played or used, and a specific implementation manner of the embodiment of the present invention is not limited.

Further, when the user plays or triggers the media information through the client, the client reports the media information triggered each time to the server and stores the media information to form a media information base, and when the server receives an association request generated based on user behavior, the server can acquire M pieces of media information of the same media information type from the media information base triggered by the user according to the specified media information type.

S102, preprocessing the playing data of the M pieces of media information to obtain M pieces of effective data, wherein the M pieces of effective data comprise first data corresponding to one piece of media information to be associated in the M pieces of media information and second data corresponding to other M-1 pieces of media information.

After the server acquires the M pieces of media information according to the media information category specified by the association request, since the server calculates the degree of correlation of the data to be associated based on the user behavior, and the behavior of the user for the media information is playing, the server can acquire the playing data corresponding to each of the M pieces of media information in the M pieces of media information. Since there may be some malicious behaviors generated by the user that the user carelessly or maliciously triggers many times in a media information library of the server to generate the user playing data ("dirty data"), the server needs to pre-process the dirty data in the M playing data to obtain M valid data. The M effective data comprise first data corresponding to one piece of media information to be associated in the M pieces of media information and second data corresponding to other M-1 pieces of media information.

Optionally, the M effective data are effective data based on user behavior obtained after data preprocessing and filtering.

Optionally, the valid data in the embodiment of the present invention may include: media information category, media information name, user playing the media information, etc.

It should be noted that, in the embodiment of the present invention, the correlation between the media information to be associated and other M-1 media information is determined for the processing of M valid data.

Illustratively, one sequence (valid data) can be obtained for each video (media information) as follows:

Type,Videoid:Uin1,Uin2,Uin3...

wherein, type is a video type (media information category), video is a video name (media information name), and uin is a user who has viewed the video.

S103, performing operation between at least two pieces of media information on the M effective data according to a first preset strategy based on the appointed media information category to obtain a characteristic value for representing the correlation between the first data and the second data.

After the server preprocesses the playing data of the M pieces of media information to obtain M pieces of valid data, the server may perform arithmetic processing on the M pieces of valid data to obtain various relationship states of at least two pieces of media information in the M pieces of media information. Specifically, the server performs an operation between at least two pieces of media information on the M pieces of valid data according to a first preset policy based on a specified media information category (media category to be associated), so as to obtain a characteristic value for representing a correlation between the first data and the second data.

The determination of the characteristic value of the correlation between two media information (e.g. first data: media information X to be correlated with second data: other media information Y) is explained in detail below.

Optionally, the first preset policy is used for calculating various relationship states between the two pieces of media information. The first preset strategy specifically comprises: x ^ N Y, X-X ^ N Y, Y-X ^ N Y, N-X ^ U.

Thus, the dotted line shown in fig. 4 is that the server obtains the characteristic value for characterizing the correlation between X and Y by the first preset policy.

Similarly, the server can obtain the eigenvalues for representing the correlation between X and other M-2 pieces of media information according to the above process, so as to obtain 4 eigenvalues for representing the correlation between the first data and the second data.

And S104, obtaining a first correlation value for representing the correlation of the media information according to the characteristic value for representing the correlation between the first data and the second data and a second preset strategy.

After the server performs operation between at least two pieces of media information on the M pieces of effective data based on the specified media information categories according to a first preset strategy to obtain characteristic values for representing the correlation between the first data and the second data, because the second data are the effective data corresponding to the other M-1 pieces of media information respectively, and one piece of media information corresponds to one piece of effective data, the server obtains M-1 characteristic values for representing the correlation between the first data and the second data through the calculation, and then the server can obtain M-1 first correlation values for representing the correlation between the media information according to the characteristic values for representing the correlation between the first data and the second preset strategy respectively.

With the first embodiment of the present invention, as shown in fig. 5, a schematic diagram of interaction between a corresponding client and a server is shown, and a second preset policy for calculating a first relevance value is set in the server in fig. 5.

It should be noted that the second preset policy in the embodiment of the present invention is used to represent a relationship between the correlation between the media information to be associated and one other media information. The second preset strategy may be: equation (1).

Wherein, A is a first correlation value used for representing the correlation of the media information; p () is an existing association rule function.

Then, the server obtains M-1 first correlation values for representing the correlation of the media information according to M-1 characteristic values for representing the correlation between the first data and the second data and a second preset strategy, that is, the server obtains M-1 first correlation values between the media information to be correlated and other M-1 media information.

S105, according to the first correlation value used for representing the correlation of the media information, the media information correlated with the media information to be correlated is determined from the M-1 media information.

After the server obtains M-1 first correlation values for representing the correlation of the media information according to the characteristic value for representing the correlation between the first data and the second data and a second preset strategy, the server can determine the media information correlated with the media information to be correlated from the M-1 media information according to the M-1 first correlation values for representing the correlation of the media information, because the M-1 first correlation values are used for representing the correlation between the media information to be correlated and other M-1 media information.

Specifically, the server may rank the M-1 first relevance values, and select media information corresponding to a highest ranked first relevance value, where the selected media information is media information associated with media information to be associated.

It should be noted that the number of the media information selected by the server to be associated with the media information to be associated is determined by the association request, and a specific implementation process will be described in detail in the following embodiments.

It can be understood that, in the embodiment of the present invention, the higher the first relevance value is, the better the relevance between the media information corresponding to the first relevance value and the media information to be correlated is represented, so that the server may select, from the M-1 media information, at least one piece of media information with better relevance to the intra-body information to be correlated.

Example two

An embodiment of the present invention provides an information processing method, as shown in fig. 6, the method may include:

s201, when an association request generated based on user behavior is received, acquiring M pieces of media information according to media information types specified by the association request; m is a positive integer greater than 1.

S202, preprocessing the playing data of the M pieces of media information to obtain M pieces of effective data, wherein the M pieces of effective data comprise first data corresponding to one piece of media information to be associated in the M pieces of media information and second data corresponding to other M-1 pieces of media information.

Type,Videoid:Uin1,Uin2,Uin3...

S203, performing operation between at least two pieces of media information on the M effective data according to a first preset strategy based on the appointed media information category to obtain a characteristic value for representing the correlation between the first data and the second data.

S204, obtaining a first correlation value for representing the correlation of the media information according to the characteristic value for representing the correlation between the first data and the second data and a second preset strategy.

S205, sorting the first correlation values corresponding to the second data according to the numerical values.

After the server obtains M-1 first correlation values for representing the correlation of the media information according to the characteristic value for representing the correlation between the first data and the second data and a second preset strategy, the M-1 first correlation values obtained by the server are numerical values, and because the server in the embodiment of the invention wants to find the media information with better correlation with the media information to be correlated, the server selects a plurality of second data with higher first correlation values, thereby determining the media information (corresponding to the second data) with higher correlation with the media information to be correlated. Then, the server may sort the M-1 first correlation values according to the magnitude of the value, and obtain a sorting result.

It should be noted that the sorting rule in the embodiment of the present invention is not limited, and may be sorted according to decreasing numerical values, or sorted according to increasing numerical values.

Illustratively, assuming that M is 4, M media information is A, B, C and D respectively, where a is the media information to be associated, the server obtains a first correlation value between a and B of 0.8, a first correlation value between a and C of 0.5, and a first correlation value between a and D of 0.7 according to the above process, and the server performs sorting by increasing the numerical value to obtain sorting results of 0.8, 0.7, and 0.5. Wherein 0.8 may correspond to B, 0.7 may correspond to D, and 0.5 may correspond to C.

S206, according to the association request and the sequencing result, determining the media information associated with the media information to be associated from the M-1 media information.

Optionally, the association request may further include requesting data having a correlation with H first data, where H is a positive integer greater than or equal to 1 and less than M.

And the server sorts the first relevance values corresponding to the second data according to the numerical values to obtain a sorting result. Because the association request further includes the number H of the media information with high correlation with the media information to be associated, which needs to be determined by the server, the server can select the H media information corresponding to the highest H first correlation values from the M-1 media information according to the association request and the sorting result.

Specifically, the determining, by the server, the media information associated with the media information to be associated from the M-1 media information according to the association request and the sorting result may include: and the server determines H maximum first correlation values in the sequencing result according to the value of H and the sequencing result. And the server determines H pieces of media information corresponding to the H maximum first correlation values as the media information associated with the media information to be associated.

It should be noted that, since the M-1 first correlation values are used for representing the correlation between the media information to be correlated and other M-1 media information, the server may determine the media information correlated with the media information to be correlated from the M-1 media information according to the M-1 first correlation values used for representing the correlation of the media information.

Illustratively, assuming that M is 4, H in the association request is 2, M pieces of media information are A, B, C and D, respectively, where a is the media information to be associated, the server obtains a first correlation value between a and B of 0.8, a first correlation value between a and C of 0.5, and a first correlation value between a and D of 0.7 according to the above process, and the server performs sorting according to the numerical increment to obtain sorting results of 0.8, 0.7, and 0.5. Wherein 0.8 may correspond to B, 0.7 may correspond to D, and 0.5 may correspond to C. Since H is 2, the server selects the first two 0.8 and 0.7 of the ranking result, and thus, the server can determine the media information B corresponding to 0.8 and the media information D corresponding to 0.7 as the media information related to a.

Further, after the server determines the H pieces of media information with high correlation with the media information to be associated, the server issues the H pieces of media information to the client through the network, so that the client displays the H pieces of media information on the current playing page for the user to select and use.

EXAMPLE III

An embodiment of the present invention provides an information processing method, as shown in fig. 7, the method may include:

s301, when receiving a correlation request generated based on user behavior, acquiring M pieces of media information according to media information types specified by the correlation request; m is a positive integer greater than 1.

S302, the playing data of the M pieces of media information are preprocessed to obtain M pieces of effective data, and the M pieces of effective data comprise first data corresponding to one piece of media information to be associated in the M pieces of media information and second data corresponding to other M-1 pieces of media information.

Type,Videoid:Uin1,Uin2,Uin3...

S303, performing operation between at least two pieces of media information on the M effective data according to a first preset strategy based on the appointed media information category to obtain a characteristic value for representing the correlation between the first data and the second data.

S304, obtaining a first correlation value for representing the correlation of the media information according to the characteristic value for representing the correlation between the first data and the second data and a second preset strategy.

S305, according to a first correlation value used for representing the correlation of the media information, the media information correlated with the media information to be correlated is determined from the M-1 media information.

S306, the media information associated with the media information to be associated is issued.

After the server determines the media information associated with the media information to be associated from the M-1 media information according to the first relevance value for representing the relevance of the media information, the server can issue the determined media information associated with the media information to be associated to the client, so that the client can display the media information associated with the media information to be associated on a trigger interface of the media to be associated for the user to select and use.

It should be noted that the form of the media information sent by the server to the client in the embodiment of the present invention may be various forms such as a name, a picture, or a URL of the media information, and the specific sending form of the media information is determined by the form of the requested media information carried in the association request.

Example four

An embodiment of the present invention provides an information processing method, as shown in fig. 8, the method may include:

s401, when receiving a correlation request generated based on user behavior, acquiring M pieces of media information according to media information types specified by the correlation request; m is a positive integer greater than 1.

S402, preprocessing the playing data of the M pieces of media information to obtain M pieces of effective data, wherein the M pieces of effective data comprise first data corresponding to one piece of media information to be associated in the M pieces of media information and second data corresponding to other M-1 pieces of media information.

Type,Videoid:Uin1,Uin2,Uin3...

S403, performing operation between at least two pieces of media information on the M pieces of effective data according to a first preset strategy based on the appointed media information type to obtain a characteristic value for representing the correlation between the first data and the second data.

S404, obtaining a first correlation value for representing the correlation of the media information according to the characteristic value for representing the correlation between the first data and the second data and a second preset strategy.

S405, according to the first data and a third preset strategy, obtaining a second correlation value used for representing the correlation of the first data in the M effective data.

After the server obtains the first correlation value for characterizing the correlation of the media information according to the feature value for characterizing the correlation between the first data and the second preset policy, in order to more accurately characterize the correlation value of the media information to be correlated, the server further characterizes the correlation between the first data and the second data in M valid data, and specifically, for the first data, the server may obtain the second correlation value for characterizing the correlation of the first data in the M valid data according to the first data and the third preset policy.

It should be noted that the third preset policy in the embodiment of the present invention is used to characterize the relationship (i.e., correlation) of the first data among the M valid data.

Illustratively, the first data: and the media information X to be associated.

Optionally, the third preset policy may be: equation (2).

Wherein B is a second correlation value,

is M-X.

S406, obtaining a third correlation value of the correlation of each datum in the second datum in the M effective data according to the second datum and a fourth preset strategy.

After the server obtains the first correlation value for representing the correlation of the media information according to the characteristic value for representing the correlation between the first data and the second preset strategy, in order to more accurately describe the correlation value of the media information to be correlated, the server further describes the correlation between the first data and the second data in M effective data, and the server can calculate M-1 third correlation values because the second data corresponds to M-1 media information. Specifically, for the second data, the server may obtain, according to the second data and a fourth preset policy, M-1 third correlation values of the correlation of the data corresponding to each piece of media information in the second data among the M effective data.

It should be noted that the fourth preset policy in the embodiment of the present invention is used to characterize the relationship (i.e., correlation) of each data in the second data among the M valid data.

Illustratively, Y is any one of the second data corresponding to one of the other M-1 pieces of media information.

Optionally, the fourth preset policy for each piece of second data, namely the first data, may be: equation (3).

Wherein C is a third correlation value,

is M-Y.

It should be noted that in the information processing method provided in the embodiment of the present invention, S405 and S406 may be two optional steps after S404 and before S407, or two parallel steps after S404 and before S407. When S405 and S406 are optional steps after S404 and before S407, in this embodiment of the present invention, after S404 and before S407, it may be S405, or after S404 and before S407, it may be S406, specifically, which step the information processing method provided by this embodiment of the present invention includes is determined by actual setting of the server, and this embodiment of the present invention is not limited. When the information processing method provided by the embodiment of the present invention includes steps S405 and S406, steps S405 and S406 are parallel steps after S404 and before S407, and the embodiment of the present invention does not limit the execution sequence of steps S405 and S406.

S407, obtaining a total correlation value for representing the correlation of the media information according to the first correlation value, the second correlation value, the third correlation value and a fifth preset strategy.

After the server obtains the first, second, and third correlation degrees, the server may obtain a more accurate total correlation value (representing the total correlation value associated with the media information to be associated) for representing the correlation of the media information according to the first, second, and third correlation degrees and a fifth preset policy.

With the fourth embodiment of the present invention, as shown in fig. 9, a schematic diagram of interaction between a corresponding client and a server is shown, and a fifth preset policy for calculating a total correlation value is set in the server in fig. 9.

It should be noted that, because the information processing method provided in the embodiment of the present invention may include the second correlation value, the third correlation value, or the second correlation value and the third correlation value, the fifth preset policy in S407 is different for the three different cases.

Specifically, when the information processing method provided in the embodiment of the present invention includes only the second correlation value, the fifth preset policy is: equation (4).

sim＝A/B (4)

Where sim is the total correlation value, A is the first correlation value, and B is the second correlation value.

When the information processing method provided by the embodiment of the present invention includes only the third correlation value, the fifth preset policy is: equation (5).

sim＝A/C (5)

Where sim is the total correlation value, A is the first correlation value, and C is the third correlation value.

When the information processing method provided by the embodiment of the present invention includes the second correlation value and the third correlation value at the same time, the fifth preset policy is: comparing the sizes of B and C, if B > C, the fifth preset strategy is as follows: formula (5); if C is greater than B, the fifth preset strategy is: equation (4).

Wherein sim is the total correlation value, a is the first correlation value, B is the second correlation value, and C is the third correlation value.

It should be noted that, no matter which way the above is adopted, the server can derive the total correlation value for characterizing the correlation of the media information.

S408, according to the relevance value for representing the relevance of the media information, the media information relevant to the media information to be relevant is determined from the M-1 media information.

After the server obtains the total correlation values for representing the correlation of the media information, since the second data are the valid data corresponding to the other M-1 pieces of media information respectively, and one piece of media information corresponds to one piece of valid data, the total correlation values for representing the correlation of the media information obtained by the server through the above calculation are M-1 pieces, and thus, the server can obtain the M-1 total correlation values for representing the correlation of the media information.

It should be noted that the M-1 total correlation values obtained by the server are numerical values, and since the server in the embodiment of the present invention wants to find the media information with better correlation with the media information to be correlated, the server may select some of the second data with higher total correlation values, so as to determine the media information (corresponding to the second data) with higher correlation with the media information to be correlated.

Specifically, the server may rank the M-1 total correlation values according to the numerical value (that is, the server ranks the total correlation values corresponding to the second data according to the numerical value), and obtains a ranking result, and the server may determine, according to the association request and the ranking result, media information associated with the media information to be associated from the M-1 media information.

It should be noted that, after the server sorts the total correlation values corresponding to the second data according to the numerical values, a sorting result is obtained. Because the association request further includes the number H of the media information with high correlation with the media information to be associated, which needs to be determined by the server, the server can select H media information corresponding to the highest H total correlation values from the M-1 media information according to the association request and the sorting result.

Specifically, the determining, by the server, the media information associated with the media information to be associated from the M-1 media information according to the association request and the sorting result may include: and the server determines H maximum total correlation values in the sequencing result according to the value of H and the sequencing result. And the server determines H pieces of media information corresponding to the H maximum total correlation values as the media information associated with the media information to be associated.

It should be noted that, since the M-1 total correlation values are used for representing the correlation between the media information to be correlated and the other M-1 media information, the server can determine the media information correlated with the media information to be correlated from the M-1 media information according to the M-1 total correlation values used for representing the correlation of the media information.

It can be understood that, the higher the total correlation value in the embodiment of the present invention is, the better the relevance between the media information corresponding to the total correlation value and the media information to be correlated is represented, so that the server may select at least one piece of media information with better relevance to the intra-body information to be correlated from the M-1 pieces of media information.

Illustratively, assuming that M is 4, H in the association request is 2, M pieces of media information are A, B, C and D, respectively, where a is the media information to be associated, the total correlation value of a and B obtained by the server according to the above process is 0.9, the total correlation value of a and C is 0.6, and the total correlation value of a and D is 0.8, the server performs sorting according to the numerical increment, and the sorting results of 0.9, 0.8, and 0.6 are obtained. Wherein 0.9 may correspond to B, 0.8 may correspond to D, and 0.6 may correspond to C. Since H is 2, the server selects the first two 0.9 and 0.8 of the ranking result, and thus, the server can determine that the media information B corresponding to 0.9 and the media information D corresponding to 0.8 are the media information related to a.

And S409, issuing the media information associated with the media information to be associated.

After the server determines the media information associated with the media information to be associated from the M-1 media information according to the total relevance value for representing the relevance of the media information, the server can issue the determined media information associated with the media information to be associated to the client, so that the client can display the media information associated with the media information to be associated on a trigger interface of the media to be associated for the user to select and use.

EXAMPLE five

As shown in fig. 10, an embodiment of the present invention provides a server 1, where the server 1 may include:

an obtaining unit 10, configured to, when the receiving unit 11 receives an association request generated based on a user behavior, obtain M pieces of media information according to a media information category specified by the association request; and M is a positive integer greater than 1.

A calculating unit 12, configured to pre-process the playing data of the M pieces of media information acquired by the acquiring unit 10 to obtain M pieces of valid data, where the M pieces of valid data include first data corresponding to one piece of media information to be associated in the M pieces of media information and second data corresponding to other M-1 pieces of media information; performing operation between at least two pieces of media information on the M effective data according to a first preset strategy based on the appointed media information type to obtain a characteristic value for representing the correlation between the first data and the second data; and obtaining a first correlation value for representing the correlation of the media information according to the characteristic value for representing the correlation between the first data and the second data and a second preset strategy.

A determining unit 13, configured to determine, according to the first correlation value for characterizing the correlation of media information calculated by the calculating unit 12, media information associated with the media information to be associated from the M-1 pieces of media information acquired by the acquiring unit 10.

Optionally, the calculating unit 12 is further configured to obtain, after obtaining a first correlation value used for characterizing correlation of media information according to the feature value used for characterizing correlation between the first data and the second data and a second preset policy, a second correlation value used for characterizing correlation of the first data in the M valid data according to the first data and a third preset policy acquired by the acquiring unit 10; and/or obtaining a third correlation value of the correlation of each data in the second data in the M effective data according to the second data acquired by the acquisition unit 10 and a fourth preset strategy; and obtaining a total correlation value for representing the correlation of the media information according to the first correlation value, the second correlation value, the third correlation value and a fifth preset strategy.

Optionally, as shown in fig. 11, the server 1 further includes: a sorting unit 14.

The sorting unit 14 is configured to sort the first correlation values corresponding to the second data calculated by the calculating unit 12 according to the magnitude of the first correlation values.

The determining unit 13 is specifically configured to determine, according to the association request received by the receiving unit 11 and the sorting result obtained by the sorting unit 14, media information associated with the media information to be associated from the M-1 media information acquired by the acquiring unit 10.

Optionally, the association request received by the receiving unit 11 includes a request for data having correlation with H first data, where H is a positive integer greater than or equal to 1 and less than M.

The determining unit 13 is further configured to determine, according to the value of H received by the receiving unit 11 and the sorting result obtained by the sorting unit 14, H maximum first correlation values calculated by the calculating unit 12 in the sorting result.

The determining unit 13 is further specifically configured to determine the H pieces of media information corresponding to the H largest correlation values as the media information associated with the media information to be associated.

Optionally, as shown in fig. 12, the server 1 further includes: a transmitting unit 15.

The sending unit 15 is configured to, after the determining unit 13 determines, according to a first relevance value used for representing relevance of media information, media information associated with the media information to be associated from the M-1 pieces of media information, issue the media information associated with the media information to be associated, which is determined by the determining unit 13.

Optionally, the M pieces of media information acquired by the acquiring unit 10 include one piece of media information to be associated and other M-1 pieces of media information, and the piece of media information to be associated and the other M-1 pieces of media information belong to the same category.

The M effective data calculated by the calculating unit 12 are effective data based on user behavior obtained after data preprocessing and filtering.

As shown in fig. 13, in practical applications, the obtaining unit 10, the calculating unit 12, the determining unit 13, and the sorting unit 14 may be implemented by a processor 16 located on a server, specifically, a Central Processing Unit (CPU), a Microprocessor (MPU), a Digital Signal Processor (DSP), a Field Programmable Gate Array (FPGA), or the like, the transmitting unit 15 may be implemented by a transmitter 17, the receiving unit 11 may be implemented by a receiver 18, the receiver 18 and the transmitter 17 may be implemented by a receiving all-in-one machine, the server further includes a memory 19, specifically, the first data and its software code, the first correlation value and its software code, the second correlation value and its software code, the third correlation value and its software code, the total correlation value and its software code, and the feature value and its software code may be stored in the memory 19, the memory 19, transmitter 17, receiver 18 may be coupled to the processor 16, wherein the memory 19 is configured to store executable program code comprising computer operating instructions, and the memory 19 may comprise a high speed RAM memory and may further comprise a non-volatile memory, such as at least one disk memory.

It should be noted that the server 1 in the embodiment of the present invention and the server 22 in fig. 2 refer to the same server.

As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of a hardware embodiment, a software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, optical storage, and the like) having computer-usable program code embodied therein.

The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

The above description is only a preferred embodiment of the present invention, and is not intended to limit the scope of the present invention.

Claims

1. An information processing method characterized by comprising:

wherein the first preset strategy is to calculate an intersection of the first data and the second data, data except the intersection in the first data, data except the intersection in the second data, and data except the intersection in the M valid data;

2. The method according to claim 1, wherein after obtaining the first correlation value for characterizing the correlation of the media information according to the feature + value for characterizing the correlation between the first data and the second predetermined policy, the method further comprises:

3. The method according to claim 1, wherein the determining media information associated with the media information to be associated from the M-1 media information according to the first relevance value for characterizing relevance of the media information comprises:

4. The method according to claim 3, wherein the association request includes data that requests H number of the first data to have correlation, where H is a positive integer greater than or equal to 1 and less than M, and the determining media information associated with the media information to be associated from the M-1 number of media information according to the association request and the sorting result includes:

5. The method according to claim 1, wherein after determining the media information associated with the media information to be associated from the M-1 media information according to the first relevance value for characterizing relevance of the media information, the method further comprises:

6. The method of claim 1,

the M pieces of media information comprise one piece of media information to be associated and other M-1 pieces of media information, and the piece of media information to be associated and the other M-1 pieces of media information belong to the same category;

7. A server, comprising:

8. The server according to claim 7,

the calculating unit is further configured to obtain a first correlation value used for characterizing correlation of media information according to the feature value used for characterizing correlation between the first data and the second data and a second preset policy, and then obtain a second correlation value used for characterizing correlation of the first data in the M effective data according to the first data and a third preset policy acquired by the acquiring unit; and/or obtaining a third correlation value of the correlation of each data in the second data in the M effective data according to the second data acquired by the acquisition unit and a fourth preset strategy; and obtaining a total correlation value for representing the correlation of the media information according to the first correlation value, the second correlation value, the third correlation value and a fifth preset strategy.

9. The server of claim 7, further comprising: a sorting unit;

10. The server according to claim 9, wherein the association request includes a request for data having a correlation with H first data, where H is a positive integer greater than or equal to 1 and less than M;

11. The server of claim 7, further comprising: a transmitting unit;

12. The server according to claim 7,

the M pieces of media information acquired by the acquisition unit comprise one piece of media information to be associated and other M-1 pieces of media information, and the piece of media information to be associated and the other M-1 pieces of media information belong to the same category;

13. A computer-readable storage medium having stored thereon executable instructions for causing a processor, when executed, to implement the method of any one of claims 1-6.