CN116137677A

CN116137677A - Theme recommendation method of intelligent television, computer equipment and readable storage medium

Info

Publication number: CN116137677A
Application number: CN202310141780.0A
Authority: CN
Inventors: 李鸣; 肖云; 曾泽基; 郁延书
Original assignee: Future Tv Co ltd
Current assignee: Future Tv Co ltd
Priority date: 2023-02-17
Filing date: 2023-02-17
Publication date: 2023-05-19

Abstract

The application provides a theme recommendation method of an intelligent television, computer equipment and a readable storage medium, and belongs to the technical field of intelligent televisions. The method comprises the following steps: acquiring historical behavior information of a target user, wherein the historical behavior information comprises: searching behavior information, playing behavior information and clicking behavior information; determining the score of the target user for each tag according to the historical behavior data; determining the score of the target user for each topic according to the score of the target user for each label; and pushing the theme to the target user according to the score of the target user for each theme. The method and the device can achieve personalized theme recommendation effects of thousands of people and thousands of faces.

Description

Theme recommendation method of intelligent television, computer equipment and readable storage medium

Technical Field

The application relates to the technical field of intelligent televisions, in particular to a theme recommendation method, computer equipment and a readable storage medium of an intelligent television.

Background

In recent years, with The popularization of The internet, internet television (Top, OOT) has been increasingly used. Internet television refers to a method for transmitting multimedia audio-visual services including video, audio, graphics, text, data and the like based on the public internet, and providing the public with safe, reliable, interactive and manageable multimedia audio-visual services through terminal receiving devices such as televisions, set-top boxes and the like. The intelligent television belongs to an internet television and can provide rich functions for users. Theme recommendation is an important function of the smart television, so how to perform theme recommendation is a problem worthy of research.

In the prior art, the topic recommendation mode of the intelligent television mainly comprises the steps of generating a recommendation topic through operation experience of operators and current popular trends, manually screening out program composition topics conforming to core topics based on the recommendation topic, and finally sending topic configuration to the intelligent television to be presented to a user.

However, in the method for recommending the smart tv theme based on the related art, the definition of the recommended theme is performed according to the operation experience and subjective knowledge of the operator, and this theme recommending method is time-consuming and laborious, and the personal preference of the operator may also interfere with the judgment result. And secondly, the topic programs and topic arrangements recommended by the intelligent television topic recommendation method are different in size, so that the user experience is poor, and the recommendation efficiency is low. Therefore, the related technology has the problems of uniform recommendation effect, poor user experience and low recommendation efficiency.

Disclosure of Invention

The patent refers to the field of 'electric digital data processing'.

In a first aspect of the embodiments of the present application, a method for recommending a theme of an intelligent television is provided, where the method for recommending a theme of an intelligent television includes: acquiring historical behavior information of a target user, wherein the historical behavior information comprises: searching behavior information, playing behavior information and clicking behavior information; determining the score of the target user for each tag according to the historical behavior data; determining the score of the target user for each topic according to the score of the target user for each label; and pushing the theme to the target user according to the score of the target user for each theme.

As a possible implementation manner, the determining the score of the target user for each tag according to the historical behavior data includes:

according to the historical behavior data, determining the times of each behavior of the target user for each tag and the times of each behavior of the target user for all videos;

and determining the score of the target user for each tag according to the behavior weight of each behavior, the number of times each behavior occurs to the target user for each tag and the number of times each behavior occurs to the target user for all videos.

As one possible implementation manner, the determining the score of the target user for each tag according to the action weight of each action, the number of times the target user takes each action for each tag, and the number of times the target user takes each action for all videos includes:

determining a first ratio of the number of times that the target user generates a first behavior aiming at a first tag in different periods to the number of times that the target user generates the first behavior aiming at all videos in different periods, wherein the first tag is any tag, and the first behavior is any historical behavior of the target user;

Determining a product of the behavior weight of the first behavior and the first ratio to obtain a score of a combination of the first tag and the first behavior;

and determining the score of the target user for the first tag according to the score of the first tag and the combination of the behaviors.

As a possible implementation manner, the determining the score of the target user for each topic according to the score of the target user for each label includes:

sorting the scores of the target users for various labels, and obtaining preference labels of the target users according to sorting results;

determining the number of matching of the preference label of the target user and the topic labels of various topics on the preference label;

and determining the score of the target user for each topic according to the tag matching weight of the preference tag, the number of matching of the preference tag of the target user and topic tags of various topics on the preference tag and the K value of the preference tag.

As one possible implementation manner, the determining the score of the target user for each topic according to the tag matching weight of the preference tag, the number of matching of the preference tag of the target user with the topic tags of various topics on the preference tag, and the K value of the preference tag includes:

Determining a second ratio of the number of matching of a first preference tag of the target user and a topic tag of a first topic on the first preference tag to the K value of the first preference tag, wherein the first preference tag is any preference tag of the target user, and the first topic is any topic;

determining the product of the tag matching weight of the first preference tag and the second ratio to obtain the score of the first theme on the first preference tag;

and determining the score of the target user for the first theme according to the scores of the first theme on the preference labels.

As a possible implementation manner, the pushing the theme to the target user according to the score of each theme of the target user includes:

sorting the scores of the target users for various topics, and obtaining a preference topic sequence of the target users according to the sorting result, wherein the preference topic sequence comprises preference topics which are sequentially arranged;

determining an initial video set of each preference topic according to topic tags of each preference topic;

determining the score of each video in the initial video set of each preference theme according to the historical behavior data of the target user;

Determining a target video set of each preference theme according to the score of each video in the initial video set of each preference theme, wherein the number of videos in the target video set is smaller than that in the initial video set;

determining the score of each preference topic according to the score of each video in the target video set of each preference topic;

sorting the preference topics according to the scores of the preference topics, and determining a target topic sequence according to the sorting result;

pushing the target subject sequence to the target user.

As a possible implementation manner, the determining, according to the historical behavior data of the target user, the score of each video in the initial video set of each preference theme includes:

according to the historical behavior data, determining the times of each behavior of the target user for each video and the times of each behavior of the target user for all videos;

and determining the score of the target user for each video according to the behavior weight of each behavior, the times of each behavior of the target user for each video and the times of each behavior of the target user for all videos.

As one possible implementation manner, the determining the score of the target user for each video according to the behavior weight of each behavior, the number of times the target user takes each behavior for each video, and the number of times the target user takes each behavior for all videos includes:

determining a product of the behavior weight of the first behavior and the third ratio to obtain a score of a combination of the first video and the first behavior;

and determining the score of the target user for the first video according to the score of the first video and the combination of the behaviors.

In a second aspect of the embodiments of the present application, a theme recommendation apparatus for an intelligent television is provided, where the theme recommendation apparatus for an intelligent television includes: the acquisition module is used for acquiring historical behavior information of the target user, wherein the historical behavior information comprises: searching behavior information, playing behavior information and clicking behavior information; the first determining module is used for determining the score of the target user for each label according to the historical behavior data; the second determining module is used for determining the score of the target user for each theme according to the score of the target user for each label; and the pushing module is used for pushing the topics to the target user according to the scores of the target user for each topic.

In a third aspect of the embodiments of the present application, there is provided a computer device, where the computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, where the computer program is executed by the processor to implement the steps of the subject recommendation method of the smart television set described in the first aspect.

In a fourth aspect of the embodiments of the present application, there is provided a computer readable storage medium storing a computer program, where the computer program is executed by a processor to implement the steps of the method for recommending a theme of a smart tv according to the first aspect.

The beneficial effects of the embodiment of the application include: after the historical behavior information of the target user is obtained, the score of the target user for each tag can be determined according to the historical behavior information, and the score of the target user for each topic can be determined according to the score of the target user for each tag, so that topic recommendation can be performed to the target user according to the score of the target user for each topic. According to the historical behavior information, the preference of the target user for each tag can be determined, and the preference of the target user for the theme is identified according to the preference, so that the theme which is most matched with the user can be recommended to different users, the personalized requirements of the users are met, and the user experience is greatly improved. Meanwhile, accurate theme pushing can be achieved without any manual operation by operators, so that pushing efficiency can be greatly improved.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the embodiments will be briefly described below, it being understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered limiting the scope, and that other related drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

Fig. 1 is an application scenario schematic diagram of a theme recommendation method of an intelligent television provided in an embodiment of the present application;

fig. 2 is a flowchart of a theme recommendation method of a smart television provided in an embodiment of the present application;

fig. 3 is a label scoring flowchart of a topic recommendation method of an intelligent television provided in an embodiment of the present application;

fig. 4 is a flowchart of a first tag score of a topic recommendation method of a smart tv according to an embodiment of the present application;

fig. 5 is a topic score flowchart of a topic recommendation method of an intelligent television provided in an embodiment of the present application;

fig. 6 is a first topic score flowchart of a topic recommendation method of a smart tv provided in an embodiment of the present application;

fig. 7 is a theme pushing flowchart of a theme recommendation method of an intelligent television provided in an embodiment of the present application;

Fig. 8 is a video scoring flowchart of a topic recommendation method of a smart tv provided in an embodiment of the present application;

fig. 9 is a first video scoring flowchart of a topic recommendation method of a smart tv provided in an embodiment of the present application;

fig. 10 is a schematic structural diagram of a device for recommending a theme of an intelligent television according to an embodiment of the present application;

fig. 11 is a schematic structural diagram of a computer device according to an embodiment of the present application.

Detailed Description

For the purposes of making the objects, technical solutions and advantages of the embodiments of the present application more clear, the technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is apparent that the described embodiments are some embodiments of the present application, but not all embodiments. The components of the embodiments of the present application, which are generally described and illustrated in the figures herein, may be arranged and designed in a wide variety of different configurations.

Thus, the following detailed description of the embodiments of the present application, as provided in the accompanying drawings, is not intended to limit the scope of the application, as claimed, but is merely representative of selected embodiments of the application. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments herein without making any inventive effort, are intended to be within the scope of the present application.

At present, a theme recommendation mode of the intelligent television mainly depends on the operation experience of background operators and the current popular trend to generate a recommendation theme, then programs meeting a core theme are manually screened out based on the recommendation theme to form a theme, and finally the theme configuration is sent to the intelligent television to be presented to a user. However, the method for recommending the intelligent television theme is defined by recommending the theme according to the operation experience and subjective cognition of the background operator, and the theme recommending method is time-consuming and labor-consuming, and personal preference of the operator can also interfere with the judgment result. And secondly, the topic programs and topic arrangements recommended by the intelligent television topic recommendation method are different in size, so that the user experience is poor, and the recommendation efficiency is low. Therefore, the results of the recommended topics are uniformly distributed, the user experience is poor, and the recommendation efficiency is low.

Therefore, the embodiment of the application provides a theme recommendation method of an intelligent television, which is used for identifying preference labels of users based on historical behavior information of the users, determining the score of each theme by the users based on the preference labels, and accurately recommending the theme to the users according to the score of each theme, so that the theme which is most matched with the users is recommended to different users, and personalized requirements of the users are met. Meanwhile, pushing efficiency can be greatly improved.

Fig. 1 is an application scenario schematic diagram of a theme recommendation method of an intelligent television 10 provided in an embodiment of the present application, where the application scenario relates to a plurality of intelligent televisions 10 installed on a user side, and a cloud server 20 communicatively connected to the intelligent televisions 10.

As can be seen from fig. 1, the cloud server 20 may serve multiple smart televisions 10 at the same time, and the smart televisions 10 serve as terminal receiving devices of users to display the results of theme recommendation. The topic recommendation method can be executed on the cloud server 20, the cloud server 20 is accessed through an internet network, the cloud server 20 is delivered and used on the basis of information technology (Information Technology, abbreviated as IT) by utilizing a cloud computing technology, topic resources can be obtained through the network according to the needs of users, the topic resources can be expanded infinitely, and topic resources required by the users can be obtained at any time. Meanwhile, the cloud server 20 utilizes server virtualization to centrally manage a central processing unit (Central Processing Unit, abbreviated as CPU), a memory, a hard disk and the like, so that the theme resources are dynamically allocated according to the needs, and the utilization efficiency of the theme resources can be effectively improved. Therefore, the cloud server 20 can push the theme most suitable for the user needs to the smart television 10 for display according to the characteristics of the end user based on the method of the embodiment of the present application, and brings good use experience to the user.

Fig. 2 is a flow chart of a method for recommending a theme of an intelligent television according to an embodiment of the present application, and an execution subject of the method may be the cloud server 20 described above. As shown in fig. 2, the method includes:

s201, acquiring historical behavior information of a target user, wherein the historical behavior information comprises: search behavior information, play behavior information, and click behavior information.

Optionally, when the target user uses the smart television, interaction behavior between the user and the smart television occurs, that is, historical behavior data with corresponding attributes are generated along with the interaction behavior, and the smart television 10 stores the historical behavior data into the service database. Further, the historical behavior information of the target user is obtained by reading the historical behavior data from the business database of the target user.

It should be noted that the above-mentioned historical behavior information may be historical behavior information within a preset period of time, and the preset period of time may be, for example, the last 1 day, the last 7 days, the last 15 days, the last 30 days, and the like.

Optionally, the historical behavior information of the target user includes search behavior information, play behavior information and click behavior information. The search behavior information may include searched video content, searched video age, searched video language, searched video director, searched video time, etc.; the play behavior information may include the content of the video being played, the type of video being played, the language of the video being played, the director of the video being played, the time of the play, etc.; click behavior information may include: click video content, click video language, video type of search, click video director, click video time, etc.; but is not limited thereto.

S202, determining the score of the target user for each label according to the historical behavior data.

Alternatively, a plurality of tags may be preset, and a corresponding tag may be configured for each theme and each video. By way of example, the tag may include a primary category, a secondary category, a lead actor, a director, etc., wherein the primary category may include movies, television shows, etc., and the secondary category may include comedy, love, drama, etc.

Taking video as an example, when the smart television uploads a new video, a user can manually configure a corresponding tag for the video. It should be noted that the tags of each theme and video may be one or more.

Optionally, a score of the target user for each tag may be determined based on historical behavior data of the target user.

S203, determining the score of the target user for each topic according to the score of the target user for each label.

Alternatively, multiple topics may be preset, and these topics are stored in a topic library, and a corresponding tag is configured for each topic. By way of example, the subject matter may include humans, history, nature, and the like.

Optionally, the score of the target user for each topic is further determined according to the score of the target user for each tag.

S204, according to the score of each theme of the target user, theme pushing is carried out on the target user.

Optionally, the topic pushing is to set tags for videos with historical behaviors of the target user based on the historical behavior information of the target user, so that each tag score can be determined. And then, according to the score of the target user for each label, the score of the target user for each theme can be determined, the theme pushing result is determined through the theme score, and the theme pushing result is pushed to the target user through the pushing interface of the intelligent television.

For example, when the historical behavior information of the target user mostly occurs in a movie, a television series or documentaries with a personal topic, the score of the target client for the tags of the movie, the television series or the documentaries can be determined, and further, the score of the target user for the personal topic can be determined. Therefore, the method can obtain that the pushing of the target user is a film, a television play, a documentary and the like related to the human text, and the theme pushing is realized through the pushing interface of the intelligent television. It should be noted that the theme pushed may be one theme or multiple themes.

In this embodiment, after the historical behavior information of the target user is obtained, the score of the target user for each tag may be determined according to the historical behavior information, and the score of the target user for each topic may be determined according to the score of the target user for each tag, so that topic recommendation may be performed to the target user according to the score of the target user for each topic. According to the historical behavior information, the preference of the target user for each tag can be determined, and the preference of the target user for the theme is identified according to the preference, so that the theme which is most matched with the user can be recommended to different users, the personalized requirements of the users are met, and the user experience is greatly improved. Meanwhile, accurate theme pushing can be achieved without any manual operation by operators, so that pushing efficiency can be greatly improved.

Fig. 3 is a flowchart of each tag score of the topic recommendation method of the smart tv according to the embodiment of the present application, as shown in fig. 3, the step S202 may include:

s2021, according to the historical behavior data, determining the times of each behavior of the target user for each tag and the times of each behavior of the target user for all videos.

Optionally, the historical behavior data of the target user may be analyzed to obtain a tag classification of the video in which the target user has historical behavior. Meanwhile, the historical behavior information of the target user is acquired from the business database of the target user, so that the times of each type of historical behavior information of the target user on each label can be obtained, and the times of each type of historical behavior of the target client on the video with the historical behavior can also be obtained.

Taking the historical behavior of the target user of nearly 1 day as an example, the target user searches 5 times for comedy movies and 3 times for loving TV shows, plays 1 time each of comedy movies and loving TV shows, and clicks 3 times each of comedy movies and loving TV shows. The search behavior of the target user on the first-level tag movie can be obtained for 5 times, the play behavior is 1 time, the click behavior is 3 times, and the like. The method can also obtain 5 times of searching behavior, 1 time of playing behavior and 3 times of clicking behavior of the comedy video, 3 times of searching behavior, 1 time of playing behavior and 3 times of clicking behavior of the love video of the target user.

Optionally, through historical behavior information of the target user, the number of times each behavior of the target user occurs for each tag can be determined, and the number of times each behavior of the target user occurs for each video can also be determined.

S2022, determining the score of the target user for each tag according to the behavior weight of each behavior, the number of times each behavior occurs for each tag by the target user and the number of times each behavior occurs for all videos by the target user.

Optionally, a corresponding behavior weight may be preset for each type of historical behavior information of the target client, where different behavior information reflects different interest preference degrees of the target user, for example, the interest preference of the target user is reflected by the search behavior and the play behavior compared with the click behavior, and then the interest preference of the target user is reflected by the play behavior compared with the search behavior, so that in order to better attach to the interest preference of the user, the behavior weight is preset according to the degree of reflecting the interest preference of the target user, and the higher the reflection degree, the higher the weight should be set.

Taking the historical behavior example of the target user for 1 day, presetting the behavior weight of the playing behavior as 100, the behavior weight of the searching behavior as 80 and the behavior weight of the clicking behavior as 60; the target user searches for comedy videos 5 times, plays 1 time and clicks 3 times, searches for love videos 3 times, plays 1 time and clicks 3 times. According to the behavior weight of the target user, the score of the target user for the movie tag can be obtained by combining the times of each behavior of the target user for the movie tag.

Optionally, the number of times that each behavior occurs to each tag by the target user and the number of times that each behavior occurs to all videos by the target user are determined, and the score of each tag by the target user is determined by combining the behavior weights corresponding to each behavior, so that the accuracy of the score can be ensured.

Hereinafter, a process of determining the score of the target user for each tag will be described.

Fig. 4 is a flowchart of a first tag score of a topic recommendation method for a smart tv according to an embodiment of the present application, as shown in fig. 4, a process for determining a score of a target user for each tag according to a behavior weight of each behavior, a number of times each behavior occurs for each tag by the target user, and a number of times each behavior occurs for all videos by the target user may include:

s401, determining a first ratio of the times of first behaviors of the target user aiming at the first labels in different periods to the times of first behaviors of the target user aiming at all videos in different periods, wherein the first labels are any labels, and the first behaviors are any historical behaviors of the target user.

S402, determining a product of the behavior weight of the first behavior and the first ratio value to obtain a score of a combination of the first tag and the first behavior.

S403, determining the score of the target user for the first label according to the score of the first label and the combination of the behaviors.

Alternatively, the score of the target user for each tag may be obtained by the following formula (1). Specifically, the following formula (1) is called for each label, and the score of the label can be obtained.

Score _uj ＝∑ _tεT ∑ _a∈A w _a *cnt _t j _a /all_cnt _ta (1)

Wherein Score _uj Representing the score of target user u for tag j, A represents the behavior set of target user u, w _a Representing the behavior weight of the behavior type a of the target user u, T represents the use of different periodic dimensions, and T represents any periodic dimension belonging to T. cnt _tia When the cycle dimension is taken as t, the target user u generates behavior times with a behavior type of a to the tag j in the cycle circumference t; all_cnt _ta And when the period dimension is taken as t, the target user u generates the behavior times of which the behavior type is a for all videos in the period circumference t.

Wherein the label j is taken from the first class [ TV drama, movie, etc. ], the second class [ comedy, love, scenario, etc. ], and the directorThe method comprises the steps of obtaining any one of a plurality of labels of a target user on multiple dimensions such as a primary label, a secondary label, a director, a year, a region, a language and the like. The behavior set A comprises search behavior, click behavior, play behavior and the like, wherein a is any one of behavior types of the search behavior, the click behavior, the play behavior and the like; w (w) _a The behavior weight of the behavior type a representing the occurrence of the target user u is different because the different behavior types reflect the interest preference degree of the target user, such as the search behavior and the playing behavior are stronger than the interest preference of the target user reflected by the clicking behavior, and the playing behavior is stronger than the interest preference of the target user reflected by the search behavior, so that the behavior weight is preset according to the degree reflecting the interest preference of the target user in order to better fit the interest preference of the user, the higher the reflecting degree is, the higher the weight is supposed to be set, and the historical behavior data of the target user u is required to be combined for setting; t represents different cycle dimensions, and in order to comprehensively obtain a final recommendation result by comprehensively analyzing long-term behavior data and short-term behavior data of the target user u, for example, four different cycle dimensions of the last 1 day, the last 7 days, the last 15 days and the last 30 days can be adopted.

Since the action weight participates in the calculation of the score of the target user u for each tag, if the target user u performs actions with higher weight on the video a multiple times, for example, plays the video a or searches the video a multiple times, other videos a 'similar to the video a are given higher recommendation weights, and the video a' is more recommended.

Examples are as follows: the system presets that the behavior type a is the behavior weight of the playing behavior as 100, when the target user u is in the condition that j is a comedy label, the number of times of playing the behavior a of the target user u in comedy videos is 100 times in nearly 30 days, and the number of times of playing the behavior a of the target user u in all videos is 150 times in nearly 30 days; the number of times of playing behavior a of the comedy video by the target user u in the near 15 days is 60, and the number of times of playing behavior a of the comedy video by the target user u in the near 15 days is 90. Knowing the behavior of the playing behavior a from the playing behavior information of the target userThe weight is 100, then in the period dimension of nearly 30 days, the frequency of comedy videos with playing behaviors is 100 times, the frequency of all videos with playing behaviors is 150, a first ratio of 100/150 can be obtained, and in the period dimension, the relation between the behavior weight of the playing behaviors and the video occurrence behaviors in the period dimension can be obtained by the formula (1), a first combined score of 100 x (100/150) can be obtained; similarly, in the period dimension of 15 days, the frequency of playing comedy videos is 60 times, the frequency of playing all videos is 90, the ratio of 60/90 can be obtained, and in the period dimension, the relation between the action weight of the playing action and the video occurrence in the period dimension can be obtained by the formula (1), 100 (60/90); finally, the preference Score of the target user u in the label j of about 30 days and about 15 days is obtained as Score _ui =100×100/150) +100×60/90=133.33 (taking the two decimal places).

In this embodiment, different weights are set for different rows of data of the target user in a targeted manner, and the preference of the target user is finally determined mainly according to the degree of interest preference response of the behavior weights to the target user and by comprehensively analyzing long-term behavior data and short-term behavior data of the target user. Therefore, the recommendation result is more comprehensive and is closer to the requirements of target clients, and the user satisfaction is improved.

Fig. 5 is a topic score flowchart of a topic recommendation method of a smart tv according to an embodiment of the present application, where, as shown in fig. 5, the step S203 may include:

s2031, sorting the scores of the target users for various labels, and obtaining the preference labels of the target users according to the sorting result.

Optionally, the scores of the target users for each label are sequentially arranged from high to low, the ordered result can well reflect the preference of the target users, and the preference labels of the target users are fetched from top to bottom according to the ordered result.

Taking the example of playing videos of the smart television in approximately 7 days, the behavior weight of the playing behavior is preset to be 100, and 100 videos are played by the target user in 7 days, wherein comedy 30, love 20 and cartoon 50. The score of the label available according to the above formula (1) is: comedy 30 minutes, love 20 minutes, cartoon 50 minutes, and the arrangement sequence is as follows: cartoon 50 points > comedy 30 points > love 20 points, and the preference label of the user is cartoon, or cartoon and comedy. It should be noted that the preference label of the target user may be one or more.

Optionally, the preference label of the target user can be determined according to the score sorting result of the target user for each label, and the preference distribution of the target can be identified.

S2032, determining the number of matching preference labels of the target user with topic labels of various topics on the preference labels.

Optionally, a plurality of theme labels may be preset first and stored in the theme library. The topic library information comprises topic id, topic name, topic label and the like, wherein the topic label comprises a plurality of labels of each dimension such as primary classification, secondary classification, director, year, region, language and the like. The theme tag may also be refined, where the primary category tag includes tv dramas, movies, etc., and the secondary category may include comedy, love, drama, etc.

Taking the preference label of the target user as a cartoon movie as an example, 5 topics with the topic label being the cartoon label are screened out from a topic library according to the preference label of the cartoon; and screening 3 theme labels from the theme set of the movies according to the other preference label movies of the target user, wherein the matching number of the theme labels conforming to the preference labels of the target user is 3. It is worth to say that the preference labels are screened in no sequence and can be adjusted at will.

Optionally, the matching number of the theme labels conforming to the preference labels of the target user can be finally obtained by matching the preference labels of the target user with the theme labels of the theme library.

S2033, determining the score of the target user for each topic according to the tag matching weight of the preference tag, the matching quantity of the preference tag of the target user and the topic tags of various topics on the preference tag and the K value of the preference tag.

Optionally, a corresponding tag matching weight may be preset for the preference tag of the target, and the preference degree of the preference tag may be reflected by the tag score according to the preference tag obtained by the tag score sorting result. Accordingly, different tag matching weights may be set for the respective tags depending on the tag score from high to low. The preference tags may be topK tags in the tag ranking result, wherein different K values may be preset based on different tags. For example, the video tag with the historical behavior of the target user u is classified into a first class classification tag including a television series, a movie, etc., and a second class classification tag including a comedy, a love, a scenario, etc., and the K value of the corresponding first class classification tag is 2, the K value of the second class classification tag is 3, and the K values are preset in sequence.

Taking the preference label of the target user as a cartoon movie as an example, presetting the matching weight of the cartoon movie label as 100, enabling the cartoon movie to belong to a first class classification label K value as 2, and determining the score of the target user for the theme conforming to the cartoon movie label according to the number of target preference labels matched with the theme label as 3.

Optionally, topic labels of topics can be allocated from the topic library to be matched with preference labels of the target user, so as to obtain the number of topic labels matched with the preference labels of the target user. In combination with the tag matching weights of the preference tags, the K values of the preference tags can determine the target user's score for each topic. The topic which is most in line with the preference label of the target user can be screened out, the quality of the recommendation effect can be ensured, and the recommendation efficiency is improved.

Hereinafter, a process of determining a score of a target user for each topic will be described.

Fig. 6 is a flowchart of a first topic score of a topic recommendation method of an intelligent tv provided in an embodiment of the present application, as shown in fig. 6, where the process of determining a score of a target user for each topic according to a tag matching weight of a preference tag, the number of matching preference tags of the target user with topic tags of various topics on the preference tag, and a K value of the preference tag includes:

S601, determining a second ratio of the number of matching of the first preference label of the target user and the topic label of the first topic on the first preference label to the K value of the first preference label, wherein the first preference label is any preference label of the target user, and the first topic is any topic.

S602, determining the product of the tag matching weight of the first preference tag and the second ratio to obtain the score of the first theme on the first preference tag.

S603, determining the score of the target user for the first theme according to the scores of the first theme on the preference labels.

Alternatively, the score of the target user for each topic may be obtained by the following formula (2). Specifically, the following formula (2) is called for each topic, and the score of the topic can be obtained.

Score _us ＝∑ _l∈L w _l *cnt _l /k _l (2)

Wherein Score _us A score representing the subject s for the target user u; l represents a theme label dimension set; w (w) _l Representing tag matching weights with tag dimensions of I tags; cnt _l The number of matching in the tag dimension I of the preference tag of the target user u and the topic tag of the topic s is represented; k (k) _l The tag K value representing the preference tag of the target user in the tag dimension I.

Wherein L represents a label dimension set, which comprises a primary classification, a secondary classification, a director, a year, a region, a language and other label dimensions, and L is any label dimension in L. w (w) _l The label dimension is the label matching weight of the I label, and different label matching weights are preset for labels with different dimensions according to actual service conditions and service experience.

Examples are as follows: assume that the tag matching weight for the primary class is 100 and that for the secondary class is 80. The primary classification preference tag K value is 2 and the secondary classification preference tag K value is 3. The first class preference label of the target user u is [ TV drama, movie ], and the second class preference label is [ love, action, comedy ]. The labels of the subject s are: the primary class label is a movie and the secondary class label is [ comedy |love ]. As can be seen from the formula (2), the matching score of the tag information of the preference tag and the subject tag of the target user u is 100×1/2+80×2/3= 103.33 (taking the last two decimal places).

Fig. 7 is a theme pushing flowchart of a theme recommendation method of an intelligent tv provided in the embodiment of the present application, as shown in fig. 7, the step S204 may include:

s2041, sorting scores of target users aiming at various topics, and obtaining a preference topic sequence of the target users according to sorting results, wherein the preference topic sequence comprises preference topics which are sequentially arranged.

Optionally, the scores of the target users for the various topics are ranked, and the scores of the target users for the various topics are obtained according to preference distribution of the target users for the various topics. The sequencing result can well feed back the preference topic sequences of the target users, wherein the preference topic sequences are the sets of preference topics which are sequentially arranged.

Taking the behavior data of a target user in seven days as an example, presetting the tag matching weight of a first classified tag film as 100, the tag matching weight of a second classified tag comedy as 80, wherein the K value of a first class classified tag is 2, and the K value of a second class classified tag is 3. The target users watch 5 comedy movies and 3 love dramas in seven days, the number of matches between the theme labels and the movies with the first class classification labels is 1, and the number of matches between the theme labels and the dramas with the first class classification labels is 2; the number of comedy matches with the secondary category labels in the theme label is 1, and the number of love matches with the secondary category labels in the theme label is 1. As can be seen from the above formula (2), the topic of comedy movies has a score of 100×1/2+80×1/3=76.67, and the topic of love dramas has a score of 100×2/2+80×1/3= 126.67. And obtaining the preference theme arrangement of the target user according to the theme scores, wherein love TV drama > comedy movies.

Optionally, the result of the ranking of the scores of the target users for various topics can exactly obtain the result of the arrangement of the preferred topics of the target users, and the sequence of the preferred topics is the result of the sequential arrangement of the preferred topics.

S2042, determining an initial video set of each preference theme according to the theme label of each preference theme.

Optionally, the initial video set is that the preference labels of the target users are matched with the topic labels of the topic library, and finally the topic labels which accord with the preference labels of the target users are screened out, wherein the topics corresponding to the topic labels are the preference topics of the target users, and the videos contained in the preference topics are used as the initial video set.

Taking the behavior data of the target user in seven days as an example, obtaining that one preference theme of the target user is love TV drama, and then taking all videos under the theme label which accords with the preference theme of the love TV drama in the theme library as an initial video set of the preference theme of the target user.

Alternatively, an initial video set for each preferred topic may be determined based on the topic tags for the preferred topics.

S2043, determining the score of each video in the initial video set of each preference theme according to the historical behavior data of the target user.

Optionally, different behavior weights are set based on historical behavior data of the target user, and the score of each video in the initial video set corresponding to each preference topic can be determined in combination with the topic score of each preference topic.

S2044, determining a target video set of each preference theme according to the scores of the videos in the initial video set of each preference theme, wherein the number of videos in the target video set is smaller than that in the initial video set.

Alternatively, a fixed number N of videos may be set as the target video set, where the fixed number N may be set to 100, 50, 10, etc., and slightly adjusted according to the specific number of initial videos. And sorting the scores of the videos in the initial video collection of the preferred topics of the target user, and sequentially taking a fixed number N of videos from the score as the target video collection, wherein the number of videos in the target video collection is necessarily smaller than that in the initial video collection.

Optionally, the target video set may be determined by sequentially arranging the scores according to the score of each video in the initial video set of each preference theme, where the number of videos in the target video set is smaller than the number of videos in the initial video set.

S2045, determining the score of each preference theme according to the score of each video in the target video set of each preference theme.

Optionally, based on the score of each video in the target video set of each preference topic, the scores of all the videos corresponding to the preference topic corresponding to the video are aggregated and averaged, so that the score of each preference topic can be obtained.

Taking historical behavior data of a target user for seven days as an example, assume that the score of the target user on the video A in a target video set in a preferred theme comedy movie is 100 and the score of the video B is 80; the score for video C in the target video set in the preferred subject war movie was 70 and the score for video D was 80. The preferred subject comedy movie now has a score of (100+80)/2=90 and the preferred subject war movie has a score of (70+80)/2=75.

Alternatively, a score for each preference topic may be derived based on the score for each video in the target video set for the respective preference topic.

S2046, sorting the preference topics according to the scores of the preference topics, and determining a target topic sequence according to the sorting result.

Optionally, the target topic sequence is to aggregate and average to obtain corresponding preference topic scores based on each video score in the target video set, and sequentially order the preference topic scores from high to low to obtain a new arrangement result, where the arrangement result is the target topic sequence.

Taking historical behavior data of a target user for seven days as an example, assume that the score of the target user on the video A in a target video set in a preferred theme comedy movie is 100 and the score of the video B is 80; the score for video C in the target video set in the preferred subject war movie was 70 and the score for video D was 80. The preferred subject comedy movie now has a score of (100+80)/2=90 and the preferred subject war movie has a score of (70+80)/2=75. The preference topic scores are ranked at this time. Comedy movie > war movie, a new theme sequence is obtained.

Alternatively, the scores of the preferred topics may be sequentially arranged from high to low to obtain a new topic sequence, which may be used as the target topic sequence.

S2047, pushing the target subject sequence to a target user.

Optionally, the target topic sequence is used as a final topic recommendation result, the topic sequence is sent to the smart television of the target client, and finally the smart television displays the topic recommendation result.

In this embodiment, the preference topic sequence of the target user is obtained according to the score ranking of the target user for each topic, and the initial video collection conforming to the preference topic is screened out depending on the determined preference label. And processing the historical behavior information of the target user to obtain the score of each video in the initial video, sorting and screening out a target video set according to the score of the initial video, and meanwhile, aggregating and averaging the scores of the target videos to obtain the preference theme score. And sequencing the preference topics, determining a recommended target topic sequence, sending the recommended result to a target user through a wireless communication technology, and displaying the recommended result by the intelligent television. And screening the themes of the theme library layer by layer according to the historical behavior information of the target user and the interest preference, so that the theme recommendation effect in front of thousands of people is finally realized, the recommendation efficiency is improved, and the operation burden is reduced.

Fig. 8 is a video scoring flow chart of a topic recommendation method of an intelligent television according to an embodiment of the present application, and as shown in fig. 8, the process of determining the score of each video in the initial video set of each preference topic according to the historical behavior data of the target user includes:

s801, according to historical behavior data, the number of times each behavior of a target user occurs for each video and the number of times each behavior of the target user occurs for all videos are determined.

Optionally, the service database of the target user is processed to obtain the historical behavior data of the target user, so that the accurate times and time of the occurrence of the historical behavior information of the target user and the video content corresponding to the historical behavior data can be obtained. The accurate times of the target user for generating the search behavior information, the play behavior information and the click behavior information aiming at each video can be obtained based on the historical behavior information of the target user.

Taking the historical behavior of the target user in seven days as an example, the target user watches 15 movies in 7 days, searches 5 movies, clicks 10 videos including comedy movies, love movies, war movies and the like, wherein the target user plays 7 comedy movies, 3 love movies and 5 war movies; searching 3 comedy movies, 1 love movies and 1 war movies; clicking comedy movie 5, love movie 3, war movie 2. It can be known that the target user has 7 play actions, 3 search actions and 5 click actions for comedy movies; aiming at the love movie, 3 playing behaviors, 1 searching behavior and 3 clicking behaviors occur; aiming at war movies, 5 playing behaviors, 1 searching behavior and 2 clicking behaviors occur; the target user has 15 play actions, 5 search actions, 10 click actions for all videos.

Optionally, based on the historical behavior information of the target user, the number of times each behavior of the target user occurs for each video and the type of each behavior of the target user occurs for all videos may be identified.

S802, determining the score of the target user for each video according to the behavior weight of each behavior, the number of times each behavior occurs to the target user for each video and the number of times each behavior occurs to the target user for all videos.

Optionally, after the historical behavior information of the target user is subjected to data processing, interest preference of the target user can be intuitively presented according to the historical behavior information of the target user, and different behavior weight parameters are set for the search behavior information, the play behavior information and the click behavior information of the target user. And then, according to the behavior weight set by each type of historical behavior information of the target user, combining the accurate times of the target user for searching the behavior information, playing the behavior information and clicking the behavior information or the times of the target user for searching the behavior information, playing the behavior information and clicking the behavior information of all videos, and comprehensively obtaining the score of the target user for each video.

Optionally, the score of the target user for each video can be obtained exactly by combining the behavior weight of each behavior of the target user, the number of times each behavior of the target user occurs for each video, and the number of times each behavior of the target user occurs for all videos.

It should be noted that one action may occur on each video or multiple actions may occur simultaneously.

Fig. 9 is a first video scoring flowchart of a topic recommendation method of a smart tv provided in an embodiment of the present application, as shown in fig. 9, the step S502 may include:

s901, determining a third ratio of the number of times of first behaviors of a target user aiming at a first video to the number of times of first behaviors of the target user aiming at all videos, wherein the first video is any video in an initial video set, and the first behaviors are any historical behaviors of the target user.

S902, determining the product of the behavior weight of the first behavior and the third ratio to obtain a score of the combination of the first video and the first behavior.

S903, determining the score of the target user for the first video according to the score of the first video and the combination of the behaviors.

Alternatively, the score of the target user for each video may be obtained by the following formula (3). Specifically, the following formula (3) is called for each video, so that the score of the label can be obtained.

Score _ui ＝∑ _tεT ∑ _a∈A w _a *cnt _tia /all_cnt _ta (3)

Wherein Score _ui Representing the score of target user u for video i, A represents the behavior type set of target user u, w _a Representing the behavior weight of the behavior type a of the target user u, T represents the use of different periodic dimensions, and T represents any periodic dimension belonging to T. cnt _tia When the period dimension is taken as t, the target user u generates behavior times with a behavior type of a for the video content i in the period circumference t; all_cnt _ta And when the period dimension is taken as t, the target user u generates the behavior times of which the behavior type is a for all videos in the period circumference t.

The video i is taken from any video content in an initial video collection obtained by matching the target user preference tag with the topic library topic tag; the behavior type A comprises search behavior, click behavior, play behavior and the like, wherein a is any one of the behavior types of the search behavior, the click behavior, the play behavior and the like; w (w) _a The behavior weight of the behavior type a representing the occurrence of the target user u is different because the different behavior types reflect the interest preference degree of the target user, such as the search behavior and the playing behavior are stronger than the interest preference of the target user reflected by the clicking behavior, and the playing behavior is stronger than the interest preference of the target user reflected by the search behavior, so that the behavior weight is preset according to the degree reflecting the interest preference of the target user in order to better fit the interest preference of the user, the higher the reflecting degree is, the higher the weight is supposed to be set, and the historical behavior data of the target user u is required to be combined for setting; t represents different cycle dimensions, and in order to comprehensively obtain a final recommendation result by comprehensively analyzing long-term behavior data and short-term behavior data of a target user u, four different cycle dimensions of the last 1 day, the last 7 days, the last 15 days and the last 30 days are adopted.

Because the action weight participates in the interest modeling scoring calculation of the target user u, if the target user u performs actions with higher weight on the video a multiple times, for example, plays the video a or searches the video a multiple times, the cloud computing algorithm may assign higher recommendation weight to other videos a 'similar to the video a, and is more prone to recommending the video a'.

Examples are as follows: the system presets that the behavior type a is that the behavior weight of the playing behavior is 100, when the target user u is in the condition of the video content i, the frequency of playing the behavior a of the target user u in the video content i is 30 times in the near 30 days, and the frequency of playing the behavior a of the target user u in all the video contents in the near 30 days is 150 times; the number of times of playing actions a of the target user u on the video content i in the near 15 days is 15 times, and the number of times of playing actions a of the target user u on all the video content in the near 15 days is 90 times. The formula (3) can know that the number of times of playing actions a of the target user u in the video content i is 30 times and the number of times of playing actions a of all the video content i is 150 times in the period dimension of about 30 days, so that a third ratio of 30/150 can be obtained; the relation between the behavior weight of the playing behavior and the video content i in the period dimension can be known by the formula (3), and a first combined score of 100 (30/150) can be obtained; similarly, in the period dimension of 15 days, the number of times of playing behavior a of the target user u in the video content i is 15 times, the number of times of playing behavior a of all the video content i is 90 times, the ratio of 15/90 can be obtained, in the period dimension, the combined Score of 100 x (15/90) can be obtained by the behavior weight of the playing behavior and the behavior relationship of the video content i in the period dimension, and finally the preference Score of the target user u in the video content i of 30 days and 15 days is Score _uj =100 (30/150) +100 (15/90) =36.67 (taking the decimal point two decimal places).

In this embodiment, different weights are purposefully set for different rows of data of the target user, and the video content which is the most suitable for the interest of the target user is finally obtained by comprehensively analyzing the long-term behavior data and the short-term behavior data of the target user and mainly sorting the interest scores of the target user for the video content. Therefore, the method can realize real-time recommendation of the topics and the topic contents which are the most in line with the interests of the user for the user, realize personalized recommendation effect of the topics of thousands of people and thousands of faces, and greatly improve recommendation efficiency.

The following describes a device, equipment, a computer readable storage medium, etc. for implementing the subject recommendation method of the smart tv, and specific implementation processes and technical effects of the subject recommendation method are referred to above, which are not described in detail below.

Fig. 10 is a schematic structural diagram of a theme recommendation apparatus of an intelligent television according to an embodiment of the present application, and referring to fig. 10, the apparatus includes:

an obtaining module 1001, configured to obtain historical behavior information of a target user, where the historical behavior information includes: searching behavior information, playing behavior information and clicking behavior information;

A first determining module 1002, configured to determine a score of the target user for each tag according to the historical behavior data;

a second determining module 1003, configured to determine a score of the target user for each topic according to the score of the target user for each tag;

and the pushing module 1003 is configured to push the topics to the target user according to the score of each topic of the target user.

As an alternative embodiment, the first determining module 1002 is specifically configured to:

according to the historical behavior data, determining the times of each behavior of a target user aiming at each tag and the times of each behavior of the target user aiming at all videos;

Determining the product of the behavior weight of the first behavior and the first ratio to obtain a score of the combination of the first tag and the first behavior;

and determining the score of the target user for the first label according to the score of the first label and the combination of the behaviors.

As an alternative embodiment, the second determining module 1003 is specifically configured to:

and determining the score of the target user for each topic according to the tag matching weight of the preference tag, the number of matching of the preference tag of the target user and the topic tags of various topics on the preference tag and the K value of the preference tag.

determining a second ratio of the number of matching of the first preference tag of the target user and the topic tag of the first topic on the first preference tag to the K value of the first preference tag, wherein the first preference tag is any preference tag of the target user, and the first topic is any topic;

As an alternative embodiment, the pushing module 1004 is specifically configured to:

sorting scores of the target users aiming at various topics, and obtaining a preference topic sequence of the target users according to sorting results, wherein the preference topic sequence comprises preference topics which are sequentially arranged;

Pushing the target subject sequence to a target user.

according to the historical behavior data, determining the times of each behavior of a target user for each video and the times of each behavior of the target user for all videos;

and determining the score of the target user for each video according to the behavior weight of each behavior, the number of times each behavior occurs to the target user for each video and the number of times each behavior occurs to the target user for all videos.

determining a third ratio of the number of times of first behaviors of a target user aiming at a first video to the number of times of first behaviors of the target user aiming at all videos, wherein the first video is any video in an initial video set, and the first behaviors are any historical behaviors of the target user;

determining a product of the behavior weight of the first behavior and a third ratio to obtain a score of a combination of the first video and the first behavior;

a score for the target user for the first video is determined based on the score for the first video and the combination of behaviors.

The foregoing apparatus is used for executing the method provided in the foregoing embodiment, and its implementation principle and technical effects are similar, and are not described herein again.

The above modules may be one or more integrated circuits configured to implement the above methods, for example: one or more application specific integrated circuits (Application Specific Integrated Circuit, abbreviated as ASICs), or one or more microprocessors, or one or more field programmable gate arrays (Field Programmable Gate Array, abbreviated as FPGAs), etc. For another example, when a module above is implemented in the form of a processing element scheduler code, the processing element may be a general-purpose processor, such as a central processing unit (Central Processing Unit, CPU) or other processor that may invoke the program code. For another example, the modules may be integrated together and implemented in the form of a system-on-a-chip (SOC).

Fig. 11 is a schematic structural diagram of a computer device according to an embodiment of the present application. Referring to fig. 11, a computer apparatus includes: a memory 1101, and a processor 1102, wherein the memory 1101 stores a computer program executable on the processor 1102, and the processor 1102 implements the steps of any of the various method embodiments described above when executing the computer program.

The present application also provides a computer readable storage medium storing a computer program which, when executed by a processor, implements the steps of the respective method embodiments described above.

Optionally, the present application further provides a program product, such as a computer readable storage medium, including a program, which when executed by a processor is configured to perform the subject recommendation method embodiment of any of the above-mentioned smart televisions.

In the several embodiments provided by the present invention, it should be understood that the disclosed apparatus and method may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, e.g., the division of elements is merely a logical functional division, and there may be additional divisions of actual implementation, e.g., multiple elements or components may be combined or integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or units, which may be in electrical, mechanical or other form.

The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed over a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in hardware plus software functional units.

The integrated units implemented in the form of software functional units described above may be stored in a computer readable storage medium. The software functional unit is stored in a storage medium, and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) or a processor (english: processor) to perform part of the steps of the methods of the embodiments of the invention. And the aforementioned storage medium includes: u disk, mobile hard disk, read-Only Memory (ROM), random access Memory (Random Access Memory, RAM), magnetic disk or optical disk, etc.

The foregoing is merely a specific embodiment of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can easily think about changes or substitutions within the technical scope of the present application, and the changes or substitutions are covered by the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

The foregoing description is only of the preferred embodiments of the present application and is not intended to limit the same, but rather, various modifications and variations may be made by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. The topic recommendation method of the intelligent television is characterized by comprising the following steps of:

acquiring historical behavior information of a target user, wherein the historical behavior information comprises: searching behavior information, playing behavior information and clicking behavior information;

determining the score of the target user for each tag according to the historical behavior data;

determining the score of the target user for each topic according to the score of the target user for each label;

and pushing the theme to the target user according to the score of the target user for each theme.

2. The method of claim 1, wherein determining the score of the target user for each tag based on the historical behavioral data comprises:

3. The method according to claim 2, wherein determining the score of the target user for each tag according to the behavior weight of each behavior, the number of times each behavior occurs for each tag by the target user, and the number of times each behavior occurs for all videos by the target user, comprises:

4. The method according to claim 1, wherein determining the score of the target user for each topic based on the score of the target user for each tag, comprises:

5. The method according to claim 4, wherein determining the score of the target user for each topic according to the tag matching weight of the preference tag, the number of matching of the preference tag of the target user with topic tags of various topics on the preference tag, and the K value of the preference tag, comprises:

6. The method according to any one of claims 1-4, wherein the pushing of the topics to the target user according to the score of the target user for each topic is characterized by comprising:

pushing the target subject sequence to the target user.

7. The method of claim 6, wherein determining the score for each video in the initial video set for each preferred topic based on historical behavioral data of the target user comprises:

8. The method according to claim 7, wherein determining the score of the target user for each video according to the behavior weight of each behavior, the number of times each behavior occurs for each video by the target user, and the number of times each behavior occurs for all videos by the target user, comprises:

Determining a third ratio of the number of times that the target user generates a first behavior for a first video to the number of times that the target user generates the first behavior for all videos, wherein the first video is any video in the initial video set, and the first behavior is any historical behavior of the target user;

9. A computer device, comprising: memory, a processor, in which a computer program is stored which is executable on the processor, when executing the computer program, implementing the steps of the method of any of the preceding claims 1 to 8.

10. A computer readable storage medium, characterized in that it has stored thereon a computer program which, when executed by a processor, implements the steps of the method according to any of claims 1 to 8.