KR102137887B1 - Vod contents recommendation server and method for iptv to combine the preference information of the iptv and ott services - Google Patents

Abstract

Disclosed is an IPTV movie VOD content recommendation server and a method thereof through a combination of movie VOD preference information of a mobile OTT service of an IPTV service. The content recommendation method performed by the IPTV content recommendation server according to an embodiment includes: collecting preference data for VOD content watched through IPTV and VOD content watched through OTT, respectively; Calculating predicted preferences of VOD content related to a user by integrating the collected preference data; And recommending VOD content based on the calculated prediction preference.

Description

IPTV Movie VOD Content Recommendation Server and Method through Combining Movie VOD Preference Information of Mobile OTT Service of IPTV Service{VOD CONTENTS RECOMMENDATION SERVER AND METHOD FOR IPTV TO COMBINE THE PREFERENCE INFORMATION OF THE IPTV AND OTT SERVICES}

The following description relates to a method of providing a VOD content recommendation service to an IPTV user by combining preference data for VOD content of an IPTV service and a mobile OTT service.

The IPTV recommendation system is difficult to measure the preference for products compared to other recommendation systems. In other recommendation system domains, there is a lot of information that can measure the user's preferences through data directly input by the user, and the preferences can often be measured with explicit information and implicit information. The explicit information indicates the user's rating information for the VOD content, and the suggestive information can be measured through the user's behavior information. Representative suggestive information includes the number of times the content has been downloaded and the number of times a web page has been visited. In the case of IPTV, unlike other domains, it is troublesome for users to input preference data. Unlike PCs and mobiles, IPTVs do not have a user-friendly input device such as a mouse or touch interface, so there are many limitations on the UI, so it is easy to obtain explicit or suggestive information that can measure preferences for content required in the recommendation system. not. Therefore, there is a disadvantage in that an accurate recommendation result cannot be provided to a user due to a data scarcity problem.

The present invention aims to solve the data scarcity problem, which is a fundamental problem of the IPTV movie VOD content recommendation system. We intend to solve the problem of data scarcity and the inaccuracy of recommendation results that occur when using only IPTV movie VOD content preference data by integrating IPTV movie VOD content preference data with mobile OTT movie VOD content preference data.

The content recommendation method performed by the IPTV content recommendation server comprises: collecting preference data for VOD content viewed through IPTV and VOD content viewed through OTT, respectively; Calculating predicted preferences of VOD content related to a user by integrating the collected preference data; And recommending VOD content based on the calculated prediction preference.

In the step of collecting the preference data for the VOD content viewed through the IPTV and the VOD content viewed through the OTT, the IPTV movie VOD content data collected from the IPTV data providing server, the preference data for the IPTV movie VOD content and the OTT And receiving the OTT movie VOD content data collected from the data providing server and the preference data for the OTT movie VOD content, collecting genre-specific information for each movie VOD content, and each according to the collected genre-specific information. Preference data for each genre may be stored.

The step of calculating the predicted preferences of the VOD content related to the user by integrating the collected preference data is to integrate the preference data for the IPTV movie VOD contents and the OTT movie VOD contents collected from the OTT data providing server. Accordingly, the method may include generating integrated VOD preference data and applying the generated integrated VOD preference data to a recommendation algorithm.

The step of calculating the predicted preferences of the VOD contents related to the user by integrating the collected preference data, the movie considering the average preference data for each genre for each of the integrated VOD preference data, IPTV movie VOD contents and OTT movie VOD contents And deriving a prediction preference of the VOD content, and an average preference value for each genre can be applied as a weight to derive the prediction preference of the VOD content.

The step of recommending VOD content based on the calculated prediction preference may include recommending VOD content of a predetermined rank or higher to a user by sorting the calculated prediction preference value in descending order.

The IPTV content recommendation server includes a data collection unit that collects preference data for VOD content viewed through IPTV and VOD content viewed through OTT, respectively; A computing unit that integrates the collected preference data and calculates prediction preferences of VOD content related to a user; And a recommendation unit that recommends VOD content based on the calculated prediction preference.

The data collection unit receives IPTV movie VOD content data collected from an IPTV data providing server, preference data for IPTV movie VOD content, OTT movie VOD content data collected from an OTT data providing server, and preference data for OTT movie VOD content Including, and collecting information for each genre for each movie VOD content, preference data for each genre according to the collected information for each genre may be stored.

The operation unit generates the integrated VOD preference data as the preference data for the IPTV movie VOD content and the preference data are collected in the OTT movie VOD content collected from the OTT data providing server, and the generated integrated VOD preference data is added to the recommendation algorithm. Can be applied.

The calculating unit includes deriving prediction preferences of movie VOD contents by considering average preference data for each genre for each of the integrated VOD preference data, IPTV movie VOD contents, and OTT movie VOD contents, and deriving prediction preferences of the VOD contents In order to do so, the average preference value for each genre can be applied as a weight.

The recommender may recommend the VOD content of a predetermined rank or higher to a user by sorting the calculated prediction preference value in descending order.

The data scarcity problem of the IPTV movie VOD content recommendation system can be solved by combining the preference data for IPTV movie VOD contents and OTT movie VOD contents, and it is possible to recommend VOD contents with higher performance.

In addition, on the contrary, it can be applied to the OTT movie VOD content recommendation system, and the performance of the OTT movie VOD content recommendation can be improved by adding the IPTV movie VOD content preference value.

1 is a view for explaining the configuration of a VOD content system according to an embodiment.
2 is a block diagram illustrating a configuration of an IPTV content recommendation server according to an embodiment.
3 is a diagram for explaining a process of recommending VOD content from the IPTV content recommendation server according to an embodiment.
4 is a flowchart illustrating a method of recommending VOD content in an IPTV content recommendation server according to an embodiment.

Hereinafter, embodiments will be described in detail with reference to the accompanying drawings.

1 is a view for explaining the configuration of a VOD content system according to an embodiment.

The VOD content system 100 may include an IPTV content recommendation server 110, an IPTV data provision server 120, an OTT data provision server 130, and a set-top box 140. At this time, the set-top box 140 can transmit the preference for VOD content generated by the user to the IPTV data providing server 120 as a receiving device for IPTV, and of IPTV to recommend suitable VOD content for IPTV viewers. The preference for the VOD content may be received from the IPTV data providing server 120 and transmitted to the IPTV content recommendation server 110. In addition, the set-top box 140 may receive the preference for the OTT VOD content from the OTT data providing server 130 by requesting the OTT VOD content data to the OTT data providing server 130, the received OTT VOD content The preference for the IPTV can be transmitted to the IPTV content recommendation server 110.

For example, the set-top box 140 may transmit the preference for the movie VOD content generated by the user to the IPTV data providing server 120 as a receiving device for IPTV, and provide IPTV movie VOD contents suitable for IPTV viewers. For recommendation, the movie VOD preference may be received from the IPTV data providing server 120 and transmitted to the IPTV content recommendation server 110. In addition, the set-top box 140 receives the preference for the OTT movie VOD content from the OTT data providing server 130 by requesting the OTT movie VOD content data to the OTT data providing server 130, the IPTV content recommendation server 110 Can be sent to. At this time, as a VOD content, for example, various genres such as a movie, music, drama, and entertainment may exist, and preferences (IPTV VOD content preference and OTT VOD content preference) for each genre VOD content may be collected. Hereinafter, in an embodiment, a method of collecting and recommending movie VOD contents will be described to help understanding of the description among VOD contents. In addition, it will be described, for example, that the IPTV content server recommends the movie VOD to the IPTV service. In the same way, a movie VOD content can be recommended by the OTT content server to the OTT service.

The IPTV data providing server 120 mainly collects data for IPTV movie VOD content and user preference data for IPTV movie VOD content, and the OTT data providing server 130 provides data for OTT movie VOD content and OTT movie VOD It is possible to mainly collect user preference data for content. In addition, the IPTV data providing server 120 and the OTT data providing server 130 may also collect preference data for each movie VOD content and also collect genres for movie VOD contents to collect preference data for each genre.

2 is a block diagram illustrating the configuration of an IPTV content recommendation server according to an embodiment, and FIG. 4 is a flowchart illustrating a method of recommending VOD content in an IPTV content recommendation server according to an embodiment.

The processor of the IPTV content recommendation server 110 may include a data collection unit 210, a calculation unit 220, a recommendation unit 230, and a storage unit 240. The components of the processor may be expressions of different functions performed by the processor according to a control command provided by the program code stored in the IPTV content recommendation server. The processor and components of the processor may control the IPTV content recommendation server to perform steps 410 to 440 included in the method for recommending VOD content in FIG. 4. At this time, the processor and the components of the processor may be implemented to execute instructions according to the code of the operating system included in the memory and the code of at least one program.

The processor may load program code stored in a file of a program for a method of recommending IPTV content into memory. For example, when a program is executed in the IPTV content recommendation server, the processor may control the IPTV content recommendation server to load program code from a program file into a memory under control of an operating system. In this case, each of the processor and the data collection unit 210, the operation unit 220, the recommendation unit 230, and the storage unit 240 included in the processor executes a corresponding portion of the program code loaded in the memory to execute a subsequent step. It may be different functional representations of the processor for executing the fields (410 to 440).

The data collection unit 210 may collect basic information about IPTV movie VOD contents and OTT movie VOD contents. For example, the data collection unit 210 may collect IPTV movie VOD content data from the IPTV data providing server (410), and may collect preference data for IPTV movie VOD content collected from the IPTV data providing server. (411).

The data collection unit 210 may collect OTT movie VOD content data from the OTT providing server (420), and may collect preference data for the OTT movie VOD content collected from the OTT data providing server (421). At this time, the data collection unit 210 may collect IPTV movie VOD content data and OTT movie VOD content data at the same time, for a period of time or in a predetermined order. In addition, the data collection unit 210 may collect information for each genre of the movie VOD content, and may collect preference data for each genre according to the collected information for each genre.

As such, the data collection unit 210 may transfer the collected data to the operation unit 220 to calculate the recommended movie VOD content and may store it in the storage unit 300. The storage unit 240 may classify and store collected data for each user.

In operation 430, the operation unit 220 may combine user preference data for VOD content of IPTV and OTT. The calculating unit 220 may calculate the predicted preference of movie VOD content related to a user by integrating each collected VOD preference data. The operation unit 220 may generate IPTV movie VOD preference data by integrating IPTV movie VOD preference data and OTT movie VOD preference data, and apply the generated integrated movie VOD preference data to a recommendation algorithm to calculate movie VOD content suitable for a user. have. In other words, the operation unit 220 may calculate a movie VOD content preference value suitable for the user, and predict a movie VOD content related to the user based on the calculated movie VOD content preference value. As such, the calculated movie VOD content preference value may be stored in the storage unit 240 for the next recommendation. The storage unit 240 may store the predicted movie VOD content and the predicted movie VOD preference value as calculated by the calculator 220.

The recommendation unit 230 may recommend IPTV content to the user. The recommendation unit 230 may recommend the movie VOD content to the user who is using the IPTV service. In this case, the movie VOD content for one genre of the movie VOD contents may be recommended, and the movie VOD content for a plurality of genres may be recommended. Alternatively, the recommended movie VOD contents may be arranged based on the ranking and provided as a list. For example, a list of recommended movie VOD contents for one genre may be provided, and a list of recommended movie VOD contents for each of a plurality of genres may be provided as a list. Accordingly, the user can select and watch one of the recommended movie VOD contents.

3 is a diagram for explaining a process of recommending VOD content from the IPTV content recommendation server according to an embodiment.

The IPTV content recommendation server can measure the predicted preference for movie VOD contents by combining the preference for VOD contents of IPTV and the preference for VOD contents of OTT. The IPTV content recommendation server may consider the preference value for each of the movie VOD contents and the average preference for each genre of the movie VOD contents in order to provide the recommended movie VOD contents suitable for the user.

The IPTV content recommendation server may integrate a preference value for IPTV movie VOD content and a preference value for OTT movie VOD content. Then, the average preference value for each genre for the IPTV movie VOD content and the average preference value for each genre for the OTT movie VOD content may be integrated.

라고 하고, OTT 영화 VOD 콘텐츠에 대한 선호도 값을

라고 할 때, 사용자의 영화 VOD 콘텐츠에 대한 최종 선호도 값(

)은 수학식 1과 같이 나타낼 수 있다. The following explains how to integrate preference values. User's IPTV movie VOD content preference value

And the preference value for OTT movie VOD content.

Is the final preference value for the user's movie VOD content (

) Can be expressed as in Equation 1.

Equation 1:

와

이 하나(각각)만 존재할 경우, 하나의 선호도 값(각각의 선호도 값)을 사용자의 영화 VOD 콘텐츠에 대한 선호도로 취급할 수 있다. 만약,

와

이 모두 존재할 경우, 두 값의 평균으로 VOD 콘텐츠에 대한 선호도 값을 측정할 수 있다.

와

이 모두 존재하지 않는 경우, VOD 콘텐츠에 대한 선호도 값을 0으로 처리할 수 있다. Prior to the description of Equation 1, when the ranges of the two preference values are different, the ranges of the two preference values may be adjusted to a certain range or may be equally adjusted. To briefly explain Equation 1,

Wow

If only one (each) exists, one preference value (each preference value) may be treated as a preference for the user's movie VOD content. if,

Wow

If both are present, the preference value for VOD content can be measured as the average of the two values.

Wow

If all of these do not exist, the preference value for VOD content can be treated as 0.

)의 가중치로 적용할 수 있다. 예를 들면, 두 장르의 VOD에 대한 선호도 값이 동일하다고 가정하였을 때, 특정 장르에 대한 선호도에 따라 어떤 장르의 VOD 콘텐츠를 더 선호하는지 측정할 수 있다. 이에 따라, VOD 콘텐츠 자체의 선호도 값에 장르에 대한 선호도를 가중치로 하여 추천 알고리즘에 적용시킬 수 있다. 영화에 복수 개의 장르(예를 들면, 코믹, 로맨스, 액션 등)가 존재하기 때문에 장르별 가중치를 부여하기 위하여 각 장르별 평균 선호도 벡터와 영화 장르 벡터가 필요하다. 영화 장르 벡터는 총 n(n은 자연수)개의 장르가 정의되어 있다고 가정할 때, 해당 영화가 가지고 있는 장르를 표현하는 벡터로 존재하는 장르에 1로 표시할 수 있다. 예를 들면, 액션, SF, 스릴러 장르의 영화라고 했을 때, 영화 장르 벡터에서 액션이 2번째, SF가 3번째, 스릴러가 n번째 요소라고 하면, 영화 장르 벡터

는 다음과 같이 표현할 수 있다.Then, the IPTV content recommendation server measures the preference value for each genre to determine the final preference value (

). For example, when it is assumed that the preference values for the VODs of two genres are the same, it is possible to measure which genre of VOD content is preferred according to the preference for a specific genre. Accordingly, the preference value of the genre can be applied to the recommendation algorithm by using the preference value of the VOD content itself as a weight. Since there are multiple genres (e.g., comics, romance, action, etc.) in a movie, the average preference vector for each genre and the movie genre vector are required to weight each genre. Assuming that a total of n (n is a natural number) genre is defined as a movie genre vector, a vector representing a genre of the movie may be represented as 1 in the genre. For example, if a movie is an action, SF, or thriller genre, and the action is 2nd, SF is 3rd, and thriller is the nth element in the movie genre vector, the movie genre vector

Can be expressed as

Equation 2:

In the IPTV content recommendation server, a user's average preference vector for a genre may be expressed as Equation 3 and Equation 5 as a preference value of the IPTV movie VOD content for the genre and an OTT movie VOD content. At this time, the range of the value of the preference can be changed between 0 and 1, and the preference value of the IPTV movie VOD content and the OTT movie VOD content derived by changing the range of the preference value are as shown in Equation 4 and Equation 6. Can be represented.

Equation 3:

Equation 4:

Equation 5:

Equation 6:

The weight of the genre may be calculated as an average of the preference values of all genres. When the total number of genres of movie i (i is a natural number) is N (N is a natural number), the genre weight of preference for IPTV movie VOD content is Equation 7, and the genre weight of preference for OTT movie VOD content Can be expressed as Equation 8.

Equation 7:

Equation 8:

)은 기존의 선호도 값에서 장르의 가중치를 가산함으로써 도출될 수 있다. 이때, 기존의 VOD 콘텐츠 선호도의 낮은 구간의 평점이 높은 구간의 평점보다 높은 점수가 나오지 못하도록 평점의 간격 d를 장르의 가중치에 곱해준다. 예를 들면, 평점이 1, 2, 3, 4, 5인 추천 시스템에서 d=1이고, 이와 같이 평점의 간격을 설정함으로써 평점이 4인 선호도가 평점이 5인 선호도보다 높은 점수로 순위화되는 것을 방지할 수 있다. Final preference value (

) Can be derived by adding the weight of the genre from the existing preference value. At this time, the interval d of the rating is multiplied by the weight of the genre so that the score in the lower section of the existing VOD content preference does not exceed the rating in the higher section. For example, d=1 in a recommendation system having a rating of 1, 2, 3, 4, 5, and by setting the interval of the ratings as described above, a rating with a rating of 4 is ranked with a score higher than a rating with a rating of 5 Can be prevented.

Equation 9:

Equation 10:

By applying the newly measured values to Equation 1 through Equations 9 and 10, the final preference value for the user's movie i can be measured, and the measured final preference value can be applied to the recommendation algorithm. Movie VOD content may be recommended by applying the final preference value measured in this way to the recommendation algorithm.

In the case of the recommendation algorithm, the IPTV content recommendation server can utilize a latent factor model technique such as matrix factorization or tensor factorization, which generally shows high performance in a recommendation system.

Referring to FIG. 3, performance is deteriorated when there is a lot of null data in the user-content matrix. Accordingly, by considering not only the movie VOD preference data on the IPTV service but also the movie VOD preference data on the OTT service and additionally the average preference for the IPTV and OTT movie genres, the integrated movie VOD content preference is calculated to calculate the integrated user-content matrix. Can be created. The generated user-content matrix can be applied to the recommendation algorithm to measure the predicted movie VOD content preference value, and recommend the user to recommend VOD content having a predetermined rank or higher based on the predicted preference value.

The device described above may be implemented with hardware components, software components, and/or combinations of hardware components and software components. For example, the devices and components described in the embodiments may include, for example, processors, controllers, arithmetic logic units (ALUs), digital signal processors (micro signal processors), microcomputers, field programmable gate arrays (FPGAs). , A programmable logic unit (PLU), microprocessor, or any other device capable of executing and responding to instructions, may be implemented using one or more general purpose computers or special purpose computers. The processing device may run an operating system (OS) and one or more software applications running on the operating system. In addition, the processing device may access, store, manipulate, process, and generate data in response to the execution of the software. For convenience of understanding, a processing device may be described as one being used, but a person having ordinary skill in the art, the processing device may include a plurality of processing elements and/or a plurality of types of processing elements. It can be seen that may include. For example, the processing device may include a plurality of processors or a processor and a controller. In addition, other processing configurations, such as parallel processors, are possible.

The software may include a computer program, code, instruction, or a combination of one or more of these, and configure the processing device to operate as desired, or process independently or collectively You can command the device. Software and/or data may be interpreted by a processing device or to provide instructions or data to a processing device, of any type of machine, component, physical device, virtual equipment, computer storage medium or device. Can be embodied in The software may be distributed on networked computer systems, and stored or executed in a distributed manner. Software and data may be stored in one or more computer-readable recording media.

The method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded on a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, or the like alone or in combination. The program instructions recorded on the medium may be specially designed and constructed for the embodiments or may be known and usable by those skilled in computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical media such as CD-ROMs, DVDs, and magnetic media such as floptical disks. -Hardware devices specifically configured to store and execute program instructions such as magneto-optical media, and ROM, RAM, flash memory, and the like. Examples of program instructions include high-level language code that can be executed by a computer using an interpreter, etc., as well as machine language codes produced by a compiler.

As described above, although the embodiments have been described by a limited embodiment and drawings, those skilled in the art can make various modifications and variations from the above description. For example, the described techniques are performed in a different order than the described method, and/or the components of the described system, structure, device, circuit, etc. are combined or combined in a different form from the described method, or other components Alternatively, even if replaced or substituted by equivalents, appropriate results can be achieved.

Therefore, other implementations, other embodiments, and equivalents to the claims are also within the scope of the following claims.

Claims (10)

In the content recommendation method performed by the IPTV content recommendation server,
Collecting preference data for VOD content viewed through IPTV and VOD content viewed through OTT, respectively;
Calculating predicted preferences of VOD content related to a user by integrating the collected preference data; And
Recommending movie VOD content based on the calculated prediction preference
Including,
The step of collecting the preference data for the VOD content viewed through the IPTV and the VOD content viewed through the OTT, respectively,
IPTV movie VOD content data collected from the IPTV data providing server, preference data for IPTV movie VOD content and OTT movie VOD content data collected from the OTT data providing server, and preference data for OTT movie VOD content received, and each movie Collecting genre-specific information on VOD content and storing preference data for each genre according to the collected genre-specific information
Including,
The step of calculating the predicted preference of the VOD content related to the user by integrating the collected preference data,
As the preference data for the IPTV movie VOD contents and the preference data are integrated into the OTT movie VOD contents collected from the OTT data providing server, integrated VOD preference data is generated, and the generated integrated VOD preference data is applied to the recommendation algorithm, and the In order to derive the predicted preference of the movie VOD contents by considering the average preference data for each genre of the integrated VOD preference data, the IPTV movie VOD contents, and the OTT movie VOD contents, the average preference value for each genre is derived to derive the predicted preference of the movie VOD contents Step to apply as weight
Content recommendation method comprising a. delete delete delete According to claim 1,
Recommending the VOD content based on the calculated prediction preference,
Recommending VOD content of a predetermined rank or higher to a user by sorting the calculated value of predicted preference in descending order
Content recommendation method comprising a. In the IPTV content recommendation server,
A data collection unit for collecting preference data for VOD content viewed through IPTV and VOD content viewed through OTT, respectively;
A computing unit that integrates the collected preference data and calculates prediction preferences of VOD content related to a user; And
A recommendation unit that recommends movie VOD content based on the calculated prediction preference
Including,
The data collection unit,
IPTV movie VOD content data collected from the IPTV data providing server, preference data for IPTV movie VOD content and OTT movie VOD content data collected from the OTT data providing server, and preference data for OTT movie VOD content received, and each movie Collecting information by genre for the VOD content, and storing preference data for each genre according to the collected genre information,
The calculation unit,
As the preference data for the IPTV movie VOD contents and the preference data are integrated into the OTT movie VOD contents collected from the OTT data providing server, integrated VOD preference data is generated, and the generated integrated VOD preference data is applied to the recommendation algorithm, and the In order to derive the predicted preference of the movie VOD contents by considering the average preference data for each genre of the integrated VOD preference data, the IPTV movie VOD contents, and the OTT movie VOD contents, the average preference value for each genre is derived to derive the predicted preference of the movie VOD contents Weighted
IPTV content recommendation server. delete delete delete The method of claim 6,
The recommendation unit,
As the values of the calculated prediction preferences are sorted in descending order, VOD content of a predetermined rank or higher is recommended to the user.
IPTV content recommendation server.

Images