WO2014046649A1

WO2014046649A1 - Initial recommendation system seeding

Info

Publication number: WO2014046649A1
Application number: PCT/US2012/056009
Authority: WO
Inventors: Sandilya Bhamidipati
Original assignee: Thomson Licensing
Priority date: 2012-09-19
Filing date: 2012-09-19
Publication date: 2014-03-27

Abstract

A recommendation engine obtains login information from a user and extracts their likes and dislikes relating to media content directly from a social media site. This information is extracted from the social media site to jump start the recommendation engine. The extracted information is associated with a user who is initializing a product that offers recommendations. This permits the product to offer more relevant recommendations quicker than waiting for sufficient user interactions.

Description

INITIAL RECOMMENDATION SYSTEM SEEDING

BACKGROUND

[0001] The existing recommendation systems rely on the users to perform some kind of activity/ interact with the system like rate a movie or buy a product to recommend new products based on people who performed similar actions bought or watched. The problem is that when a new user arrives, he has to explicitly rate items to get recommendations. This is a very long time consuming process to build a knowledge base about the user and until such a knowledge base is built, he/she cannot get good recommendations.

SUMMARY

[0002] Social media information is leveraged to initially seed recommendation systems. Information about a user, particularly their interests in media content such as movies and TV shows, etc., can be utilized to cold start or "seed" the

recommendation system in a quick and efficient manner. The extracted information can also be communicated to other users of a particular social media site. In a similar manner, preference information can be extracted from other users on the social media site that are associated (e.g., friends, relatives, etc.) with the user. This information can be combined and used to seed a recommendation engine that provides media content recommendations to first time users of a product.

[0003] The above presents a simplified summary of the subject matter in order to provide a basic understanding of some aspects of subject matter embodiments. This summary is not an extensive overview of the subject matter. It is not intended to identify key/critical elements of the embodiments or to delineate the scope of the subject matter. Its sole purpose is to present some concepts of the subject matter in a simplified form as a prelude to the more detailed description that is presented later.

[0004] To the accomplishment of the foregoing and related ends, certain illustrative aspects of embodiments are described herein in connection with the following description and the annexed drawings. These aspects are indicative, however, of but a few of the various ways in which the principles of the subject matter can be employed, and the subject matter is intended to include all such aspects and their equivalents. Other advantages and novel features of the subject matter can become apparent from the following detailed description when considered in conjunction with the drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0005] FIG. 1 is an example of a recommendation system using an embodiment.

FIG. 2 is a flow diagram of a method 200 of seeding a recommendation system.

DETAILED DESCRIPTION

[0006] The subject matter is now described with reference to the drawings, wherein like reference numerals are used to refer to like elements throughout. In the following description, for purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of the subject matter. It can be evident, however, that subject matter embodiments can be practiced without these specific details. In other instances, well-known structures and devices are shown in block diagram form in order to facilitate describing the embodiments.

[0007] As used in this application, the term "component" is intended to refer to hardware, software, or a combination of hardware and software in execution. For example, a component can be, but is not limited to being, a process running on a processor, a processor, an object, an executable, and/or a microchip and the like. By way of illustration, both an application running on a processor and the processor can be a component. One or more components can reside within a process and a component can be localized on one system and/or distributed between two or more systems. Functions of the various components shown in the figures can be provided through the use of dedicated hardware as well as hardware capable of executing software in association with appropriate software.

[0008] Many of the next generation TV guides/TV interaction systems allow the TV user to let people in his/her social media networks know what they are watching either by tweeting using Twitter or post a message on their social media sites, such as Facebook wall and the like. This clearly shows that the TV user is willing to share his personal information through these social media networks with his friends/followers. This allows the following implementation for recommendation system cold starts.

[0009] As a lot of information is already present about the user on these networks and not harnessed by any of the TV applications, the user has to perform considerable number of actions/interactions with the system to build a profile for him so that the system can give good recommendations on what the user likes to watch. This is a time taking process and results in a degraded TV recommendation experience. The implementation is to extract information about the user with respect to the movies and TV shows watched from the social network and utilize it as in input to the TV guide, so that it will get a preliminary idea on what the user likes based on his interaction and the people he is connected to in his social network. In one example, the data to be used is obtained from a social media site such as, for example, Facebook due to its graph structure of linking people and partially due to its global presence and availability.

[0010] The other advantage of using a social media site similar to Facebook is it will let the user know what his/her friends like along with what the user liked in the past. Additional advantages that can be considered are that the user can have a clear idea on where the recommendations are coming from and increase the trust put in the system, it encourages the user to interact with the system seamlessly and help in building the profile in a faster and efficient way and the user does not have to interact with a TV guide per se to get recommendations. Any social media activity will let the system understand the user and provide the recommendations.

[0011] A quick algorithm on how the process works is as follows:

Step 1: A new user purchases a TV system and would like to interact/use it for entertainment purposes.

Step 2: The system can ask the user to input his/her social media site account (e.g., Facebook) so that they can let their friends know about their new interactive system and at the same time let people in his/her network to recommend shows/movies. Step 3: This step is the information extraction from a social media site after the user allowed the application to access the information from his account. The data being extracted can be as simple as TV shows and movies the user and people in his networked "like" and the number of people who share similar interests.

Step 4: Step 3 output can be used as input for the TV guide to start searching for movies/ TV shows which are similar to what is obtained.

Step 5; Use this data as a good supplementary until the system is confident enough that it gained enough interaction from the user. The process can be continued so that the user will have a better TV viewing experience. [0012] FIG. 1 illustrates an embodiment 100 of the above recommendation system. In this example, a new user 102 interacts with a product for a first time 104. The product can be, but is not limited to, video devices such as televisions, set top boxes, mobile devices (cell phones, media players, etc.) and the like. The user 102 is prompted to enter login information associated with their social media site 106. The user information is sent to the recommendation engine 110 which then interacts with the social media site 114 to obtain media content metadata 112. The metadata information can include, but is not limited to, a user's likes and/or dislikes of various media content and/or a user's friend's likes and/or dislikes, etc. The recommendation engine 110 can also communicate with the social media site 114 to communicate a user's interests to their friends to obtain a larger seed of information. Thus, the user's information can be combined with their friends information to create a larger pool of information for initially seeding the recommendation engine 110. Once the seeding is accomplished the recommendation engine 100 provides the recommendations to the user via a recommendations screen 108 and the like. At this point, the user 102 can further select likes and/or dislikes on the recommendation screen 108 to further enhance the information utilized by the recommendation engine 110. [0013] In view of the exemplary systems shown and described above, methodologies that can be implemented in accordance with the embodiments will be better appreciated with reference to the flow charts of FIG. 2. While, for purposes of simplicity of explanation, the methodologies are shown and described as a series of blocks, it is to be understood and appreciated that the embodiments are not limited by the order of the blocks, as some blocks can, in accordance with an embodiment, occur in different orders and/or concurrently with other blocks from that shown and described herein. Moreover, not all illustrated blocks may be required to implement the methodologies in accordance with the embodiments.

[0014] FIG. 2 is a flow diagram of a method 200 of seeding a

recommendation system. The method starts 202 by establishing a first use by a given user 204. This can be accomplished by detecting that it is the first time the product has been turned on, it is the first time that a user has logged on, and/or it is detected that it is the first time recommendations have been requested, etc. The user's login information for a social media site associated with the user is then obtained 206. The log in information establishes what type of account the user has as well as which sites the user is associated with. More than one account can be utilized per user.

Preference information associated with the user is then extracted from the social media site 208. The preference information includes, but is not limited to, the user's likes and/or dislikes of media content and the like. The user's preference information can also be communicated to others associated with the user on the social media site. The recommendation system is then seeded with the extracted preference information 210, ending the flow 212. In an alternative embodiment, the preference information from other users of the social media site that are associated with the user can be extracted as well. This information can then be combined with the user's preference information to better seed the recommendation system. Once the recommendations are generated they can be displayed to the user. The user can also choose to interact with the recommendations to better fine tune the recommendations, etc.

[0015] What has been described above includes examples of the

embodiments. It is, of course, not possible to describe every conceivable combination of components or methodologies for purposes of describing the embodiments, but one of ordinary skill in the art can recognize that many further combinations and permutations of the embodiments are possible. Accordingly, the subject matter is intended to embrace all such alterations, modifications and variations that fall within the spirit and scope of the appended claims. Furthermore, to the extent that the term "includes" is used in either the detailed description or the claims, such term is intended to be inclusive in a manner similar to the term "comprising" as "comprising" is interpreted when employed as a transitional word in a claim.

Claims

1 A system that provides initial recommendation information, comprising:

a recommendation engine that interfaces with a user and obtains login information for a user associated social media site, extracts the user's preference information relating to media content from the social media site and generates recommendations based on the extracted preference information.

2. The system of claim 1 further comprising:

a recommendations screen that displays the recommendations based on the user's social media site preferences generated by the recommendation engine.

3. The system of claim 1, wherein the preference information includes at least one of the user's likes and the user's dislikes of media content.

4. The system of claim 1, wherein the recommendation engine communicates the user's preference information to others associated with the user on the social media site.

5. The system of claim 4, wherein the recommendation engine extracts preference information from other users of the social media site that are associated with the user.

6. A method for seeding a recommendation system, comprising:

establishing a first use by a given user;

obtaining the user's login information for a social media site associated with the user;

extracting preference information associated with the user from the social media site; and

seeding the recommendation system with the extracted preference

information.

7. The method of claim 6, wherein the preference information includes at least one of the user's likes and the user's dislikes of media content.

8. The method of claim 6 further comprising:

communicating the user's preference information to others associated with the user on the social media site.

9. The method of claim 6 further comprising:

extracting preference information from other users of the social media site that are associated with the user.

10. The method of claim 6, further comprising:

displaying recommendations to the user based on their social media site preference information.

11. A system that seeds a media content recommendation engine, comprising:

a means for establishing a first use by a given user;

a means for obtaining the user's login information for a social media site associated with the user;

a means for extracting preference information associated with the user from the social media site; and

a means for seeding the recommendation system with the extracted preference information.

12. The system of claim 11 further comprising:

a means for displaying recommendations to the user based on their social media site preference information.