Browse Prior Art Database

Method for Dynamic Selection and Ranking of Television channels based on Auxiliary information, Personalized User Preferences, Demographic, Social media and Channel content

IP.com Disclosure Number: IPCOM000247770D
Publication Date: 2016-Oct-06
Document File: 3 page(s) / 43K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is the method for ranking and selection of television channels using User demographic, Users explicit preferences, historically watched channels, External information on various programs running on different channels, Users internet social activity, Users internet browsing patterns

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 51% of the total text.

Page 01 of 3

Method for Dynamic Selection and Ranking of Television channels based on Auxiliary information, Personalized User Preferences, Demographic, Social media and Channel content

Current system lacks smart way of recommending television channels to the viewers according to their user preferences as well as demographic information .

Dynamic processing of channel content requires selective ranking of channels and this can be done using user demographic, user explicit preferences, historically watched channels, external information on various programs running on different channels, users internet social activity, users internet browsing patterns.

Proposed article uses following methodology to come up with selective ranking of the channels


1. Build User profile: Build the user's profile by compiling the user's interests, Channel viewing pattern history and Social media information.

•This channel viewing pattern history is retrieved from the Television broadcasters or the set-top box equipment providers. This viewing pattern would include the channel type being watched, program in the channel etc. during various time periods. One can also collect review information on the

past programs that the user has watched.

•Modern set-top boxes (even televisions) now include social media apps by which the viewers interact to the internet. This user's interaction with the apps of the set-top box (or television) could also provide information about the user - like gender, age range etc.

•If set-top boxes do not have these apps, the social media information (twitter, facebook ids for example) from the users can be taken to get the information at the backend server.


2. Build Channel profile

•Build the channels profile by taking into consideration, the channel content, genre, content maturity, content viewership as well as the multi-modal information of the content .

•Multi-modal analysis of the channel content uses Image Processing (for

person, object detection), Video processing (for frame and scene segmentation), Audio processing (speech-to-text), OCR and Text processing. •Entity, topic and relation extraction can be done using existing state-of-the-art algorithms in computer vision; such as Natural Language Processing for entity extraction, Information retrieval for Index & Search, Machine Learning ranking.

1


Page 02 of 3


3. Collaboration and Content based filtering

•In addition to the above, the viewing pattern of different users at different time periods, such as themes, content type etc. is gathered from the television broadcasters and set-top box companies.

•...