The large number of channels available on television can make the process of searching for the appropriate TV channel cumbersome and time consuming. This disclosure describes techniques to make real-time personalized recommendations for live TV programming, implemented with the user’s permission. The recommendations are based on the application of a trained machine learning model. Upon being turned on, the TV is automatically tuned to a channel showing the program that is identified as being most likely to be watched by the user, chosen from all programs available at that time. Further, the user is shown a dynamically updated personalized shortlist of other channels that are broadcasting programs of likely interest to the user. Further, a notification can be shown to alert the user when a channel other than the one currently being watched starts broadcasting a program the user is likely to find of greater interest that the one on the current channel.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Calvo, Marcos and Pastor, Joan, "Dynamically updated personalized recommendations for broadcast television", Technical Disclosure Commons, (September 04, 2019)