AI-Powered Personalized Media Suggestion System Share: Bit Adventure Score Awaiting client review n/a Date Published 3 July 2023 Reading Time 2-Minute Read An AI-based suggestion system that is a revolutionary new way to personalize media consumption. The Challenge Our client approached us with the challenge of developing a system that goes beyond traditional methods of recommending content based on past views. They wanted a system that could take into account external factors such as time, day of the week, events in a user’s calendar, and feedback from a user. The Solution Our team rose to the challenge and developed a system that continuously gathers media consumption information from all of a user’s Apple devices. Such information can include the podcasts a user has been listening to, as well as TV shows or films they have watched. The system builds vectors based on this information, as well as user preferences that were added during the initial setup, allowing for highly personalized and accurate content suggestions. By comparing the vectors of different users, the system can find matches and suggest content that people with similar tastes are likely to enjoy. To achieve this, we used advanced data manipulation tools, such as pandas and machine learning techniques, to train the suggestion mechanism. We also leveraged the Pyatv library to facilitate communication with Apple devices. The system also incorporated user feedback, allowing it to adapt and improve over time. The Result This AI-based suggestion system offers users a new and exciting way to discover and consume media, providing a truly personalized and dynamic experience. It can be integrated into any streaming or media platform and customized to meet a client’s specific needs.