Recommending when to switch TV channels to maximise the number of seen interesting live events

From Master Projects
Jump to: navigation, search


About Recommending when to switch TV channels to maximise the number of seen interesting live events

  • This project has been fulfilled.
  • This project fits in the following Bachelor programs: {{#arraymap:|, |xXx|bachelorproject within::xXx|,}}
  • This project fits in the following masterareas: {{#arraymap:Technical Artificial Intelligence, Computational Intelligence and Selforganisation, Information Sciences|, |xXx|project within::xXx|,}}


Description

Let us imagine you sit in front of a TV, watching a football match of a certain championship (other, but comparable, situations are also accepted for the project). After 20 minutes, you see that your match stands 0 - 0, while there have been a number of goals scored in other matches being played in parallel to yours. Should you keep watching the same match in the hope that someone will score, or should you change channel? This is provided that you are interested in maximising the number of goals (or other live events) - this is the main assumption of the project. The project will involve studying the literature on event prediction, recommender systems, profiling, probabilistic models, and then designing a solution to the above described problem. Various levels of (user/channel/team) profiles, various costs of switching are also to be investigated, and corresponding recommender system versions are to be evaluated.