A hybrid semantic method for enhancing movie recommendations

With the abundance and similarity of data on the World Wide Web (WWW), search and recommendation tasks have become harder jobs. This leads researchers to intensify their efforts to discover new ways for searching and recommendations. One of the most modern methods to enhance the accuracy of search and recommendations is by exploiting semantic information and relations that are included in or have been extracted through reasoning from the data. This paper describes a hybrid recommendation method for movies that exploits semantic relations and hidden associations from multiple resources (ontologies such as MO and Movie night) referring to movies to recommend movies to the user that are relevant to his/her preferences. An offline experiment that has been conducted on a set of the UK and USA movies showed that our suggested method has overcome other state of the art methods in this area.