Recommendations platforms is a reasonably new concept that involves TV viewers being recommended specific shows based on their viewing preferences.
The idea behind recommendation engine technology is that by keeping viewers satisfied in an increasingly competitive market, revenues will be increased. The increase in the number of programmes available coupled with the rise in different methods to watch content means that the recommendations technology battle is only going to go on.
Some features of a recommendations engine
Content analysis
A content analysis will provide a detailed analysis on all programmes, extracting the relevant key tags and comparing the key tags between the different shows. This is used to produce relationships based on the content of the shows metadata. Mechanisms such as this can be entirely automated producing powerful results without the need for complex metadata schemes
User activity and preferences
Platforms can usually gather anonymous intelligence on the viewer’s behaviour. This information is relayed back into the engine, with different levels of importance placed upon things such as the preferences within email reminders that are sent to the viewer
Editorial
At some point, regardless of how powerful a recommendation engine is the viewer will want to make their own manual TV show recommendations. This may be to back a new programme or series, or to filter out a competitor’s show. The platform will normally include a simple interface to add and remove, and also modify the viewers automatic recommendations
Another way that recommendations are made is through the use of social media. The idea behind this is that people will watch shows that friends and family have suggested to them. This differs from other recommendation methods in that they are not generally relevant to the user who is receiving the recommendation other than based upon the precept of who has made the recommendation.
As competition to attract viewers continues to grow, more TV service providers will have to find better solutions to keep users from becoming disinterested and moving on to a competitor.
Tags: Engines, Recommendation