Article Preview
TopThe current generation of recommendation methods can be classified into: Content-based Recommendations, Collaborative Recommendations, Knowledge-based Recommendations, Demographic Recommendations and Hybrid Recommendation.
Collaborative filtering (CF) is a common Web technique for generating personalized recommendations. Examples of its use include Amazon, iTunes, Netflix, LastFM, StumbleUpon, and Delicious, which are websites offering various multimedia services (Hoppe, 2009).While being the most successful recommendation technique to date (Resnick et al., 1994), it has difficulty dealing with new users. This is because there needs to be some correlation or better still, relationship among existing users. On the other hand, pure content-based recommendations ignore the preferences of other users (Schein, Popescul, & Ungar, 2002). Therefore, items that contain terms that relate well to the search and which are statistically less likely to be common are suggested first (Adomavicius & Tuzhilin, 2005). However, content-based filtering techniques are known to have issues involving limited content data and information, and concentration on bounded data. Knowledge-based recommendation gain leverage on recommendation tasks by using explicitmodels of both the user of the system and the products being recommended (Towle & Quinn, 2000). Demographic Recommendations categorizes the user based on personal attributes and makes recommendations based on demographic classes, e.g. college students, teenagers, women, men, etc. Grundy’s system (Guttman et al., 1998) is an example of a demographic filtering recommender system which recommended books based on personal information gathered through an interactive dialogue. A Hybrid recommendation system combines two or more recommendation techniques to gain better system optimization and fewer of the weaknesses of any individual ones(Prasad & Kumari, 2012). Such combination could be: Weighted, Switched, Featured, Cascaded, Augmented, Levelled or Mixed.