Recommender systems are used for providing personalized recommendations based on the user profile and previous behaviour. Recommender systems such as Amazon, Netflix, and YouTube are widely used in the Internet Industry.
Collaborative filtering and content based filtering are the prime approaches in providing recommendation to the users. Both of them are the best applicable in specific scenarios because of their respective properties.
Popular sites using recommender systems
Types of algorithms used in Recommender System
Collaborative filtering:-It recommends items based on the similarity measures between users and items. The system recommends those items that are preferred by a similar category of users.
Collaborative filtering has many advantages
1. It is content-independent
2. In CF people make explicit ratings so real quality assessment of items is done.
3. It provides effective recommendations because it is based on user’s similarity rather than item’s similarity.
Content based filtering:-It is based on the profile of the user’s preference and the item’s description. In CBF, to describe items we use keywords apart from user’s profile to indicate users preferred likes or dislikes. In other words CBF algorithm recommend items or similar to those items that were liked in the past. It examines previously rated items and recommends the best matching item.
Demographic Filtering:In demographic filtering recommendations is established on a demographic profile of the user. Here recommendation is based on the information provided by the user is considered to be similar according to demographic parameter such as nationality, age, gender etc.
Hybrid filtering:The hybrid filtering is a combination of more than one filtering approach. The hybrid filtering approach is introduced to overcome some common problem that are associated with above filtering approaches such as cold start problem, overspecialization problem and sparsity problem. Another motive behind the implementation of hybrid filtering is to improve the accuracy and efficiency of recommendation process.
Major challenges in recommender system
- Data sparsity
- Vulnerability to attacks
The following research can be followed for better understanding:
“Survey on Collaborative Filtering, Content-based Filtering and Hybrid Recommendation System”