Article Preview
TopMost of the conflict resolving systems use Collaborative Filtering or Content-based methods or hybrid conflict resolving methods to predict new items of interest for a user (Herlocker, Konstan, & Ried, 2002; Feng, Xian, & Feng, 2004; Rong & Bin, 2007). The system called Tapestry is often associated with the genesis of computer-based conflict resolving, recommendation, and collaborative filtering systems. In Tapestry (Goldberg et al., 1992), users were able to annotate documents with arbitrary text comments and other users could then query based on the comments of other users. The key attribute of this system is that it allowed recommendations to be generated based on a synthesis of the input from many other users. Making recommendations based on the opinions of like minded users rather than filtering items based on content has become known as collaborative filtering. The collaborative filtering paradigm which began with Tapestry was later automated in a number of projects (Resnick & Varian, 1997; Balabanovic & Shoham, 1997; Karypis, 2001; Herlocker et al., 2004; Wang et al., 2007).
The main advantage of collaborative filtering is the ability to make serendipitous recommendations (Herlocker et al., 2004). Most systems use a notion of inter –user distance and thus can define “neighbors” for a user. If an item of a particular genre is highly preferred by user’s neighbor, then that item could be recommended even if the recommendee has no previous experience with items of that genre.