Serendipity Workshop

Article: http://www.sciencedirect.com/science/article/pii/S221083271400026X

Automated web usage data mining and recommendation system using K-Nearest Neighbor (KNN) classification method

Used on-line to identify users click stream data, matching it to a particular user group and recommend a tailored browsing option.

K-Nearest Neighbor:

learning by analogy → learnt by comparing a test tuple with a set of similar (close, wrt a distance) training tuples ⇒ classified based on the class of their $k$-closest neighbors

• Lazy learner: stores the given training tuples and waits until given a test tuple, then performs generalization

• instance based learner: does less work when presented with training tuples and more work during classification and prediction ⟹ computational expensive (unlike eager learners, constructing classification model before receiving the test tuple)

Distance used: Euclidian distance