Web information analysis and searching

Link Pattern Modeling - II

nLongest Repeating Subsequence Model [Pitkow 99]

nSurfing paths can be represented as n-grams <X1, X2,…Xn> to indicate sequences of page clicks by a population of users visiting a web site

nFind Longest repeating subsequence

nE.g., 135 in 15135213544

nMatch the performance accuracy of the one-hop Markov model while reducing the complexity by nearly 33%


	Use Markov chains model and eigen-vector decomposition techniques to make Link prediction and path analysis and to generate hubs
	Sarukkai chose Markov chains, which had been enormously successful in sequence matching/generation, as mathematical model.

	Sariukkai mentioned that the key of navigation lied in ‘personalization’, and he believed that this approach would lead to a satisfactory solution to navigating the huge World Wide Web.