HPM: A Hybrid Model for User’s Behavior Prediction Based on N-Gram Parsing and Access Logs

Algorithm 3

BipartiteGraphGen.

	Input: processed logs (PL)
	Output: N-gram associated click-graph (NC-graph)
	Begin
(1)	Read (PL);
(2)	Q ← Read queries from PL;
(3)	U ← Read URLs from PL;
(4)	Calculate click count Cq,u for each pair using PL;
(5)	C-graph ← create an edge between with label ;
(6)	For each query do
(7)	Nq ← Parser (q); //parsing of query into N-grams
(8)	NC-graph ← Create an edge between
(9)	EndFor
(10)	Return (NC-graph);
	End