Intention Modeling for Web Navigation

A novel global optimization method referred to as the multi-step dynamic n-gram model is proposed for predicting the user’s next intended action while he/she is surfing the Web. Unlike the traditional n-gram model, in which the predicted action is taken as the ultimate goal and is only determined by its previous n actions, our method predicts the next action that lies on the optimal path that leads to the ultimate goal. Experiments show that the prediction accuracy of our proposed method can achieve up to 3.65% (or about 11% relative) improvement to the traditional one-step n-gram model.