In this work, we apply a clustering technique to integrate the contents of items into the item-based collaborative filtering framework. The group rating information that is obtained from the clustering result provides a way to introduce content information into collaborative recommendation and solves the cold start problem. Extensive experiments have been conducted on MovieLens data to analyze the characteristics of our technique. The results show that our approach contributes to the improvement of prediction quality of the item-based collaborative filtering, especially for the cold start problem.