In this paper, we deal with a word-of-mouth recommender system with a human-friendly robot. First, we explain the background of conventional recommendation systems, and we discuss the availability of smart devices and robot partners for information support. Next, we describe the issues caused in the conventional recommendation method, and propose propose a word-of-mouth recommender system. Furthermore, we propose a preference extraction technique based on interactive evolutionary computation. Finally, we show preliminary experimental results and comparison results using the proposed method, and discuss the effectiveness of word-of-mouth recommender system.