In contrast to traditional social network services (SNSs), event-based social networks determine close friendships (CFs) of users who share experiences and emotions with candidate friends in offline events. However, we could not provide feedback to cyberspace regarding the place, time, and target of a user realizing friendship since there is no technique for conveniently measuring the evocation of friendship during offline events. In this research, we propose a method of estimating the evocation of friendship using the similarity in the pulse rate variabilities (PRVs) of users when empathy is evoked between them. The user can be made aware of friendship estimated automatically through machine learning by wearing a wristwatch-type pulsimeter. CFs are more likely to evoke empathy than superficial friendships (SFs). To demonstrate the usefulness of this method, we conducted an experiment assuming an event where a group of four people are enjoying their time in an amusement park. From the experimental results, we showed that the similarity of the PRVs in CFs is greater than that in SFs when the favorability rating is high and the users like each other. Moreover, we showed that the proposed method estimated the evocation of friendship during the attraction experience with an f-measure of 0.74 at maximum and during an offline event with a mean f-measure of 0.78. The results showed the usefulness and effectiveness of this method.