A Method for Recommending People who Provide \Unexpected Information" Based on Activities on Social Media

Now we can find expected information easily with well-developed search engines and recommendation engines. However, there is no established method which can find “Unexpected Information.” “Unexpected Information” is information such that we don’t clearly want but once we get it, it stimulates our curiosity. On the other hand, in our daily life, we have conversations with some people and sometimes get “Unexpected Information” from them. In this research, we propose a method for recommending people who provide “Unexpected Information” on social media. The method predicts new people who provide “Unexpected Information” based on the other users’ activities. We have implemented the method on Twitter, and evaluated the effectivity comparing to other similar systems. 1. は じ め に