An evaluation of page segment recommendation system using user's notes and N-Gram models

Web content searching is a daily activity of almost everyone. Often, it occurs several times a day. A number of people need to make sense out of a huge amount of Web pages in order to complete their jobs. Many others also have to rely on it. A number of research works in sensemaking have demonstrated the needs for supporting tools in Web content searching. In this paper, NorCost, a system that recommends relevant page segments, is proposed. The system emphasizes helping people to complete their sensemaking tasks without having to go through every detail of the Web page themselves as such tasks could takes long time to finish. The evaluation of NorCost is carried out to assess its accuracy as well as time taken to process the recommendation.