Identifying real-life complex task names with task-intrinsic entities from microblogs

Abstract

Recently, users who search on the web are targeting to more complex tasks due to the explosive growth of web usage. To accomplish a complex task, users may need to obtain information of various entities. For example, a user who wants to travel to Beijing, should book a flight, reserve a hotel room, and survey a Beijing map. A complex task thus needs to submit several queries in order to seeking each of entities. Understanding complex tasks can allow a search engine to suggest related entities and help users explicitly assign their ongoing tasks. © 2014 Association for Computational Linguistics.