Searching for Movies: An Exploratory Analysis of Movie-related Information Needs

Despite a surge in popularity of work on casual leisure search, some leisure domains are still relatively underrepresented. Movies are good example of such a domain, which is peculiar given the popularity of movie-centered websites and discovery services such as IMDB, RottenTomatoes, and Netflix. In this paper, we present an exploratory analysis of IMDB movie discussion threads that contain requests for movies to watch. Through emergent coding we produce a taxonomy of relevance aspects for movie search and selection. Our analysis shows that topical aspects, such as content, metadata, and known-item search, are important for movie selection practices. Other requests focus more on recommendation and feature many subjective relevance aspects, such as the tone of a movie or its intended audience. This suggests efficient access to movies is likely to require different information access paradigms to satisfy all the movie-related information needs expressed in the threads.