Towards better understanding of academic search

Academics have relied heavily on search engines to identify and locate research manuscripts that are related to their research areas. Many of the early information retrieval systems and technologies were developed while catering for librarians to help them sift through books and proceedings, followed by recent online academic search engines such as Google Scholar and Microsoft Academic Search. In spite of their popularity among academics and importance to academia, the usage, query behaviors, and retrieval models for academic search engines have not been well studied. To this end, we study the distribution of queries that are received by an academic search engine. Furthermore, we delve deeper into academic search queries and classify them into navigational and informational queries. This work introduces a definition for navigational queries in academic search engines under which a query is considered navigational if the user is searching for a specific paper or document. We describe multiple facets of navigational academic queries, and introduce a machine learning approach with a set of features to identify such queries.