Introduction
Information retrieval may be considered as the quest for content relevant to a given information need. This is typically expressed by query terms submitted to a search engine. However, the performance of existing search engines has been shown to be far from users’ satisfaction in their precision and recall. For example, a search engine often returns thousands of results in response to a query, but the results contain information irrelevant to the user's information needs. This phenomenon can be attributed to a few causes. First, errors may be introduced during query formulation. Such errors may include lexical/spelling errors and syntactic errors in which the query is expressed in a form not compatible with the query language (Willson and Given, 2010). This could be because users are unfamiliar with the syntax and semantics associated with the particular search interface (Belkin, 2000). Second, users may not be able to adequately express their information needs as query terms. Here, there may be a disparity between users’ query terms that express their information needs and those used in the system to describe the information sources (Furnas et al., 1987). Additionally, users could fail to choose terms at a proper conceptual level to specify their information needs.
Ensuring that information needs are met efficiently and effectively is critical, given the increasing popularity of social computing applications, which has empowered users to create and share content on the web. Such user-generated content may include text (e.g. blogs, wikis), multimedia (e.g. YouTube) and even organizational/navigational structures providing personalized access to web content (e.g. social tags). The result is that people have now come to depend more on the web in searching for information. However, the amount of information and its growth is a double-edged sword, due to the problem of information overload, leading to a situation where users are swamped with too much information, resulting in the difficulty of sifting through the materials in search of relevant content.
Some may argue that the above problems may be less complex in digital libraries, especially when a collection is more focused and, typically, there are stakeholders who ensure the quality and the relevancy of the collection.