A Proactive Recommendation System for Writing in the Internet Age

With the use of the computers, the task of writing is intertwined with the task of searching for information that can be relevant for the document that is being written, however very little research has been done to understand how the two tasks intertwine. In this paper we present an initial attempt to develop a model of writing and information seeking with computers and to develop helpful software that can improve the quality of the information searched and the written paper. Proactive Recommendation System (PRS) can relieve authors from explicit searching by means of automatically searching, retrieving and recommending information relevant to the text currently being written, and therefore PRS can be helpful to writers. However it is also possible that there are some moments during writing in which presenting proactive information can be an interruption rather than a help. In our research, we have used the PRS IntelliGent to investigate its impact in the different stages of writing. We found that when IntelliGent offers relevant information the time to task completion is shorter and the quality of the written product increases compared with the control situations in which writers have to look actively for information. We discuss these findings in the context of developing models and tools that integrate searching and writing processes when using computers as the writing environment.