Inferring the Public Agenda from Implicit Query Data

Traditionally, implicit feedback measures are used to evaluate the performance of a particular information retrieval system. This research instead takes a distinctly applied approach to the use of implicit feedback, and extends the inference from aggregate query data to the social and political sciences. Using the three months prior to the 2008 election as a test scenario, the analysis here assesses daily fluctuations in search coverage of candidates and issues as predicted by the amount of news coverage, proximity to election day, and public opinion poll ratings of the candidates. Findings indicate that aggregate shifts in topical search queries may in fact be a useful, inexpensive indicator of political interest.