The Effects of Time on Query Flow Graph-based Models for Query Suggestion
Ranieri Baraglia, Franco Maria Nardini, Carlos Castillo, Raffaele Perego, Debora Donato, Fabrizio Silvestri · 2010
A recent query-log mining approach for query recommenda-tion is based on Query Flow Graphs, a markov-chain rep-resentation of the query reformulation process followed by users of Web Search Engines trying to satisfy their informa-tion needs. In this paper we aim at extending this model by providing methods for dealing with evolving data. In fact, users ’ interests change over time, and the knowledge extracted from query logs may suffer an aging effect as new interesting topics appear. Starting from this observation val-idated experimentally, we introduce a novel algorithm for updating an existing query flow graph. The proposed so-lution allows the recommendation model to be kept always updated without reconstructing it from scratch every time, by incrementally merging efficiently the past and present data.