Implicit relevance feedback for context-aware information retrieval in UbiLearning environments

Abstract

Ubiquitous Learning (UbiLearning) environments heavily employ mobile devices to empower users with mobility and tooling support to learn anytime, anywhere. Introducing mobile devices in educational settings imposes constraints on search behavior due to limited resources on these devices such as small screens and restricted input functionalities. Targeting this scenario, this paper proposes an architecture for implicit relevance feedback which considers users' work context to expand search queries in order to satisfy information needs with greater precision. Evaluation results conducted over a local test collection enriched with contextual features reveal improvements on average precision when compared to a pseudo relevance feedback baseline.