Spatio-Temporal Evolution of Scientific Knowledge

In this work we take a first step towards the problem of integrating the content and the spatio-temporal aspects of the evolution of the (published) scientific knowledge. A lot of research has been invested in developing tools and search engines that will enable more efficient querying of relevant medical (and broader scientific) data from various perspectives, spanning from retrieval of similar documents/images to HCI-based flexible query-answering systems. Variety of methodologies have been developed, founded on knowledge-bases, statistics, semantic similarity, etc. and quite a few systems are available (e.g., Medline). Parallel to this, another body of research works has emerged over the past couple of decades, targeting the efficient management of mobility and spatio-temporal data. What motivates this work is the observation that fusing the data (and corresponding techniques) developed in these two broad research fields could enable novel categories of queries that can be used to investigate various evolving spatio-temporal relationships between particular scientific topics. We present a novel model and a formalization of this confluence, in what we call Knowledge-Evolution Trajectories (KET). We also provide a preliminary proof-of-concept implementation that enables answering novel categories of queries pertaining to KET data with a few initial observations regarding the impact of different data-representation approaches.