Recommender-based enhancement of discovery in Geoportals

In many cases web search engines like Google are still used for discovery of geographic base information. This can be explained by the fact that existing approaches for Geo-information retrieval still face significant challenges. Discovery in currently available Geoportals is usually restricted to text-based search based on keywords, title and abstract as well as applying spatial and temporal filters. Furthermore, user context as well as search results of other users are not incorporated. In order to improve the quality of search results we propose to extend the suitable searching matches in Geoportals with user behaviour and to present them as non-directly linked recommendations like in e.g. Amazon's “Customers Who Bought This Item Also Bought” approach. As shown in the proof-of-concept EU FP7 EnerGEO Geoportal, it guarantees results that are not in the data itself but rather derived from the context of other users’ searches and views.