Recommender system

From PangaWiki
Jump to: navigation, search
PANGAEA recommender system scheme

Recommender system improves the discovery of research datasets in the PANGAEA data portal. It uses the metadata of datasets (e.g., title, abstract, related publication, authors, topics, projects, devices, campaign, location, time) and the user interactions (e.g., data search and download) extracted from the server logs to deliver two types of recommendations, i.e., similar datasets and users who are interested in this item are also interested in...

Building a data recommender on top of Elasticsearch enhances the scalability and maintainability of the recommender system.