Title: MUDROD: Improving Ocean Data Discovery Through Enhanced Searching Relevance
Presenting Author: Chaowei Phil Yang
Organization: George Mason University
Co-Author(s): Yongyao Jiang, Yun Li, Kai Liu, Edward M Armstrong, Thomas Huang, David Moroni, and Chris Finch, Lewis Mcgibbney

Abstract:
MUDROD (Mining and Utilizing Dataset Relevancy from Oceanographic Datasets to Improve Data Discovery and Access) is a search relevance project funded by NASA AIST (NNX15AM85G). The objectives are to 1) analyze web logs to discover the semantic relationships between datasets and user queries, 2) construct a knowledge base by combining an existing ontology and user browsing patterns, 3) and improve data discovery by providing better ranking results, recommendations, and ontology-based navigation. The specific functionalities of the MUDROD system include a) web log ingesting (Li et at. 2017); b) session reconstruction (Jiang et at. 2016a); c) vocabulary semantic relationship extraction (Jiang et al. 2016b); d) ontology navigation; e) search results ranking (Jiang et al. 2017); and f) recommendation (Li et al. 2017). Through the past two years’ investigation, a team from GMU and the JPL Physical Oceanography Distributed Active Archive Center (PO.DAAC) has built the MUDROD prototype that is being integrated into a new technology domain called “PO.DAAC Labs” to improve ocean data discovery.