Metadata Topic Harmonization and Semantic Search for Linked-Data-Driven Geoportals: A Case Study Using ArcGIS Online

What is it about?

This research addresses two important challenges in geoportals, namely the topic heterogeneity brought by multiple metadata standards and the lack of established semantic search in Linked-Data-driven geoportals. To harmonize the metadata topics, we designed a machine learning workflow based on Labeled Latent Dirichlet Allocation (LLDA), and employed the standardized metadata from Data.gov as the training data. With respect to semantic search, we construct thematic and geographic matching features from the textual metadata descriptions, and train a regression model via a human participants experiment. We evaluate our methods by examining their performances in addressing the two issues. Finally, we implement a semantics enabled and Linked Data driven prototypical geoportal using a sample dataset from Esri's ArcGIS Online.

Featured Image

Why is it important?

An essential goal of geoportals is to facilitate the discovery of the available resources. Such a process relies heavily on the quality of metadata. While multiple metadata standards have been established, data contributors may adopt different standards (e.g., CSDGM and ISO 19115) when sharing their data via the same geoportal. While tools have been developed to harmonize the heterogeneous metadata, manual efforts are required to assign suitable metadata topics. The machine learning workflow we designed in this work automatizes this labor-intensive process and achieves a precision of 0.8 and a recall of 0.69. With the fast development of the Semantic Web technologies, there is a rise of Linked-Data-driven portals. Although these novel portals open up new ways to organize metadata and retrieve resources, they lack effective semantic search methods. The search method proposed in this work integrates techniques in named entity recognition (NER), latent semantic analysis (LSA), and semantic expansion. We compared its performance with human judgements, and achieves a Pearson's coefficient as 0.72.