Linked Data brings the vision of exposing and interlinking datasets on the Web by using Semantic Web standards. This vision creates the potential of adding a structured data information layer on the Web which can be consumed by both humans and applications. Consuming Linked Data today, however, can be challenging. Linked Data brings a scenario where users may need to query/search over potentially thousands of highly heterogeneous datasets.
What is it?
Treo is a natural language based semantic search engine for Linked Data. The main goal behind Treo is to abstract data consumers from the representation of the datasets, allowing expressive schema-independent and natural language queries over Linked Datasets.
How it works?
Differently from traditional question answering (QA) systems, which focus on the provision of final answers in a natural language post-processed form, Treo returns a ranked list of highly semantically related results, targeting a semantic best-effort scenario (closer from the information retrieval approaches). The center of the approach relies on a query processing mechanism which focuses on a precise and comprehensive semantic matching between user natural language queries and Linked datasets. The semantic matching approach is based on distributional semantics, a data-driven semantic model based on statistical information extracted from large corpora such as Wikipedia. By relaxing the requirements of traditional QA systems and by focusing on the semantic matching problem using information retrieval and natural language processing techniques, Treo allows users to input unconstrained natural language queries to query structured data.
- André Freitas, PhD Student
- Dr Edward Curry, Researcher & Unit Leader
- Dr Sean O’Riain, Researcher & Unit Leader
Danilo S. Carvalho, Research Intern
Fabricio F. de Faria, Research Intern
- João G. Oliveira, Research Intern