Vayianos Pertsas, Panos Constantopoulos and Ion Androutsopoulos.
Abstract: We address the automatic extraction from publications of two key concepts for representing research processes: the concept of research activity and the sequence relation between successive activities. These representations are driven by the Scholarly Ontology (SO), specifically conceived for documenting research processes. Unlike usual named entity extraction tasks, we are facing textual descriptions of activities of widely variable length, while pairs of successive activities often span different sentences. We developed and experimented with several sliding window classifiers using Logistic Regression, SVMs, and Random Forests, as well as a two-stage pipeline classifier. Our classifiers employ task-specific features, as well as word, part-of-speech and dependency embeddings, engineered to exploit distinctive traits of research publication written in English. The extracted activities and sequences are associated with other relevant information from publication metadata and stored as RDF triples in a knowledge base. Evaluation on datasets from three disciplines, Digital Humanities, Bioinformatics, and Medicine, shows very promising performance.
Keywords: Ontology Population; Information Extraction; Machine Learning Methodologies; Linked Data