We conduct large-scale experiments to investigate optimal features for classification of verbs in biomedical texts. We introduce a range of feature sets and associated extraction ...
Background: Although the UniProt KnowledgeBase is not a medical-oriented database, it contains information on more than 2,000 human proteins involved in pathologies. However, thes...
Current approaches to develop information extraction (IE) programs have largely focused on producing precise IE results. As such, they suffer from three major limitations. First, ...
Warren Shen, Pedro DeRose, Robert McCann, AnHai Do...
Data mining allows the exploration of sequences of phenomena, whereas one usually tends to focus on isolated phenomena or on the relation between two phenomena. It offers invaluab...
Extract-Transform-Load (ETL) workflows are data centric workflows responsible for transferring, cleaning, and loading data from their respective sources to the warehouse. Previous ...