Dr. Annalina Caputo, Assistant Professor

Contact Details

Office: L2.48
School of Computing, Dublin City University Glasnevin Campus, Glasnevin, Dublin 9, Ireland
T:+353 (0) 700 8056
E:annalina dot caputo at dcu dot ie

Office Hours

Wednesday 14:00-15:00

Main Research Interests

  • Intelligent Information Access
  • Text Representation and Information Retrieval
    • Temporal Dynamics
    • Personalised Information Retrieval
    • Semantic IR
    • Quantum IR
    • Cross-Language
    • Question Answering
  • Natural Language Processing
    • Word Sense Disambiguation
    • Distributional semantic models
    • Entity Linking
    • Diachronic Analysis

The universe is a labyrinth made of labyrinths. Each leads to another. And wherever we cannot go ourselves, we reach with mathematics.

Stanislaw Lem, Fiasco

I'm an Assistant Professor in the School of Computing, Dublin City University. My research is in the areas of Natural Language Processing, Information Access and Retrieval, and Machine Learning. Since my Ph.D., I have been tackling problems related to text representation and its impact on intelligent information access. I focused on different ways of representing text through Word Sense Disambiguation, Distributional Semantics, and Entity Linking. Prior to this appointment, I was an EDGE COFUND Marie Skłodowska-Curie Action research fellow in the ADAPT centre at Trinity College Dublin in the domain of Temporal Aware Personalised Information Retrieval. As part of this fellowship, I was on secondment in Ryanair, collaborating on the application of the project’s algorithms to Ryanair’s huge user base.

Before joining ADAPT, I worked as a research fellow at University of Bari Aldo Moro within the SWAP (Semantic Web Access and Personalization) research group, which focuses on Machine Learning, Artificial Intelligence, and Natural Language Processing with practical application to Information Access Systems (Information Retrieval and Recommenders). My research activity concerned Entity Linking for Intelligent Information Access: the aim of that research was to provide a semantic representation of unstructured text through named entities and their relationships mentioned in the text content. This kind of representation would enable the intelligent reasoning, discovery and retrieval of relevant pieces of information as well as the suggestion of serendipitous items.

I obtained my Ph.D. in Computer Science at University of Bari Aldo Moro defending the thesis “Semantics and Information Retrieval: Models, Techniques and Applications”. Since my Ph.D. I have been working in the area of NLP and Information Retrieval by focusing on the text representation problem, which I tackled with both Word Sense Disambiguation and Distributional Semantics. To this respect, Entity Linking represents a further step toward the concept-based representation of textual content.

I have participated in different evaluation campaigns in the areas of Information Retrieval (CLEF 2008-2009, Kaggle Allen AI Science Challenge), Lexical Semantics (SemEval 2012, *SEM 2013, SemEval 2015) and Entity Scoring and Linking (WSDM Cup 2017, NEEL 2015). I co-organised two evaluation campaigns (Question Answering for Frequently Asked Questions and Named Entity rEcognition and Linking in Italian Tweets Tasks) at EVALITA 2016 and co-organized the 1st Workshop on Temporal Dynamics in Digital Libraries at the 21st International Conference on Theory and Practice of Digital Libraries.