Tutorial: Computational modelling of metaphor

26 April, afternoon

Presenters: Ekaterina Shutova, Tony Veale

Tutorial contents

Metaphor processing is a rapidly growing area in NLP. The ubiquity of metaphor in language has been established in a number of corpus studies and the role it plays in human reasoning has been confirmed in psychological experiments. This makes metaphor an important research area for computational and cognitive linguistics, and its automatic identification and interpretation indispensable for any semantics­oriented NLP application.

Computational work on metaphor in NLP and AI ignited in the 1970s and gained momentum in the 1980s, providing a wealth of ideas on the form, structure and mechanisms of the phenomenon. The last decade has witnessed a technological leap in natural language computation, as manually crafted rules have gradually given way to more robust corpus­based statistical methods. This is also the case for metaphor research. In the recent years, the problem of metaphor modeling has been steadily gaining interest within the NLP community, with a growing number of approaches exploiting statistical
techniques. Compared to more traditional approaches based on hand­coded resources, these more recent methods boast of a wider coverage, as well as greater efficiency and robustness. However, even the statistical metaphor processing approaches largely focus on a limited domain or a subset of conceptual phenomena. At the same time, recent work on computational lexical semantics and lexical acquisition techniques, as well as a wide range of NLP methods applying machine learning to open­domain semantic tasks, opens many new avenues for creation of large­scale robust tools for the recognition and interpretation of metaphor.

The tutorial aims to:

  • introduce a CL audience to the main linguistic, conceptual and cognitive properties of metaphor;
  • cover the history of metaphor modelling and the state­of­the­art approaches to metaphor identification and interpretation
  • analyse the trends in computational metaphor research and compare different types of approaches, aiming to identify the most promising system features and techniques in metaphor modelling
  • discuss potential applications of metaphor processing in wider NLP
  • relate the problem of metaphor modelling to that of other types of figurative language

The tutorial is targeted both at NLP researchers who are new to metaphor and need a comprehensive overview of metaphor processing techniques and applications, as well as at scientists working in the field who want to stay up to date on the recent advances in metaphor research.


Ekaterina Shutova is a Postdoctoral Research Fellow at the International Computer Science Institute (ICSI) and the Institute for Cognitive and Brain Sciences (ICBS) at the University of California, Berkeley, USA. Her research is in the area of Natural Language Processing with a specific focus on metaphor and human creativity, and its computational and cognitive modeling. She is currently leading the new Metaphor Extraction research team at ICSI, the goal of which is to create robust and accurate tools that identify metaphorical expressions in unrestricted text using statistical methods. Previously, she was a Research Associate at DTAL and the Computer Laboratory, University of Cambridge, UK, where she worked on issues in computational lexical semantics. Ekaterina received her PhD in Computer Science from the University of Cambridge in 2011 and her doctoral dissertation concerned computational modeling of figurative language.

Tony Veale is an invited professor in the Division of Web Science and Technology at the Korean Advanced Institute of Science and Technology in Daejeon, South Korea, and a senior lecturer in the School of Computer Science and Informatics at University College Dublin. He leads the European Commission's international coordination action on the Promotion of the Scientific Exploration of Computational Creativity (PROSECCO), which aims to foster a greater understanding (via summer/autumn schools and code­camps) and a more robust computational modeling of creative tasks such as the generation of novel stories, poems and metaphors. He has researched the computational treatment of metaphor, similes, blending and irony for over two decades, both in academia and in industry, and has delivered courses on these topics at ESSLLI 2005, ESSLLI 2007 and at the 1st international Autumn School on Computational Creativity in Helsinki in 2011. He is the author of the 2012 monograph Exploding the Creativity Myth: The Computational Foundations of Linguistic Creativity.