Since the publication of the first edition of the Oxford Handbook of Computational Linguistics, the landscape of Natural Language Processing (NLP) has changed extensively and rapidly. Whereas the first edition certainly served the community well, now it is time for this continuously growing community—which encompasses students, lecturers researchers and software developers—to have a new and up-to-date reference book available for their studies, teaching, and research and development.
While until recently there was a widespread view that machines could only achieve incremental advances in the ways language is understood, produced, and generally processed, and that automatic processing has a long way to go, the future now holds great promise due to recent developments in the field of Computational Linguistics. The employment of Deep Learning techniques for various tasks and applications has accounted for significant improvement in performance, and has radically changed the landscape of the state of the art of NLP.
When the first edition of the Oxford Handbook of Computational Linguistics was published, NLP did not have the significant commercial importance that it does now. While many successful companies strategically invest in NLP and develop various NLP-based products such as voice assistants, machine translation programs, and search engines, we now also see smaller businesses specializing in NLP technologies. It is worth noting that commercial NLP has clearly benefited from the Deep Learning revolution.
Computational Linguistics applications nowadays are not only proving indispensable for research and in various industrial settings, including customer service and marketing intelligence, but also have a significant impact on (and are used to the benefit of) society. The list of examples is long and includes healthcare (e.g. clinical documentation and provision of communication assistance to people with disabilities), education (e.g. generation and marking of exams), and a variety of services for citizens.
The second edition is not simply an updated version of the first; it is a significantly larger and substantially revised volume which covers many recent developments. The new edition also represents a comprehensive reference source for all those interested in Computational Linguistics, from beginners who would like to familiarize themselves with the key areas of this field to established researchers who would like to acquaint themselves with the latest work in a specific area. The Handbook is organized thematically in four parts. Part I (‘Linguistic Fundamentals’) introduces disciplines which linguistically underpin Computational Linguistics. Part II (‘Computational Fundamentals: Methods and Resources’) provides an overview of the models, methods, and approaches as well as resources that are employed for different NLP tasks and applications. Part III (‘Language Processing Tasks’) presents typical tasks associated with the automatic analysis or production of language. Finally, Part IV (‘NLP Applications’) offers surveys of the most popular NLP applications.
When designing the contents of the Handbook, I wanted it to be perceived as a coherent whole and not as a collection of separate contributions. Following the approach adopted in the first edition, I took on a rather proactive editorial role by seeking to enhance the coherence of this volume through cross-referencing and maintaining a consistent structure and style throughout. I have also included a glossary of significant size, compiled with the help of the authors and mainly with the help of Emma Franklin, in the hope that this will be a useful resource for students in the field.
This comprehensive volume is intended for a diverse readership, including university academics and students, researchers in universities and industry, company managers and software engineers, computer scientists, linguists, language specialists, translators, interpreters, lexicographers, and all those interested in Computational Linguistics and NLP.
I would like to thank the many people who contributed to the completion of this time-consuming, labour-intensive, and (on the whole) very challenging project. First, I gratefully acknowledge Emma Franklin. Over the past ten years, she has assisted me with unreserved enthusiasm and professionalism. Without her, this project could have scarcely been completed. Thank you, Emma!
I would like to thank all contributors—leading lights, well-known names, and young rising stars in their fields—for their high-quality contributions. A great debt of gratitude is also owed to the leading scholars who gave their time to review the chapters in this volume. Many thanks to Sophia Ananiadou, John Bateman, Johan Bos, Lynne Bowker, Fabio Brugnara, Claire Cardie, Yejin Choi, Kenneth Church, Noa Cruz Díaz, Robert Dale, Mary Dalrymple, Iustin Dornescu, Raquel Fernandez Rovira, Joey Frazee, Joanna Gough, Le An Ha, Stephanie W. Haas, Ed Hovy, Mans Hulden, Staffan Larsson, Claudia Leacock, Marie-Claude L’Homme, Inderjeet Mani, Rada Mihalcea, Alessandro Moschitti, Philippe Muller, Roberto Navigli, Mark Jan Nederhof, Joakim Nivre, Michael Oakes, Constantin Orasan, Paul Rayson, Paolo Rosso, Mark Shuttleworth, Khalil Sima’an, Sanja Stajner, Mark Stevenson, Liling Tan, Shiva Taslimipoor, Dan Tufis, Anssi Yli-Yurä, Fabio Zanzotto, and Michael Zock.
I am indebted to Oxford University Press (OUP), who have supported me throughout this challenging project. I would like to explicitly mention Julia Steer, the OUP lead on this project and my main contact at OUP, who was always ready to help; Brad Rosenkrantz, for intervening expediently whenever there was a typesetting issue; and Vicki Sunter and Laura Heston, who similarly provided me with excellent support. My deep gratitude goes to the late John Davey, who will be remembered for inspiring this big project; it was John who introduced me to Julia.
I would like to say a big ‘thank you’ to many colleagues for their valuable help and/or advice. The list includes, but is not limited to, Natalia Konstantinova, Constantin Orasan, Ed Hovy, Robert Dale, Sebastian Padó, Dimitar Kazakov, Omid Rohanian, Shiva Taslimipoor, Victoria Yaneva, Le An Ha, Anna Iankovskaya, Sara Moze, Patrick Hanks, Payal Khullar, Jessica López Espejel, Rut Gutierrez Florido, Carmen de Vos Martin, Erin Stokes, April Harper, Burcu Can, Frédéric Blain, Tharindu Ranasinghe, Dinara Akmurzina, Ana Isabel Cespedosa Vázquez, Parthena Charalampidou, Marie Escribe, Darya Filippova, Dinara Gimadi, Milica Ikonić Nešić, Alfiya Khabibullina, Lilith Karathian, Shaifali Khulbe, Gabriela Llull, Natalia Sugrobova, Simona Ignatova, Rocío Caro, Sabi Bulent, Slavi Slavov, and Sara Villar Zafra.
The Oxford Handbook of Computational Linguistics (2nd edn)
Annotate
