I am pleased to announce that I will be at COLING’18 giving a tutorial on Negation and Speculation Detection in NLP.

Currently, negation and speculation detection is an emergent task inNLP. In recent years, several challenges and shared tasks have included the extraction of these language forms, such as the BioNLP’09 Shared Task 3, the CoNLL-2010 Shared Task, the i2b2 NLP Challenge, the SEM 2012 Shared Task and the ShARe/CLEF eHealth Evaluation Lab 2014 Task 2.

Detecting uncertain and negative assertions is relevant in a wide range of applications such as information extraction, machine translation, sentiment analysis, paraphrasing and recognising textual entailment. In addition, a special issue of Computational Linguistics has been published on negation and speculation and a book by John Benjamins Publishing Company about this topic will be published this year. This shows how computational linguistics has started to take into account the subjective aspects of language.

This tutorial is motivated by the fact that this is an emerging topic relevant for the computational linguistic community which has not previously been covered in any ACL/COLING/EMNLP/NAACL related tutorial. It would be useful for students of NLP subjects who are interested in understanding this problem in more depth as well as for researchers with an interest in these phenomena in order to improve performance in other NLP tasks. The main advantage of this tutorial is that it will not only provide an overview of the state of the art in negation and speculation recognition, but that it also introduces newly developed data sets and scripts.

The tutorial will divided into 6 different parts. A more detailed outline is described below.

  • Part 1 begins with an introduction to the definition of negation and speculation from different perspectives. After briefly providing basic notions necessary to address the problem, this part addresses the importance of processing these language forms.
  • Sections 2 and 3 go into detail about the concepts of negation and speculation, including a classification of the different types of each. In addition, the related work is analysed, providing a description and comparison of the most relevant negation and speculation detection systems found in the literature and highlighting the different approaches followed by the authors in order to solve the problem.
  • Part 4 is an in-depth description of the applications where information about negation and speculation has proven to be useful such as text mining, sentiment analysis, recognising textual entailment, machine translation and information retrieval.
  • Section 5 presents a set of relevant resources for any researcher or developer interested in the problem. It also includes information about available scripts for evaluation.
  • Finally, part 6 discussed the possibilities for future work and the open challenges in each domain.

I look forward to seeing you in Santa Fe (New Mexico, USA) 🙂