Causality is a research field with roots in philosophy, psychology, physics and application domains in medicine and knowledge engineering, which focuses on cause-effect, or causal, relations between two events or actions, in which one (the cause), causes the other (the effect). This valuable information can be used in a number of natural language processing applications such as question answering, text summarization, decision support etc. While encyclopedic knowledge can be manually encoded into causal relations, in many other domains, causality is not explicit and must be inferred from data.
The Workshop on Computational Approaches to Causality in Language will provide a forum for presentation and discussion of innovative research on all aspects of recognition, representation and the use of causal information and its processing in NLP-centered applications. The workshop encourages researchers to submit papers (long (6-pages) and short (4-pages)) that report on accomplished and ongoing work related, but not restricted, to the following topics:
- annotation of causal information in text;
- knowledge engineering with causalities;
- reasoning with causal relations;
- question answering with causal questions;
- social media and behavioral causality;
- resources for causal information;
- extracting causal relations in law and legal texts;
- extracting causal relations in clinical texts;
- causal relations in natural language discourse;
- causality for planning in natural text generation and interactive game design.
In addition, the workshop sets the goal of initiating a discussion on computational approaches of recognition of causal relations in text which will lead to a shared task in the scope of Semantic Evaluation Exercises (SemEval).