This book discusses the development of knowledge acquisition techniques in the field of text-based learning and discusses their applications in the context of education and research.
This book gives a comprehensive view of state-of-the-art techniques that are used to build spoken dialogue systems and presents dialogue modelling and system development issues relevant in both academic and industrial environments and also discusses requirements and challenges for advanced interaction management and future research.
It is argued that the single global label with which RTE examples are annotated is insufficient to effectively evaluate RTE system performance; to promote research on smaller, related NLP tasks, it is believed more detailed annotation and evaluation are needed.
This paper describes the Illinois-Columbia system that participated in the CoNLL-2014 shared task and describes the novel aspects of the system, which ranked second on the original annotations and first on the revised annotations.
Curator, an NLP management framework designed to address some common problems and inefficiencies associated with building NLP process pipelines; and Edison, a NLP data structure library in Java that provides streamlined interactions with Curator and offers a range of useful supporting functionality.
This paper presents a principled approach to this problem that builds on inducing representations of text snippets into a hierarchical knowledge representation along with a sound optimization-based inferential mechanism that makes use of it to decide semantic entailment.
A number of researchers appear to have converged on some defining characteristics of the problem, and on characteristics of practical approaches to solving it, which are an exciting time to be working in this area.
The underlying approach is described, which relates to the previous work in this area, and an improvement to the earlier method, error inflation, is proposed, which results in significant gains in performance.
Improvements of Illinois-Coref system from last year are presented, focusing on improving mention detection and pronoun coreference resolution, and a new learning protocol is presented.
The University of Illinois system that participated in Helping The authors' Own (HOO), a shared task in text correction, is described, which ranked first in all three evaluation metrics (Detection, Recognition and Correction) among six participating teams.