An overview on text coherence methods

Abstract

The increasing availability of texts generated in many aria and online information has necessitated intensive research in the area of automatic text coherence identification within the Natural Language Processing (NLP) community. Over the past two decades, the problem has been addressed from many different perspectives, in varying domains and using various paradigms such as text summarization, text simplification, text generation and machine translation. This survey intends to investigate some of the most relevant approaches both in the areas of semantic and syntactic text coherence recognition methods, giving special emphasis to empirical methods and syntactic techniques. Special attention is devoted to categorize and classify of proposed methods, as future research on coherence and cohesion of texts is strongly dependent on progress in this area.