Short Text Interpretation for User Classifications

Abstract

The texts with limited context are referred as short text. Understanding and Interpreting short texts are important to many applications, but challenges appear. The problem when handling with short text is that firstly, short texts do not always observe the syntax of a written language. As a result, traditional natural language processing tools, ranging from part-of-speech tagging to dependency parsing, cannot be easily applied. Secondly, short texts usually do not contain sufficient statistical signals to support many state-of-the-art approaches for text mining such as topic modelling. Thirdly, short texts are more ambiguous and noisy, and are therefore generated in a huge volume, which further increases the problem to handle them. Various methods for short text understanding and interpretation along with the limitations faced and, how the interpretation can be used in an effective way is discussed.