Qualitative Study in Natural Language Processing: Text Classification

Abstract

Recently Natural Language Processing (NLP) has excessive attention due to increased data available online. Although there is huge development in Arabic NLP, but still it is behind English NLP. The aim of this study is to examine the most useful qualitative research design that can be used in the field of Arabic Natural Language Processing in general and in-text classification in specific. Two research designs have been examined in detail; survey and experimental research. Both benefits and drawbacks illustrated for each research design with examples for each method. Thus to have better understanding for some of the research methods. Then deciding which one is best fit to design a qualitative study in this field. Furthermore, this paper suggested framework can be used while researching in the field of NLP. The framework suggested to start with survey or Literature review then follow it with experiment.