Classification Models of Text: A Comparative Study

摘要

Classification, as a problem, has far-reaching applications in data mining. Most notably, text classification has skyrocketed in usage and importance, having implications in education, politics, and even finance. As a process, classification generally occurs as follows: preprocessing, feature engineering, dimension decomposition, model selection, and model evaluation. This paper will provide a brief overview of each stage, as well as survey and compare popular classification algorithms.