An Exploration on Text Classification with Classical Machine Learning Algorithm

Our project “An Exploration on Text Classification with Classical Machine Learning Algorithms” aims to investigate and enhance the understanding of the application of traditional machine learning techniques in the domain of text classification. Leveraging foundational studies, such as Joachims’ work on Support Vector Machines (SVMs) for text classification, the project will build upon established methodologies and address challenges associated with high-dimensional feature spaces. Drawing inspiration from the comparative analysis conducted by McCallum and Nigam on Naive Bayes and decision trees, the research will explore the strengths and limitations of classical algorithms, emphasizing simplicity, interpretability, and efficiency. Techniques for feature selection, as proposed by Joachims in 1999, will be further examined to improve the handling of irrelevant features in text data. Additionally, the project will consider advancements in hierarchical text classification, as introduced by Forman, to enhance accuracy through the incorporation of class hierarchy.