A Comprehensive Survey of Text Classification Techniques and Their Research Applications: Observational and Experimental Insights

TL;DR

This survey provides a detailed taxonomy and evaluation of text classification techniques, highlighting their research applications and offering insights into their effectiveness through empirical and experimental analysis.

Contribution

It introduces a hierarchical taxonomy for text classification based on research fields and evaluates techniques using dual empirical and experimental approaches.

Findings

Structured taxonomy enhances understanding of classification algorithms.

Empirical assessment across four criteria offers comprehensive insights.

Experimental comparison ranks techniques within specific research sub-categories.

Abstract

The exponential growth of textual data presents substantial challenges in management and analysis, notably due to high storage and processing costs. Text classification, a vital aspect of text mining, provides robust solutions by enabling efficient categorization and organization of text data. These techniques allow individuals, researchers, and businesses to derive meaningful patterns and insights from large volumes of text. This survey paper introduces a comprehensive taxonomy specifically designed for text classification based on research fields. The taxonomy is structured into hierarchical levels: research field-based category, research field-based sub-category, methodology-based technique, methodology sub-technique, and research field applications. We employ a dual evaluation approach: empirical and experimental. Empirically, we assess text classification techniques across four critical criteria. Experimentally, we compare and rank the methodology sub-techniques within the same methodology technique and within the same overall research field sub-category. This structured taxonomy, coupled with thorough evaluations, provides a detailed and nuanced understanding of text classification algorithms and their applications, empowering researchers to make informed decisions based on precise, field-specific insights.