Selecting Systemic Features for Text Classification
Casey Whitelaw, Jon David Patrick · 2004
Systemic features use linguisticallyderived language models as a basis for text classification. The graph structure of these models allows for feature representations not available with traditional bag-of-words approaches. This paper explores the set of possible representations, and proposes feature selection methods that aim to produce the most compact and effective set of attributes for a given classification problem. We show that small sets of systemic features can outperform larger sets of wordbased features in the task of identifying financial scam documents.