Text Classification

Text classification involves assigning labels to text that denote a particular category, for example whether the sentiment expressed in this text is positive or negative. Some biomedical applications of this technique include automated indexing of biomedical research articles, assignment of billing codes to clinical notes, and classification of patient-generated text as to whether or not it indicates the presence of a condition such as dementia. This chapter provides an overview of text categorization, with explanations of some common techniques and biomedical tasks. The chapter draws connections between some challenging aspects of biomedical text processing and representational choices with the capacity to address these challenges. In addition it provides some worked examples of applications of text categorization techniques to clinical problems.