chapter  11
Text Classification
ByCharu C. Aggarwal and ChengXiang Zhai
Pages 50

There are some variations of the basic problem formulation given above for text classifcation. In the hard version of the classification problem, a particular label is explicitly assigned to the instance, whereas in the soft version of the classification problem, a probability value is assigned to the test instance. Other variations of the classification problem allow ranking of different class choices for a test instance, or allow the assignment of multiple labels [60] to a test instance. The classification problem assumes categorical values for the labels, though it is also possible to use continuous values as labels. The latter is referred to as the regression modeling problem. The problem of text classification is closely related to that of classification of records with set-valued features [34]; however, this model assumes that only information about the presence or absence of words is used in a document. In reality, the frequency of words also plays a helpful role in the classification process, and the typical domain-size of text data (the entire lexicon size) is much greater than a typical set-valued classification problem.