Multi-label classification

So far, we have seen problems in which we need to classify text into one of the classes or labels. There are text classification problems in which we might need to attach multiple categories to the same document. For example, in the Yelp reviews dataset, the reviewer might be talking about different aspects of a restaurant such as the food, ambiance, quality of service, and so on. In this case, we need to identify what categories the review belongs to so that we can understand the overall review. We will now look at some of the existing approaches to this problem.

Binary relevance

Multi-label classification for identifying L labels of a document can be transformed to an L binary classification problem. In this approach, we pass a document into the L binary classifiers, where each are trained for identifying one of the L classes. The output of the L classifiers is merged to produce a vector of class labels to which the document belongs. Even for simple models such...