Also known as: Co training
Co-training is a semi-supervised learning method that trains models using complementary and conditionally independent views of the data. It is useful when labeled examples are limited but unlabeled examples are abundant, allowing models to reinforce one another's predictions.
A semi-supervised learning approach particularly useful when all of the following conditions are true: - The ratio of unlabeled examples to labeled examples in the dataset is high. - This is a classification problem (binary or multi-class). - The dataset contains two different sets of predictive features that are independent of each other and complementary. Co-training essentially amplifies independent signals into a stronger signal. For example, consider a classification model that categorizes individual used cars as either Good or Bad. One set of predictive features might focus on aggregate characteristics such as the year, make, and model of the car; another set of predictive features might focus on the previous owner's driving record and the car's maintenance history. The seminal paper on co-training is Combining Labeled and Unlabeled Data with Co-Training by Blum and Mitchell.