Comparing Different Machine Learning Techniques for Classifying Multi Label Data

In recent years, multi-label classifications have become common. Multi label classification is a classification in which a collection of labels is associated with a single instance, which may be a variation of the classification of a single label. The problem of huge data is the classification in which each instance is of different kind which further can be identified with more than one class. The various machine learning strategies for classifying multi-label data are discussed in this paper. Many researches have been carried out that specify the grouping of multiple labels. Here we will compare various classification machine learning techniques that involve two approaches: the adapted algorithm approach and the method of problem transformation. Here we are using naive multinomial bayes and logistic regression. We use certain evaluation metrics to predict the differences as well. Better classification methods are discussed in this paper.