Application of Apriori Algorithm in Multi Label Classification
Feng Qin, Xian-Juan Tang, Zekai Cheng · 2013
Multi_label learning and application is a new hot issue in machine learning and data mining recently. In multi_label learning, a training set is composed of instances, each is associated with a set of labels, and the task is to predict all the appropriate labels of unseen instances. In this paper, the authors research on proposing Apriori algorithm to search the relationship between all labels. In the iteration process of generating frequent itemsets, compound labels with strong association are replaced by existing single labels. And then it uses ML_KNN algorithm to classify multi_label data. Finally, at the stage of predicting labels, compound labels are filled based on the relationship between labels. Experiments on emotions data set show that this method is effective.