Multi-Label Classification with PSO based Synthetic Minority Over-Sampling Technique (Psosmote) for Imbalanced Samples

Abstract

Recently, the learning from unbalanced data has emerged to be a pre-dominant problem in several applications and in that multi label classification is an evolving data mining task, learning from unbalanced multilabel data is being examined. However, the available algorithms-based SMOTE makes use of the same sampling rate for every instance of the minority class. This leads to sub-optimal performance. To deal with this problem, a new Particle Swarm Optimization based SMOTE (PSOSMOTE) algorithm is proposed. The PSOSMOTE algorithm employs diverse sampling rates for multiple minority class instances and gets the fusion of optimal sampling rates and to deal with classification of unbalanced datasets. Then, Bayesian technique is combined with Random forest for multilabel classification (BARF-MLC) is to address the inherent label dependencies among samples such as ML-FOREST classifier, Predictive Clustering Trees (PCT), Hierarchy of Multi Label Classifier (HOMER) by taking the different metrics including precision, recall, F-measure, Accuracy and Error Rate.

FG-SMOTE: Fuzzy-based Gaussian synthetic minority oversampling with deep belief networks classifier for skewed class distribution

“…Random sampling enables the classifiers with learning ability from the available data without any bias. Besides, the traditional random sampling techniques detect the data by employing a random sample by replacing the conventional data set which has less significance in enhancing the classifier performance (Priyadharshini and Pavithira, 2019). Hence, more advanced, and sophisticated techniques were developed to handle the imbalanced data and among them, SMOTE technique became a pioneer in effectively resolving the unstructured data by providing solutions to overcome the issue of imbalance.…”