Machine learning has been extensively used in the field of automation systems, and in machine learning, imbalanced data is a prevalent word in fact it is a challenging element to deal with. How to deal with this imbalanced data is a major focus for the majority of studies. In terms of balancing the data, at the data level point under-sampling, over-sampling, and their variants are widely used. Since over-sampling creates precise replicas of examples from the minority class, it may increase the risk of over-fitting. Under-sampling wipes out a significant quantity of data, making it more difficult to determine where the decision boundary between minority and majority classes lies. In this work, a novel method has been proposed that combines both under-sampling and over-sampling strategies based on the Heuristic Range-Basd Association rule and a modified Tabular Generative Adversarial Network (TGAN) known as the Ranged-based Association rule and GAN-based Hybrid (RAMGANH) to avoid those kinds of problem scores and produce a well-balanced data set. The proposed approach has been tested by using existing standard classifiers with a few standard data sets, and the results demonstrate appreciable improvements in the classifier performance than the other state of-art-method.