A research paper by Dr. Firuz Kamalov, Associate Professor, Faculty of Engineering, Applied Science Technology at Canadian University Dubai, in collaboration with Dmitry Denisov of Deloitte/Careem, on finding a solution to deal with imbalanced data using the Gamma distribution was published in the leading AI journal, Knowledge-Based Systems.
Dr. Kamalov’s and Denisov’s research addresses the issue of imbalanced class distribution which is a common problem in a number of fields including medical diagnostics, fraud detection, and others. Imbalanced data causes bias in classification algorithms leading to poor performance on the minority class data.
Their paper proposed a novel method for balancing the class distribution in data through intelligent resampling of the minority class instances. The proposed method is based on generating new minority instances in the neighborhood of the existing minority points via a gamma distribution. Their method offers a natural and coherent approach to balancing the data.
Dr. Kamalov and Denisov conducted a comprehensive numerical analysis of the new sampling technique and gained insight into a more effective solution to imbalanced data. The research concludes that the new technique offers a simple yet effective sampling approach to balance data. This research highlights CUD’s commitment to quality research in collaboration with industry.