Development of an Efficient Supervised Machine Learning Model using Concept Drift Analysis: A Review
Abhay Anil Dande, M. A. Pund · 2024
The need of efficiently working supervised machine learning models is of great importance and it is the domain of widespread research in the top Information Technology sector industry. Presently machine learning models are becoming very complex, and it is necessary to maintain the efficiency of the models in the increased complexity for better performance. Efficiency of the complex supervised machine learning models plays a significant role in the improved user experience, cost and resource optimization, real time applications, model interpretability, explainability and its reuse ability. Organizations are able to utilize the full capability of the supervised learning models and can have the expected outcome in the allied sector with efficient supervised machine leaning models. Machine learning is the part of the almost any industry and has its effect on many aspects of human life. To design the expected workable and efficient machine learning model is really challenging. It mainly includes the importance steps such as to select the proper data sets, accurate feature analysis, selecting the proper machine learning model type, architecture, hyper parameters finally designing efficient model. Concept drift is the major challenge in this design process and that needs to be analyzed with systematic approach. This review paper aims to discuss Concept drift analysis with respect to supervised machine learning models.