Data Mining: Classification and Prediction

Abstract: Data mining refers to a collection of computational techniques and methodologies designed to extract insights from extensive datasets. This involves the utilization of advanced data analysis tools to reveal the structural information underlying large datasets. Among these tools, machine learning methods play a crucial role, enabling not only the management of data but also the execution of analytical and predictive operations. Within machine learning, supervised learning stands out as a methodology that employs input data to generate two types of outputs: qualitative, which describes data classes, and quantitative, which predicts data trends. In the context of supervised learning, classification tasks yield qualitative responses, while prediction or regression tasks produce quantitative outputs. Keywords: Classification; Decision trees; Ensemble learning; Imbalance learning; Multilayer feed-forward neural network; Prediction; Support vector machines