PROBLEM DESCRIPTION

The overall goal of this project is to do something that nobody else has ever done before. We want to take an existing machine learning model and improve it so it can better solve a classification problem. Our problem is the classification of Iris flower dataset. The dataset has three different type of Iris flowers: Iris Setosa, Iris Virginica and Iris Versicolor. The dataset is a balanced and unbiased. It consists of 50 different samples of each Iris flowers. Each sample has four distinct features: length of sepals, width of sepals, length of petals and width of petals.

The dataset was originally introduced and used by British statistician and biologist Ronald Fisher, was used for linear discriminant analysis. After which is became a typical test case for many statistical classification techniques in machine learning. Ideally, linear discriminant model can only be obtained when the object species are known – class labels, this make this dataset usable for classification purposes only.