Classification of Iris Flowers

Projects Goals:

The project , classification of iris flowers is based on the data set named as the iris dataset . In this project, my primary goal is to learn the basic steps of data science and machine learning processes. Using the python environment data summarisation, data visualization, evaluating the best algorithm for the data set and making predictions, the main steps of the project, are performed. In this project, the best classification model is selected by comparing with other prediction models and the perfect model for this data set is used for better prediction.

Findings:

The iris data set is a small data set with 4 attributes and 150 rows and attributes are numeric. All of the numeric attributes are in the same units and the same scale, not requiring any special scaling or transforms to get started.

In this project, 6 prediction models are used and they are:

• Logistic Regression (LR)
• Linear Discriminant Analysis (LDA)
• K-Nearest Neighbors (KNN)
• Classification and Regression Trees (CART)
• Gaussian Naive Bayes (NB)
• Support Vector Machines (SVM)

From the efficiency comparison graph, it is apparent that KNN has the largest estimated accuracy score. Therefore, the KNN model is used for prediction.And the KNN model is used directly on the validation set and summarize the results as a final accuracy score, a confusion matrix, and a classification report.