ML-classification-task

Classical Iris flowers classification task solving by ML algorhitms.

The task was completed using the tutorial from Jason Brownlee. It helps to implement the knowledge and skills of working with Machine Learning algorithms on a concrete complete example.

Practical skills:

• loading dataset
• primary data analysis
• data visualisation
• suitable ML alghoritms selection
• model cross-validation and choosing the best model for the current task
• make predictions
• evaluate predictions

Theoretical knowledge:

• statistical parameters: mean, standard deviation, min, max, percentiles of the features;
• data visualisation methods: box and whisker plots, skatter matrix plots;
• K-fold cross-validation method;
• predictions evaluating parameters: accuracy, precision, recall, F1 score, support.

Python libraries used:

• Pandas
• Matplotlib
• Scikit-learn

Some ML algorithms appropriate for the classification task:

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