There are four different case studies. Each case study represents different concepts of Machine Learning. First the problem will be defined. Next, is the data exploration. After that, visual inspection. Model building and testing. Case studies are embeded as PDF files. For better experience try to read them on a larger screen such as desktop or larger tablet. On your phone you might not be able to grasp everything. Adjust text size using PDF + and - icons. Move mouse coursor over PDF text to scroll down. The ink to complete PDF file is under Case Title.
This case study will focus on finding a polynomial model for a data set that is described in “An Introduction to Statistical Learning with Applications in Python.” A must read book for anybody who wants to learn about Data Analysis. A free pdf download of the book can be found here.
The real procedure to determine if a person has a heart disease is an invasive process. It is costly and painfully for individuals. The objective in this study is to find a model that can help predict existence of heart problems using clinical available test results. More info about this data set and study can be also found in the following link.
This case study will focus on Classification Logical Regression (CLR). CLR is a type of analysis where the model is trying to classify the output into more than two possible outcomes. The data set is about three type of penguins.
Meaning of the confusion matrix and accuracy metrics such as precision, recall and f1-score. More can be found in this link.
A tunnel building company is using X-rays to find out rock density. They are using an automated boring machine to drill (you can watch an example video of this technique in this link). The machine boring heads on the equipment are switched based on rock density.
Information about the theory and history of the receiver operating characteristics curve can be found in this Wikipedia article.
More info about Anscombe's quartet can be found in this Wikipedia article.