Diagnosing breast cancer based on collected breast tissues. In this section, we will see how a simple k Nearest Neighbor algorithm can be very powerful in identifying malignant cancer cells.

Spam filtering algorithm using Bayesian probability. In this section, we will build and train a spam identifier using a dataset available to public at [spam dataset]. This dataset contains 5,574 real English messages, 747 of which are spam. The goal is to build an accurate spam identifier with at least 90% accuracy. Let's begin!

Applying rule based learning technique in identifying poisonous mushrooms. In this section, we will implement a poisonous mushrooms classifier using rule classification.

With decision trees technique, we will implement a binary predictor which can predict whether a bank loan will go into default based on collected information of the loan applicants.

Using neural networks, we will implement a concrete estimator to predict the strength of concrete with different mixtures. Concrete with different ingredients and ratios will result in have different compression strength.

Support Vector Machine (SVM) is one of the most used machine learning techniques when it comes to dealing with classification tasks that involve numerical data. In this example, we will observe how SVM can be a powerful solution to identifying alphabet letters!

In this section, we will revisit the Breast Cancer data set. Particularly, we will use different approaches (such as Decision Tree, Random Forest, and Support Vector Machine) other than k-NNs to predict brest cancer diagnosis.

Association Rules learning is the foundation behind recommendation engines. We will use an unsupervised machine learning technique called Market Basket Analysis by mining Association Rules betwwen items in a 30 days worth of transactions data from a groceries store.