I've conducted an analysis applying supervised and unsupervised learning techniques. These techniques are used practically to analyze the market and its behavior. For example, supervised learning algorithm could be used to predict the demand for a product based on historical sales data and other factors that might influence its demand. At the same time, an unsupervised learning algorithm could be used to identify clusters of consumers with similar purchasing habits based on their past purchase data. This could be useful for targeted marketing efforts or for identifying potential new product opportunities.
The supervised analysis was conducted on the 1994 Census bureau dataset and unsupervised learning techniques on a mall’s customers dataset.
The supervised learning algorithms were K nearest neighbor, logistic regression, tree prediction, tree predictor with Gini index, Naïve Bayes, and random forest. All the algorithms were compared according to their accuracies.
The unsupervised learning algorithm used was k-means. In order to choose the number of k, the elbow method, the average silhouette method, and the gap statistic were used.