We will definitely need Irish Whiskey - but which should we stock first?
In this capstone we'll explore the Iowa Liquor Sales dataset to gain insights into Irish wiskey sales in Iowa. We'll explore dimensionality reduction techniques as well as clustering algorithms to come up with an unsupervised learning model that will predict which Irish whiskies we should have in stock when we open our pub.
Project artifacts in this repo:
- IowaLiquorSales_IrishWhiskies.csv - Preprocessed dataset that is used for this analysis.
- Notebook-UMLDataPreProcessing - Jupyter Notebook detailing the steps used to extract the subset of data for this analysis from the original Kaggle dataset.
- Notebook-UnsupervisedLearningCapstone - Jupyter Notebook detailing the analysis conducted on the preprocessed data.
- Presentation-UnsupervisedLearningCapstone - Presentation slides outlining conclusions as presented to the DSI09 cohort for comittee review and grading.