Skip to content
This repository has been archived by the owner on Jun 12, 2019. It is now read-only.
Aidan Sawyer edited this page Nov 28, 2016 · 7 revisions

Description

This is the Data Mining project I created in fulfillment of my 'CSCI.420 - Principles of Data Mining' course in my Fall 2016 semester at RIT.

Purpose

Titled "Tapping into the Data: A Pint-by-Pint Analysis of Craft Beer Styles by Characteristics and Region", it utilizes common clustering algorithms to analyze craftcans data in an attempt to look at the correlation of a beer's International Bitterness Unit (IBU) and Alcohol by Volume (ABV) measurements to beer style, in conjunction with its state of origin.

Introduction

More information about the different steps of my process and about the data can be found in the other pages of the wiki. More in depth and formalized writings can be found in the .pdf and .Rmd files in the writeups/ subfolder of the 'code' section.

Precursors

Methods

Results

Discussion

Data Richness

I'd very much like to redo or update the analysis with richer data that includes more information that may be more conducive to mapping. Measurements like the Standard Reference Model (SRM) (a measurement of the beer's color) would be very helpful in differentiating between a traditional IPA vs a Black IPA (for example), and the Initial and Specific Gravity (a measurement of the amount of residual sugar in the beer) would be useful to bring out the barleywine, scotch ale, and Wee Heavy, and in the creation of commonly used features such as the BU:GU more useful for inter-style/family clustering.

Data Mining Process

...

Clone this wiki locally