Premium project

Clustering Bustabit Gambling Behavior

Use cluster analysis to glean insights into cryptocurrency gambling behavior.

Start Project
9 Tasks1,500 XP

Loved by learners at thousands of companies


Project Description

Have you ever wondered if you could quantify the behavior of gamblers at the casino? Some seem to win the most, some can be reckless and risky with their bets, and others are casual about the whole experience. While collecting this data from the casino might be a challenge, there is an online platform called [Bustabit](https://www.bustabit.com/play) in which gamblers can bet Bitcoin. We've collected data on thousands of Bustabit gambling sessions and tracked the user, the amount bet, the amount won, and various properties of the particular game itself. Using this data, you will perform a cluster analysis from start to finish in an attempt to group gamblers based on their gambling behavior. To complete this project, students should be comfortable with R programming, the `tidyverse` package in particular, as the data manipulation and summarization routines will use this. The dataset used includes 10,000 games of Bustabit. Each game tracks the particular gambler, the BustedAt value of the game, and the multiplier at which the gambler cashed out.

Project Tasks

  1. 1
    A preliminary look at the Bustabit data
  2. 2
    Deriving relevant features for clustering
  3. 3
    Creating per-player statistics
  4. 4
    Scaling and normalization of the derived features
  5. 5
    Cluster the player data using K means
  6. 6
    Compute averages for each cluster
  7. 7
    Visualize the clusters with a Parallel Coordinate Plot
  8. 8
    Visualize the clusters with Principal Components
  9. 9
    Analyzing the groups of gamblers our solution uncovered
Technologies
R R
Topics
Data ManipulationData VisualizationMachine LearningImporting & Cleaning Data
Eric Hare Headshot

Eric Hare

Chief Data Scientist at Omni Analytics Group
Eric Hare is the Chief Data Scientist at Omni Analytics Group, a boutique statistical consulting firm specializing in data visualization, modeling, and Shiny applications. Eric graduated from Iowa State University with a PhD in Statistics and Computer Science in 2017 under the supervision of Dr. Heike Hofmann.
See More
Lawrence Mosley Headshot

Lawrence Mosley

Founder of Omni Analytics Group
Lawrence Mosley is the Founder of Omni Analytics Group, a statistical consulting company specializing in machine learning, data strategy, Shiny development, and analytics training. He earned his PhD in Industrial Engineering at Iowa State University. Check him out on Twitter at @OmniAnalytics.
See More

What do other learners have to say?

I've used other sites—Coursera, Udacity, things like that—but DataCamp's been the one that I've stuck with.

Devon Edwards Joseph
Lloyds Banking Group

DataCamp is the top resource I recommend for learning data science.

Louis Maiden
Harvard Business School

DataCamp is by far my favorite website to learn from.

Ronald Bowers
Decision Science Analytics, USAA