Use various machine learning models to predict the btc price of dark market cocaine. Our Project_code.ipynb contains the complete workflow of our process: Data Cleaning, Exploratory Data Analysis, Feature Engineering, Modeling, and Evaluation.
Project contributors include: Shirley Li, Jingxian Li, Michael Schulze, & Mundy Reimer
Our presentation slides can be found here.
Public Data Set found here: https://www.kaggle.com/everling/cocaine-listings
Description:
The dataset is approximately 1,400 cleaned and standardized product listings from Dream Market's "Cocaine" category. It was collected with web-scraping and text extraction techniques in July 2017.
Extracted features for each listing include: product_title ships_from_to quantity in grams quality btc_price vendor details shipping dummy variables (true/false columns)
For further details on the creation of this dataset and what it contains, see the blog post here: https://medium.com/thought-skipper/dark-market-regression-calculating-the-price-distribution-of-cocaine-from-market-listings-10aeff1e89e0
Files:
Final_code: Contains all our code from modeling to visualization.
dream_market_cocaine_listings: Original CSV file we downloaded from Kaggle.
fixed_cocaine_listings: Dataset we used after cleaning the original Kaggle CSV.