marklekina / data_challenge_22w

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

DALI Data Challenge

Part 1

Describe the WIID dataset given with four visualizations and four short paragraphs. Interpret this as you will, but you should explain the choices of graph you made. Think of it as having to explain the dataset to someone who can't see the Excel sheet. How would you go about it, what insights would you mention, what do you leave out. Use whatever tools you're comfortable with

Part 2

Part 2 is more free-form, and allows you to showcase YOUR specific skillset.

Drawing from a dataset YOU have chosen (preferred but not required) or the WIID dataset, describe it quantitatively and train a model on it, reporting its predictive accuracy.

For example, you could, given a country's population, GDP, and previous scores, predict its GINI index. IMPT: Explain outliers to your model, explain its assumptions, explain the metrics used. We want to know hwo you think about data modelling and what goes into your decision-making process. You can use ML, deep learning, statistical analysis, t-tests, whatever you want! Just explain whatever you do!

Have fun!!!

Email osman.khan@dali.dartmouth.edu with questions!

About


Languages

Language:Jupyter Notebook 100.0%