Repo for various data science projects
LoanApplicationPrediction - small loan-application dataset to determine if customer should be given loan or not. Data contains mixed types (categorical, ordinal, numerical), many missing values and data entry issues.
AnswerClassifier - Quora's challenge to classify answers on their site as good or bad. Data is completely anonymized and taken from real production datasets.
AdtechClickPrediction - Predict the probability of a click given ad impression data.
BizDevLeadScoring - Score potential business development opportunities/customer leads by value.