Predictive analysis on dataset provided by Indiana University Bus Transit to create and evaluate different models and suggest possible modifications to improve the overall efficiency of the transit system.
Proprietary data from transit system is cleaned and appended with other datasets including - Weather dataset - This dataset allows to analyze the effect of weather on the number of people who travel by the buses. Class and exams schedule - It allows to see how exam schedule and class timings affect the number of people travelling by bus.
Models were tested and created using several techniques including- Linear Regression, K-means, K-Nearest Neighbour, association rule mining and Naive Bayes.