In this article, we will do a complete machine learning pipeline from getting data through APIs, performing exploratory data analysis and formulating a real-world problem into a machine learning model. The dataset we use for this post is New Zealand Crash Analysis Dataset which is updated on a quarterly basis by the Transport Agency. The dataset was last updated on October 2018 (from January 2000). It contains all traffic crashes as reported to the Transport Agency by the NZ police. However, not all crashes are reported NZ police. A big portion of minor car crashes are settled on site by the parties without reporting to the police. The level of reporting increases with the severity of the crash. Due to the nature of non-fatal crashes it is believed that these are under-reported.

more ...