Final Peer Review - vy45

[vy45]

The team tried to make use of weather data to predict the occurrence of wild fires across the US. The data source has not been mentioned, but going through the previous reports mentions it to be from Kaggle. There were nearly 1.6 million rows of data present.

• The project is well written and presented, the use of hyperlinks for navigation across the report is a very helpful addition. They made sure not to build a weapon of math destruction from the get go.

• Many different techniques were used to predict the wildfires and each technique was described succinctly.

• Good amount of effort has been taken to give a visual representation of the data set.

• A description of what the error of ~0.5 meant qualitatively would help in understanding if the occurrence of fire was predicted well or not, or was it close by hours or days etc.

• Would've been nice to see how basic regression techniques that were taught in the class performed on this data set.

• Since there was a problem of very few positive data points, with the occurrence of wild fire, focusing on a particular city like California might've helped for the over all project in general.