My notes:
- I like the abstract and how it tells exactly the steps the paper will take and the topic(s)
- Is there a reason you choose random forest over the others?(Ok, you answered this question later on)
- Do you have a hypothesis of how it compares to others in your implementation and comparisons based on research?
- make image bigger, I cannot read the words and numbers even after zooming in some
- What are the percents in the image representing?
- Maybe define what a decision tree is when you first mention them
- are you having a validation set and a test set?
- what will you use to analyze how well your forest did?
- Is there a reason that you are using sk-learn's package for decision tress over a different package or your own implementation?
- What are you going to do if random forest results don't look positive?
- What exactly are your research questions?
- current prediction rates of the data set or of the algorithm or both?
- how are you going to compare algorithms trained and tested on different data sets?
Overall, I like the set up and order of the paper. I also thought you described what a random forest is well.