Today I compared xgboost and randomforest in creating a model that predicts if a kid will be diagnosed late or early with adhd. The threashold is the top 25% of National Survey of Childrens Health kids with ADHD so “late” is considered after 9 years old. As someone who has been diagnosed with adhd “late” (at 14) this project means a lot to me. I currently have 17 factors and my accuracy is bad lol. However this is expected because I am in the beginning stages of my project. I hope to make this information available to adults and those willing to learn more through a website.
