What I like about this project
Interesting, I myself watch soccer. I like the way the develop their model. They use a linear regression to provide some intuition first before running into random forest. They also illustrate their data pretty well.
What I am concerned
I am not so familiar with random forest. Maybe you can include more details about it. But for the tree model, I remembered there are some parameters needs to be selected by cross validations. How did you set up your cost function? How did you run cross validation? What is the optimal parameters?Why do you have such low accuracy? Does some parameters plays an important role in this? You did not explain it in the report and you did not upload your code. So I don't know how do you achieve this result and why does the result look like this.
Except for the above concerns, interesting project and well done. And I look forward to see more details about it.
What I like about this project Interesting, I myself watch soccer. I like the way the develop their model. They use a linear regression to provide some intuition first before running into random forest. They also illustrate their data pretty well.
What I am concerned I am not so familiar with random forest. Maybe you can include more details about it. But for the tree model, I remembered there are some parameters needs to be selected by cross validations. How did you set up your cost function? How did you run cross validation? What is the optimal parameters?Why do you have such low accuracy? Does some parameters plays an important role in this? You did not explain it in the report and you did not upload your code. So I don't know how do you achieve this result and why does the result look like this.
Except for the above concerns, interesting project and well done. And I look forward to see more details about it.