the project is about how to predict outcome of soccer matches. The group uses three datasets, including data from 25,000 European soccer matches, betting odds for each match, and EA sports FIFA games.
Things that could be improved:
I like the fact that the group gives links of where they find the data. However, if to read this report on a paper, or a more professional setting, I would prefer to put the links of data source into appendix or into reference page. This is because, if people read on paper, they will not be able to click the link. And the link takes quite a lot of space in the paper.
For the data cleaning, I would prefer if the group talk more about why they choose to delete the rows instead of trying other approaches, such as using statistics of other columns, or prediction.
I would suggest the group to add a section in the end to talk more about their perspective about the future of this project. And of course, add a reference section about where they find all the resources they use for the project.
Things I like about the project:
I like when the group use correlation matrix to represent the correlation between different features from the data. And then from there, they combine some information together and calculate overall score.
When the group chooses model, they start from linear regression, multiclass classification, then add in regularizers. I like the process they use. And I like the fact that they try different models in order to find the best one.
I like they include random forest as part of their analysis, although it does not work well as a model. They then try to use random forest without encoding.
Overall, I think this project is great. This is a very new topic. The group does a great job in explaining the topic, so people who do not know much about soccer can understand the topic too.
the project is about how to predict outcome of soccer matches. The group uses three datasets, including data from 25,000 European soccer matches, betting odds for each match, and EA sports FIFA games.
Things that could be improved:
Things I like about the project:
Overall, I think this project is great. This is a very new topic. The group does a great job in explaining the topic, so people who do not know much about soccer can understand the topic too.