1. Project Description
Every year, professional football clubs spend billions of dollars for the transfer of players.In recent years the transfer fees paid by football clubs for their players have sky-rocketed. For instance, in 2017, Paris St-Germain signed Barcelona forward Neymar for a world record US$233 million.
Accurately assessing and predicting football player's market value can foster various applications for the football industry, such as transfer negotiations, club management, new football tactics, etc.
In this project, you will attempt to answer one of most intriguing questions in the football world: How the market value of a player is determined?
The datasets you need is attached below:
The train set should be used to build your machine learning models. In the train set, you are provided with the market value (i.e., the “ground truth”) for each player.The “label” is the market value of the players.Your model will be based on“features”like heights, total goals, nationalities, and so on.You can also create new features using your domain knowledge or data skills.
The test set should be used to see how well your model performs on unseen data. In the test set, there is no information on the players' market value.It is your job to predict the market value.For each player in the test set, use the model you trained to predict the player's market value (in USD).
No limitations or rules on model types.You can use any methods/algorithms to train your model (e.g. linear, non-linear, logistic, tree-based models, etc.).
==========================================================================================================
2. Submission Files Format
You should upload two files use the upload file link on canvasbyDec 16 (Friday) 11:59 pm. No extension will be granted.
The two files are:
Your python script
The prediction outcome
For the python script, you should submit a python file namedexactlyas"final.py"
For the prediction outcome:
- You should submit a csv file namedexactlyas"
prediction.csv
"
- The submitted csv file should haveexactly 4134 entries plus a header row.
- The file should haveexactly 2 columns:
player_id(sorted in any order)
market_value_usd(the predicted price)
I posted an
example_sumbission.csv
Download example_sumbission.csv
here as an example of what a submission file should look like.This example submission predicts that each player's market value is always total_apperance*1000(By no means this is a good prediction, I believe you can absolutely do better than this example).
==========================================================================================================
3. Evaluation
Your goal
is to predict the market price for each diamond product in the test set. Try your best to maximize thecoefficient of determination ( i.e., R-squared)of your model.The
R2" role="presentation" style="display: inline-block; line-height: normal; text-align: left; word-spacing: normal; overflow-wrap: normal; white-space: nowrap; float: none; direction: ltr; max-width: none; max-height: none; min-width: 0px; min-height: 0px; border: 0px; padding: 0px; margin: 0px; position: relative;">
provides a measure of how well observed outcomes are replicated by the model, based on the proportion of total variation of outcomes explained by the model. Specifically:
R2=SSRSST=∑(yi^−y¯)2∑(yi−y¯)2" role="presentation" style="display: inline-block; line-height: normal; text-align: left; word-spacing: normal; overflow-wrap: normal; white-space: nowrap; float: none; direction: ltr; max-width: none; max-height: none; min-width: 0px; min-height: 0px; border: 0px; padding: 0px; margin: 0px; position: relative;">
Your grade
depends on the following criteria:
