代做Fall 2024 MGMT 571 Final Project代做Python程序
- 首页 >> Algorithm 算法Fall 2024 MGMT 571 Final Project
Deadline: 11/21/2024 11:59pm EST
This project aims to apply data mining algorithms covered in the course in a real data competition. If you are new to Kaggle, please make sure to read all instructions on this link of data competition: https://www.kaggle.com/docs/competitions. You need to first register a Kaggle account before any submission.
1. We would like to develop a predictive model that combines various econometric mea- sures to predict if a firm will file for bankruptcy. The details about our final project competition, data and evaluation criterion are available in the following website. Do not share this link with others!
https://www.kaggle.com/t/67f4dfaea9fe43f4a15f12538e85e3b2
2. The final project contributes 20 points to your final score: (1) 10 points will be eval- uated based on the rank on the Private leaderboard (Not Public leaderboard!), see below the grading policy; (2) 10 points will be based on presentation.
Top 0% ~ 10% : 10/10;
Top 10% ~ 30% : 9/10;
Top 30% ~ 50% : 8/10;
Top 50% ~ 70% : 7/10;
Top 70% ~ 90% : 6/10;
Top 90% ~ 100% : 5/10.
3. Rules in Kaggle competition: (1) Only use your assigned Team Name to submit solutions; see “MGMT 571 Final Project Teams” in Brightspace for all allowed team names; (2) Maximal daily submission is 10; (3)The number of submissions eligible for the final private leaderboard is 2; (4) Only use the provided training data to train your model; (5) Only use SAS EM in your analysis. For the fairness of the whole class, using anything other than SAS EM for your competition and analysis will cause your team to receive ZERO on the Final Project. If any of these rules is broken, the final score will be discounted.
4. Please choose one team member to submit a brief description of your team’s final chosen two models for Private leaderboard and a zip file of your SAS EM project to Brightspace, before 11/25/2024, 11:59 pm EST, so that the TA can reproduce your result on Private leaderboard. If TA can not reproduce your result, your final score will be discounted.
5. The presentation will be evaluated based on the content and presentation skills. The presentation should be ≤ 5-minutes long. The content should contain the final al- gorithm used in the leaderboard, other tried algorithms, why do you choose the final one, lessons learnt etc.