I found some solutions that have the same evaluation metrics than this competition. Overall, their approach was simple solutions, even some of them such as March Machine Learning Mania 2022 - Men’s use a solution from a previous competition. Some other use Auto machine learning, and one of them won by using linear regression, the oldest solution use neural network, which makes me think that just the classical models can work perfectly with our dataset.
\nNeural Network
\n\nSimple Linear Regression
\n\nXGBoost:
\n\nAutoLGBM
\n", "messages": [ { "id": 1997457, "postDate": "2022-10-20T22:47:03.623Z", "content": "I found some solutions that have the same evaluation metrics than this competition. Overall, their approach was simple solutions, even some of them such as March Machine Learning Mania 2022 - Men’s use a solution from a previous competition. Some other use Auto machine learning, and one of them won by using linear regression, the oldest solution use neural network, which makes me think that just the classical models can work perfectly with our dataset.
\nNeural Network
\n\nSimple Linear Regression
\n\nXGBoost:
\n\nAutoLGBM
\n", "rawMarkdown": "I found some solutions that have the same evaluation metrics than this competition. Overall, their approach was simple solutions, even some of them such as **March Machine Learning Mania 2022 - Men’s** use a solution from a previous competition. Some other use Auto machine learning, and one of them won by using linear regression, the oldest solution use neural network, which makes me think that just the classical models can work perfectly with our dataset. \n\nNeural Network\n- [Quora Question Pairs 2017](https://www.kaggle.com/competitions/quora-question-pairs/discussion/34355)\n\nSimple Linear Regression\n- [Google Cloud & NCAA® ML Competition 2019-Women's](https://www.kaggle.com/code/jleecook/1st-place-regression-to-projected-mov/script)\n\nXGBoost:\n- [March Machine Learning Mania 2022 - Men’s](https://github.com/fakyras/ncaa_women_2018/blob/master/win_ncaa.R)\n\nAutoLGBM\n- [March Machine Learning Mania 2022 - Women's 2022](https://www.kaggle.com/code/rishirajacharya/mmlm22-women-tuned) ", "votes": 7 } ], "comments": [], "raw_markdown_by_id": { "1997457": "I found some solutions that have the same evaluation metrics than this competition. Overall, their approach was simple solutions, even some of them such as **March Machine Learning Mania 2022 - Men’s** use a solution from a previous competition. Some other use Auto machine learning, and one of them won by using linear regression, the oldest solution use neural network, which makes me think that just the classical models can work perfectly with our dataset. \n\nNeural Network\n- [Quora Question Pairs 2017](https://www.kaggle.com/competitions/quora-question-pairs/discussion/34355)\n\nSimple Linear Regression\n- [Google Cloud & NCAA® ML Competition 2019-Women's](https://www.kaggle.com/code/jleecook/1st-place-regression-to-projected-mov/script)\n\nXGBoost:\n- [March Machine Learning Mania 2022 - Men’s](https://github.com/fakyras/ncaa_women_2018/blob/master/win_ncaa.R)\n\nAutoLGBM\n- [March Machine Learning Mania 2022 - Women's 2022](https://www.kaggle.com/code/rishirajacharya/mmlm22-women-tuned) " } }