My collection of Kaggle projects. (Profile link: https://www.kaggle.com/changhilee)
All file formats are in the .ipynb(Jupyter Notebook) format. You can run these files with Jupyter Notebook, PyCharm IDE, Google CoLab, etc. Most datasets are too large to host on Github, so API calls will be provided in their respective READMEs.
-
Suspicious Transaction Detection(https://www.kaggle.com/changhilee/suspicious-transaction-detection) A fraud transaction detection machine learning model using gradient boosting classification with K-best feature selection and PCA for dimensionality reduction.
-
Wildfire Ignition Point Prediction(https://www.kaggle.com/changhilee/wildfire-ignition-point-prediction) A wildfire prediction machine learning model using GradientBoostingRegressor and one-hot encoding.
Note: Performed as part of school work. Course number and instructor information have been omitted to prevent plagiarism.