Mini Project - Modeling

• Course: YearDraem School 2nd
• Project period: 7 Jun. ~21 Jun., 2022
• Team members
  • 권오균(O. Kwon), 김태영(T. Kim), 양아연(A. Ryang), 이주안(J. Lee), 홍승현(S. Hong)

Project topic

IEEE-CIS Fraud Detection

https://www.kaggle.com/c/ieee-fraud-detection

Description

Fraud prevention system is actually saving consumers millions of dollars per year.
The data comes from Vesta's real-world e-commerce transactions and contains a wide range of features from device type to product features.
If successful, you'll improve the efficacy of fraudulent transaction alerts for millions of people around the world, helping hundreds of thousands of businesses reduce their fraud loss and increase their revenue.

• Data source
  • Vesta Corporation from Kaggle

The reason we chose this topic

It could be have an opportunity to practice on machine learning classification models as well we thought there are many things that can be referenced or learned because of the topic in Kaggle out of the given 5 topics.

What we want to achieve with this project

• Try to use new IDE (Visual Studio Code or PyCharm)
• Try to make and use the structural layout
• Select important features through feature engineering.
• Try to select a model based on a logical basis.
• Try to do modeling itself.

Models selected for Classification

• XGBM

Conclusion and Discussion

• An experience to realize the importance of systematic planning, execution, and time management
• Learned that much more active communication such as sharing progress is required
• Got used to it after practicing Git fork and VS Code
• It was a time to think about how to approach masked data
• I was able to practice feature engineering.
• Re-feeled the importance of EDA once again