- Title: AI competition to predict user inconvenience due to system quality changes
- Goal: Analyze unidentified system records (log and numerical data) to diagnose user discomfort with changes in system quality
- Category: Classification, Log Data
- Tool Used: python, pandas, sklearn, catboost, lightgbm, ngboost
- Period: Jan, 2021 ~ Feb, 2021
- Data: System Log Data
- Shape: (828,624 * 16), (16,554,663 * 6)
- From: Dacon(LG AI Research)
- Metric: AUC
- Public Score : 53th(12.6%) / 0.83775
- Private Score: 42th(10.0%) / 0.83557