چکیده مقاله
In machine learning, dealing with binary imbalanced data classification ischallenging due to unequal class sizes, leading to model bias We propose a unique methodthat uses filtering, ADASYN oversampling, and ENN cleaning to balance data, improveminority class accuracy, and boost overall model performance, showing significantimprovements in AUC, F1, and G mean metrics
کلیدواژهها
نویسندگان
Zahra Arefzadeh
Faculty of Data Science and Intelligent Systems, Persian Gulf University, Bushehr, Iran
Erfan Dehghani
Faculty of Data Science and Intelligent Systems, Persian Gulf University, Bushehr, Iran
Mohammad Bozorgmehr
Faculty of Data Science and Intelligent Systems, Persian Gulf University, Bushehr, Iran
شیوه ارجاع
Arefzadeh, Zahra and Dehghani, Erfan and Bozorgmehr, Mohammad,1403,Advancing Binary Imbalanced Classification: A Novel Hybrid SamplingApproach for Noise Reduction and Data Integrity,1st International Biennial Conference of Artificial Intelligence and Data Science 2024,Bushehr
ارائهشده در
مجموعه مقالات اولین کنفرانس بین المللی دوسالانه هوش مصنوعی و علوم داده3 اردیبهشت 1403 · بوشهر