Examines class imbalance in the diabetes dataset. Applies over/under-sampling strategies (e.g., SMOTE or random resampling) and compares to baseline. Trains models and reports metrics/ROC across ...

This project implements a fraud detection system using Support Vector Machines (SVM) on an extremely imbalanced credit card dataset. The focus is on preprocessing, SMOTE oversampling, SVM kernel ...

Abstract: With the increasing number of cybersecurity threats, the detection and prevention of malicious samples have become an important research direction in the field of information security.

Abstract: To address the risk of soft tissue injury caused by saw blade penetration during freehand orthopedic osteotomies, this paper designs and implements a vibration-based penetration detection ...

宫颈癌早期筛查的机器学习模型研究。采用随机森林、XGBoost等算法，基于UCI数据集（858例，含14个特征），通过SMOTE oversampling平衡类别分布，结合Boruta特征选择筛选关键变量（如年龄、性伴侣数量、HPV感染等）。结果显示随机森林模型表现最优，AUC 0.997，灵敏度 ...

SMOTE, once a solo endeavor by Daniel Foggin, now expanded to a fuller ensemble blends traditional instrumentation — creaking strings, funeral drums, somber winds — with thunderous synth work and ...

Williams, A. and Louis, L. (2026) Cumulative Link Modeling of Ordinal Outcomes in the National Health Interview Survey Data: Application to Depressive Symptom Severity. Journal of Data Analysis and ...