Business-Analytics

Course homepage for "Business Analytics" @Korea University

Notice

• 비대면 기말고사 실시

  • 일시: 2020년 12월 22일(화) 오후 3시 30분~5시 30분 (120분)
  • 형식: Google Meet 회의 링크에 접속하여 시험시간 동안 비디오를 켠 상태에서 시험 실시
  • 시험 방식: Open Slide + Cheating Sheet (A4용지 앞뒤로 3장, 총 6페이지, 본인이 직접 필기한 자료만 인정) 
  • 링크 및 시험 문제는 당일 오후 2시 25분에 블랙보드의 이메일 기능을 이용하여 전송
  • 기말고사 제출 방식
    • 시험이 종료되는 5시 30분 시점에서 답안지(A4 단면)를 페이지별로 휴대폰으로 촬영하여 담당교수 이메일(pilsung_kang@korea.ac.kr)로 전송 (촬영 및 발송시간 고려하여 이메일 발송시간 기준 5시 35분까지 인정)
    • 12월 25일(금) 자정까지 Cheating Sheet와 답안지 원본을 스테이플러로 결합하여 담당교수 연구실로 제출(창의관 801A호, 부재중일 경우 문 아래로 밀어넣기)

• 유튜브 강의영상 요약 및 논문 재현 포스팅 기한 공지

  • 유튜브 강의영상 요약 및 논문 재현 포스팅은 12월 27일(일) 23:59분까지 제출된 버전에 대해서만 인정하도록 하겠습니다.

• Syllabus (Document, Slide, Video)
• Tutorial resources (2015)
• Tutorial resources (2016)

Schedule

Topic 1: Dimensionality Reduction

• Dimensionality Reduction: Overview (Slide, Video)
• Supervised Methods 1: Forward selection, Backward elimination, Stepwise selection (Slide, Video)
• Supervised Methods 2: Genetic algorithm (Slide, Video)
• Unsupervised Method (Linear embedding) 1: Principal component analysis (PCA) (Slide, Video)
• Unsupervised Method (Linear embedding) 2: Multi-dimensional scaling (MDS) (Slide, Video)
• Unsupervised Method (Nonlinear embedding) 1: ISOMAP, LLE (Slide, Video)
• Unsupervised Method (Nonlinear embedding) 2: t-SNE (Slide, Video)
• Tutorial 1: Supervised Method
  • Tutorial Video (장명준), Tutorial Notes (최희정), Tutorial Notes (박경찬)
• Tutorial 2: Unsupervised Method (Linear embedding)
  • Tutorial Video (정재윤), Tutorial Notes (천우진), Tutorial Notes (이정호)
• Tutorial 3: Unsupervised Method (Nonlinear embedding)
  • Tutorial Video (서승완), Tutorial Notes (양우식), Tutorial Notes (손규빈)

Topic 2: Kernel-based Learning

• Theoretical foundation (Slide, Video)
• Support Vector Machine (SVM) - Linear & Hard Margin (Slide, Video)
• Support Vector Machine (SVM) - Soft Margin (Slide, Video)
• Support Vector Regression (SVR) (Slide, Video)
• Kernel Fisher Discriminant Analysis (KFDA) (Slide, Video)
• Kernel Principal Component Analysis (KPCA) (Slide, Video)
• Tutorial 4: Support Vector Machine (SVM)
  • Tutorial Video (성유연), Tutorial Notes (유초롱), Tutorial Notes (정민성)
• Tutorial 5: Support Vector Regression (SVR)
  • Tutorial Video (이민정), Tutorial Notes (이창현), Tutorial Notes (이지윤), Tutorial Notes (신유경)
• Tutorial 6: Kernel Fisher Discriminant Analysis (KFDA)
  • Tutorial Video (조윤상), Tutorial Notes (이민형), Tutorial Notes (송재승), Tutorial Notes (조수희)
• Tutorial 7: Kernel Principal Component Analysis (KPCA)
  • Tutorial Video (채선율), Tutorial Notes (홍연정), Tutorial Notes (이선화), Tutorial Notes (최지은)

Topic 3: Anomaly Detection

• Anomaly Detection: Overview (Slide, Video)
• Density-based Anomaly Detection Part 1: Gaussian Density Estimation & Mixture of Gaussian Density Estimation (Slide, Video)
• Density-based Anomaly Detection Part 2: Parzen Window Density Estimation (Slide, Video)
• Density-based Anomaly Detection Part 3: Local Outlier Factor (LOF) (Slide, Video)
• Distance/Reconstruction-based Anomaly Detection (Slide, Video)
• Model-based Anomaly Detection Part 1: Auto-Encoder, 1-SVM, and Support Vector Data Description (SVDD) (Slide, Video)
• Model-based Anomaly Detection Part 2: Isolation Forest and Extended Isolation Forest (Slide, Video)
• (Optional) Anomaly Detection with Generative Adversarial Network (Video, presented by 김창엽)
• (Optional) Graph-based Anomaly Detection (Video, presented by 김혜연)
• Tutorial 8: Density-based novelty detection
  • Tutorial Video (전창동, 오주혁), Tutorial Notes (이인구), Tutorial Notes(Justus Aaron Benning), Tutorial Notes (음수민)
• Tutorial 9: Distance/Reconstruction-based novelty detection
  • Tutorial Video (옥명훈), Tutorial Notes (김영호), Tutorial Notes (김위백), Tutorial Notes (손주희)
• Tutorial 10: Model-based novelty detection
  • Tutorial Video (송서하, 최현율), Tutorial Notes (윤혜빈), Tutorial Notes (박혜준), Markdown File, Tutorial Notes (박성훈)

Topic 4: Ensemble Learning

• Overview (Slide, Video)
• Bias-Variance Decomposition (Slide, Video)
• Bagging (Slide, Video)
• Bagging: Random Forests (Slide, Video)
• Boosting 1 - Adaptive Boosting (AdaBoost) (Slide, Video)
• Boosting 2 - Gradient Boosting Machine (GBM) (Slide, Video)
• Boosting 3 - XGBoost (Slide, Video)
• Boosting 4 - Light GBM (Slide, Video)
• Boosting 5 - CatBoost (Slide, Video)
• (Optional) XGBoost (Video, presented by 윤훈상)
• (Optional) CatBoost (Video, presented by 김지나)
• Tutorial 11: Bagging
  • Tutorial Video (이주한), Tutorial Notes (김문수), Tutorial Notes (양형원), Tutorial Notes (윤석채)
• Tutorial 12: AdaBoost, Gradient Boosting
  • Tutorial Video (김명소), Tutorial Notes(황호선), Tutorial Notes (이주현), Tutorial Notes (유건령)
• Tutorial 13: Random Forests, Decision Jungle (임희찬, 권상현)
  • Tutorial Video (임희찬, 권상현), Tutorial Notes (고봉균) , Tutorial Notes (조영선), Tutorial Notes (정회빈)

Topic 5: Semi-supervised Learning

• SSL: Overview (Slide, Video)
• SSL: Self-training & Co-Training (Multi-view algorithm) (Slide, Video)
• SSL: Graph-based SSL (Slide, Video)
• SSL: Generative Models (Slide, Video)
• (Optional) Text Augmentation (Video, presented by 김정희)
• (Optional) Semi-supervised learning with ladder network (Video, presented by 양우식)
• (Optional) MixMatch (Video, presented by 이정훈)
• (Optional) Remixaatch & FixMatch (Video, presented by 이정훈)
• Tutorial 14: Self-training
  • Tutorial Video (김우일), * Tutorial Notes (김영준), Tutorial Notes (김진국)
• Tutorial 15: Generative models
  • Tutorial Video (강성호), Tutorial Notes (강희종), Tutorial Notes (박건빈), Tutorial Notes (이효연)
• Tutorial 16: Graph-based SSL
  • Tutorial Video (안건이), Tutorial Notes (변준형), Tutorial Notes (이수빈), Tutorial Notes (윤도균)
• Tutorial 17: Multi-view algorithm (Co-training)
  • Tutorial Video (이준헌), Tutorial Notes (문준), Tutorial Notes (진세종), Tutorial Notes (백인성)