| Group | Title | Slides |
|---|---|---|
| 1 | Music Genre Classification | |
| 2 | Quora Question-pair classification | PPTX |
| 3 | Text sentiment analysis | PPTX |
| 4 | KKBox's churn prediction | PPTX |
| 5 | Guitar playing techniques recognition | PPTX |
| 6 | Facial expression recognition | PPTX |
| 7 | Movie box prediction from forum reviews | PPTX |
| 8 | Mercari price suggestion | PPTX |
| 9 | Problem Classification in K-12 Question-Driven Learning | PPTX |
| 10 | Green bike demand analysis | PPTX |
| 11 | Taiwan stock market trend prediction | |
| 12 | Stock prediction | PPTX |
| 13 | Early Prediction to Students' Academic Performance | PPTX |
| 14 | Predicting traffic jam on the highway | PPTX |
| 15 | Traffic sign classification | PPTX |
| 16 | Predicting the first salary for fresh graduates | PPTX |
| 17 | Weather prediction | PPTX |
| 18 | Credit card fraud detection | PPTX |
| Week | Date | Content | Exercise |
|---|---|---|---|
| 1 | 9/12 | 1. Overview of the course 2. Introduction to ML 3. KNN 4. k-means | Exercise 1 (due: 9/25 23:59:59) |
| 2 | 9/19 | TA session: Introduction to Python and popular scientific libraries in Python | |
| 3 | 9/26 | 1. Distance measures 2. Decision tree | Exercise 2 (due: 10/9 23:59:59) |
| 4 | 10/3 | 1. Entropy 2. Decision tree 3. Linear regression | |
| 5 | 10/10 | Holiday | |
| 6 | 10/17 | 1. Linear regression and gradient descent 2. Linear regression and regularization (Lasso, Ridge, Elastic-net) | |
| 7 | 10/24 | 1. Matrix derivatives 2. Evaluation metrics for regression problem 3. Logistic regression | Exercise 3 (due: 11/6 23:59:59) |
| 8 | 10/31 | Invited talk (speaker: Johnson Hsieh, DSP cofounder and chief knowledge officer) | |
| 9 | 11/7 | 1. Logistic regression and gradient ascent 2. Precision, recall, ROC curve, and other measures 3. Poisson regression | |
| 10 | 11/14 | Midterm project proposal presentation | |
| 11 | 11/21 | 1. SVM 2. Lagrange multiplier 3. Regularized linear regression and classification | Exercise 4 (due:12/11 23:59:59) |
| 12 | 11/28 | Invited talks (speakers: (1) Kae Lou, data scientist at KKV; (2) Hwai-Jung Hsu, Assistant Professor at FCU) | |
| 13 | 12/5 | 1. Linear vs Kernel 2. Practical considerations 3. Ensemble methods | |
| 14 | 12/12 | Recommender systems | |
| 15 | 12/19 | Deep neural network | |
| 16 | 12/26 | Convolutional neural network | |
| 17 | 1/2 | Final project presentation | |
| 18 | 1/9 | 1. Recurrent neural network 2. final wrap-up | Final project report due |