MachineLearning @ TWCL

Transcript

1. Dive Into Machine Learning tewei 2014/10/22
2. Why this talk?
3. Why this talk? 潮!!!
4. Why this talk? 潮!!! 其實我不知道ml是什麼._.
5. Why this talk? 潮!!! 其實我不知道ml是什麼._. 上了一個星期的自虐營
6. Why this talk? 潮!!! 其實我不知道ml是什麼._. 上了一個星期的自虐營 於是我現在就在這了...
7. What’s ML? https:/ /www.youtube.com/watch?v=FC3IryWr4c8 http:/ /www.informatik.uni-hamburg.de/ML/ml-wordle.jpg
8. Types of ML
9. Types of ML Supervised (With Label)
10. Types of ML Supervised (With Label) Regression (Output Continuous)
11. Types of ML Supervised (With Label) Regression (Output Continuous) Clustering (Output Discrete)
12. Types of ML Supervised (With Label) Regression (Output Continuous) Clustering (Output Discrete) Unsupervised (Without Label)
13. Types of ML Supervised (With Label) Regression (Output Continuous) Clustering (Output Discrete) Unsupervised (Without Label) Reinforcement (Long-Term Reward)
14. Applications
15. Applications Data Mining
16. Applications Data Mining Speech Recognition
17. Applications Data Mining Speech Recognition Computer Vision
18. Applications Data Mining Speech Recognition Computer Vision Bioinformatics
19. Applications Data Mining Speech Recognition Computer Vision Bioinformatics Recommendation System
21. Sky’s the limit!!!
23. Algorithms
24. Algorithms 只有一點點，我說真的xP
25. Decision Tree ID3, C4.5 algorithm Demo!!!
27. Orange
28. Check for base cases For each attribute a Find the normalized information gain ratio from splitting on a Let a_best be the attribute with the highest normalized information gain Create a decision node that splits on a_best Recurse on the sublists obtained by splitting on a_best, and add those nodes as children of node
29. Result
30. Support Vector Machine http:/ /www.csie.ntu.edu.tw/~cjlin/ libsvm/
31. Probabilistic Graphical Models A: Markov Network; B: Bayesian Network
32. CNN Convolutional Neural Network Pattern Recognition, Computer Vision
33. Getting Started Orange http:/ /orange.biolab.si/getting-started/ Sci-Kit Learn http:/ /scikit-learn.org/stable/tutorial/ machine_learning_map/index.html http:/ /www.astroml.org/ sklearn_tutorial/general_concepts.html
34. Learn More https:/ /www.coursera.org/course/ ntumlone https:/ /www.coursera.org/course/ml https:/ /www.coursera.org/course/mmds 這些好像都是簡單的囧...
35. http:/ /nirvacana.com/thoughts/wp-content/uploads/2013/07 /RoadToDataScientist1.png