Introduction To Machine Learning

Talks about Machine Learning and Artificial Intelligence

Transcript

1. Introduction to Machine Learning by dummy using
2. Why Machine Learning? • Series Startup Netflix • My wife is on postgraduate study about material physics using ML • 1on1 mentorship target • Prakerja project about Course Recommendation
3. What is Machine Learning Machine learning (ML) is the study of computer algorithms that improve automatically through experience and by the use of data. Mitchell, Tom (1997). Machine Learning. New York: McGraw Hill. ISBN 0-07-042807-7. OCLC 36417892
4. How is machine learning different from traditional programming? Traditional Programming Machine Learning
5. How is machine learning different from traditional programming? Traditional Programming Machine Learning
6. Machine learning flow
7. Step 1: Define your machine learning problem Binary classification Multi-class classification Regression Catalog organization Generative model
8. Binary classification Multi-class classification Regression Catalog organization Generative model
9. Binary classification Multi-class classification Regression Catalog organization Generative model
10. Binary classification Multi-class classification Regression Catalog organization Generative model
11. Binary classification Multi-class classification Regression Catalog organization Generative model
12. Binary classification Multi-class classification Regression Catalog organization Generative model
13. Step 2: Acquire, get to know, & prepare your data Data types: • Tabular • Text • Sound • Image Where to get the data?: • Use a ready-to-use dataset • Extract the data by yourself Correct and Normalize data:
14. Step 3: Train your model Features Model Training
15. Step 3: Train your model Features Things that influence prediction result Example: 1. demographics, location, education history, employment history, remaining credit 2. username, anime_id, and my_score
16. Step 3: Train your model Model • A model maps examples to predicted labels • It is defined by weights that are learned during the training process • Once trained, you can use it to make predictions about data that it has never seen before
17. Step 3: Train your model Train Steps: 1. Separate data to be trained and to be evaluated 2. Iterate training by modifying feature and/or modify how to train your dragon (epoch or learning rate) Our model does not get smart right away - it needs to be “trained” Using available Library: - CoreML - fastText
18. Step 4: Use the model to make predictions
19. Prakerja Phase 1: Content Filtering Phase 2: Collaborative Filtering
20. Supervised Unsupervised
21. Demo Tutorial