Machine Learning
- 2. A Bit About Me…. Computer Consultant, Trainer and Educator Master of Science – Computer Science Artificial Intelligence and Expert Systems Interest: Automated Adaptive Learning Systems Master of Education – Education, Policy, Organization and Leadership eLearning in Higher Education Interest: Online Knowledge Communities
- 3. Learning Objectives What is Machine Learning?
- 4. Learning Objectives What is Machine Learning? What are some algorithms, types, and languages of Machine Learning?
- 5. Learning Objectives What is Machine Learning? What are some algorithms, types, and languages of Machine Learning? What are some applications for Machine Learning?
- 6. Learning Objectives What is Machine Learning? What are some algorithms, types, and languages of Machine Learning? What are some applications for Machine Learning? What are some current trends and research in Machine Learning?
- 7. Learning Objectives What is Machine Learning? What are some algorithms, types, and languages of Machine Learning? What are some applications for Machine Learning? What are some current trends and research in Machine Learning? What are some careers with Machine Learning skills?
- 8. Discussion Question What does machine learning mean to you?
- 9. What is Machine Learning? Machine learning is the subfield of computer science where computers “are given the ability to learn without being explicitly programmed” (Samuel, 1959).
- 10. Trivia Question Who Is known as the creator of modern computing?
- 11. Trivia Question Who Is known as the creator of modern computing? In the 1930’s, he described the “universal computing machine”.
- 12. Trivia Question Who Is known as the creator of modern computing? In the 1930’s, he described the “universal computing machine”. His initials are A.T.
- 14. Alan Turing Alan Turing described the “universal computing machine,” a “single machine that can be used to compute any computable sequence.” (Turing, 1936)
- 15. Trivia Question Who is one of the pioneers in Artificial Intelligence?
- 16. Trivia Question Who is one of the pioneers in Artificial Intelligence? Who was the first to illustrate machine learning.
- 17. Trivia Question Who is one of the pioneers in Artificial Intelligence? Who was the first to illustrate machine learning. His checkers-playing program was the world's first self-learning program.
- 18. Trivia Question Who is one of the pioneers in Artificial Intelligence? Was the first to illustrate machine learning. His checkers-playing program was the world's first self-learning program. His initials are A.S.
- 20. Arthur Samuel’s Gameof Checkers Arthur Samuel (1901–1990) was a pioneer of artificial intelligence research and was the first to illustrate the concept of machine learning in his Game of Checkers. His Checkers-playing Program (Samuels, 1959) appears to be the world's first self-learning program.
- 21. Look Ahead Through Tree of Possible Moves
- 22. Trivia Question Who Is known for illustrating artificial neural networks?
- 23. Trivia Question Who Is known for illustrating artificial neural networks? He created Perceptron
- 24. Trivia Question Who Is known for illustrating artificial neural networks? He created Perceptron His initials are F.R.
- 25. In The Beginning… Creatorof Modern ComputingThe Game of Checkers – Machine Learning Perceptron– Artificial Neural Network
- 26. Frank Rosenblatt’s Perceptron Frank Rosenblattcreatedthe Perceptronin 1957which was a first artificial neural network.
- 27. A Few Machine Learning Algorithms Decision Tree Learning Association Rule Learning Artificial Neural Networks
- 28. Decision Tree Learning Decision Tree Learning
- 29. Decision Tree Learning Decision Tree Learning Uses a decision tree as a predictive model, which maps observations about an item to conclusions about the item's target value.
- 30. Decision Tree Learning Decision Tree Learning Uses a decision tree as a predictive model, which maps observations about an item to conclusions about the item's target value.
- 31. Association Rule Learning Association Rule Learning
- 32. Association Rule Learning Association Rule Learning A method for discovering interesting relations between variables in large databases.
- 33. Association Rule Learning Association Rule Learning A method for discovering interesting relations between variables in large databases.
- 34. Association Rule Learning Association Rule Learning A method for discovering interesting relations between variables in large databases.
- 35. Artificial Neural Networks Artificial Neural Networks
- 36. Artificial Neural Networks Artificial Neural Networks Computations are structured in terms of an interconnected group of artificial neurons, processing information using a connectionist approach to computation.
- 37. Artificial Neural Networks Artificial NeuralNetworks Computations are structured in terms of an interconnected group of artificial neurons, processing information using a connectionist approach to computation. Modern neural networks are non-linear statisticaldata modeling tools.
- 38. Artificial Neural Networks Artificial NeuralNetworks Computations are structured in terms of an interconnected group of artificial neurons, processing information using a connectionist approach to computation. Modern neural networks are non-linear statisticaldata modeling tools.
- 39. Lisp, Prolog, et al Lisp created by John McCarthy in 1958 Prolog created by Alain Colmerauer and Philippe Roussel in 1972 Allows for the logic programming needed for traversal creation of the neural networks Recognizes the relationships between the data and their rules. Semantic nets represent knowledge in tree-like patterns connecting nodes and arcs based on these rules.
- 41. EMYCIN Expert System ( Van Melle, Shortliffe & Buchanan, 1981)
- 42. Current Machine Learning Languages MATLAB/Octave R Python Java Family C Family
- 44. Supervised Learning: Predictive Model Feature Supervised Learning Strategy Use a predictive model that is given clear instructions
- 45. Supervised Learning: Predictive Model Feature Supervised Learning Strategy Use a predictive model that is given clear instructions
- 46. Feature Supervised Learning Strategy Use a predictive model that is given clear instructions Algorithm Nearest neighbor, Naïve Bayes, Decision Trees,Regression Supervised Learning: Predictive Model
- 47. Feature Supervised Learning Strategy Use a predictive model that is given clear instructions Algorithm Nearest neighbor, Naïve Bayes, Decision Trees,Regression Supervised Learning: Predictive Model
- 48. Feature Supervised Learning Strategy Use a predictive model that is given clear instructions Algorithm Nearest neighbor, Naïve Bayes, Decision Trees,Regression Supervised Learning: Predictive Model
- 49. Feature Supervised Learning Strategy Use a predictive model that is given clear instructions Algorithm Nearest neighbor, Naïve Bayes, Decision Trees,Regression Supervised Learning: Predictive Model Decision Tree Classification
- 50. Feature Supervised Learning Strategy Use a predictive model that is given clear instructions Algorithm Nearest neighbor, Naïve Bayes, Decision Trees,Regression Supervised Learning: Predictive Model Decision Tree Classification
- 51. Feature Supervised Learning Strategy Use a predictive model that is given clear instructions Algorithm Nearest neighbor, Naïve Bayes, Decision Trees,Regression Use Predict the likelihood of an earthquakeor tornado Supervised Learning: Predictive Model
- 52. Feature Supervised Learning Strategy Use a predictive model that is given clear instructions Algorithm Nearest neighbor, Naïve Bayes, Decision Trees,Regression Use Predict the likelihood of an earthquakeor tornado Supervised Learning: Predictive Model
- 53. Feature Supervised Learning Strategy Use a predictive model that is given clear instructions Algorithm Nearest neighbor, Naïve Bayes, Decision Trees,Regression Use Predict the likelihood of an earthquakeor tornado Supervised Learning: Predictive Model
- 54. Unsupervised Learning: Descriptive Model Feature Unsupervised Learning Strategy Uses a descriptive model whereno target is set and no single featureis more importantthan the other.
- 55. Unsupervised Learning: Descriptive Model Feature Unsupervised Learning Strategy Uses a descriptive model whereno target is set and no single featureis more importantthan the other.
- 56. Feature Unsupervised Learning Strategy Uses a descriptive model whereno target is set and no single featureis more importantthan the other. Algorithm K-meansClustering Algorithm Unsupervised Learning: Descriptive Model
- 57. Feature Unsupervised Learning Strategy Uses a descriptive model whereno target is set and no single featureis more importantthan the other. Algorithm K-meansClustering Algorithm Unsupervised Learning: Descriptive Model
- 58. Feature Unsupervised Learning Strategy Uses a descriptive model whereno target is set and no single featureis more importantthan the other. Algorithm K-meansClustering Algorithm Use Predict which diseases are likely to occuralong with diabetes. Unsupervised Learning: Descriptive Model
- 59. Feature Unsupervised Learning Strategy Uses a descriptive model whereno target is set and no single featureis more importantthan the other. Algorithm K-meansClustering Algorithm Use Predict which diseases are likely to occuralong with diabetes. Unsupervised Learning: Descriptive Model
- 60. Feature Reinforcement Learning Strategy Trains itself on a continualbasis based on the environmentit is exposed to,and applies it’s enriched knowledgeto solve problems. Reinforcement Learning
- 61. Feature Reinforcement Learning Strategy Trains itself on a continualbasis based on the environmentit is exposed to,and applies it’s enriched knowledgeto solve problems. Reinforcement Learning
- 62. Feature Reinforcement Learning Strategy Trains itself on a continualbasis based on the environmentit is exposed to,and applies it’s enriched knowledgeto solve problems. Algorithm Markov Decision Process Reinforcement Learning
- 63. Feature Reinforcement Learning Strategy Trains itself on a continualbasis based on the environmentit is exposed to,and applies it’s enriched knowledgeto solve problems. Algorithm Markov Decision Process Reinforcement Learning
- 64. Feature Reinforcement Learning Strategy Trains itself on a continualbasis based on the environmentit is exposed to,and applies it’s enriched knowledgeto solve problems. Algorithm Markov Decision Process Use Self driving carsuse it to make decisionscontinuously on which routeto take and what speed to driveand so on… Reinforcement Learning
- 65. Feature Reinforcement Learning Strategy Trains itself on a continualbasis based on the environmentit is exposed to,and applies it’s enriched knowledgeto solve problems. Algorithm Markov Decision Process Use Self driving carsuse it to make decisionscontinuously on which routeto take and what speed to driveand so on… Reinforcement Learning
- 66. Machine Learning Work Flow "Machine Learning" emphasizes that the computer machine/program must do some work after it is given data. (Brand,2015)
- 67. Azure Machine Learning Workflow (Grondlund,2016)
- 68. Google and Facebook Google and Facebook use Machine Learning extensively to push their respective ads to the relevant users.
- 69. Banking and Financial Providers Banking and Financial Providers can use Machine Learning to predict the customers who are likely to default from paying loans or credit card bills.
- 70. Healthcare Providers Healthcare Providers can use Machine Learning to diagnose deadly diseases based on the symptoms of patients and tallying them with the past data of similar kind of patients.
- 71. Retailers Retailers can use Machine Learning to determine fast and slow moving products.
- 75. In Conclusion… Machine Learning is a subset of Artificial Intelligence.
- 76. In Conclusion… Machine Learning is a subset of Artificial Intelligence. It refers to the techniques involved in dealing with vast data, in the most intelligent fashion, (by developing algorithms) to derive actionable insights.
- 77. In Conclusion… Machine Learning is a subset of Artificial Intelligence. It refers to the techniques involved in dealing with vast data, in the most intelligent fashion, (by developing algorithms) to derive actionable insights. There are a wide variety of algorithms and techniques to aid in machine learning and the technique chosen is determined by what one wants the machine to learn.
- 80. Summary We offered a brief history and definition of Machine Learning
- 81. Summary We offered a brief history and definition of Machine Learning We explored different types and applications of Machine Learning
- 82. Summary We offered a brief history and definition of Machine Learning We explored different types and applications of Machine Learning We looked at current trends, research and careers in Machine Learning.
- 83. References Blank, S. (2014) Tools and Blogs for Entrepreneur.Retrievedfromhttps://steveblank.com/tools-and-blogs-for- entrepreneurs/. Chen, F. (2016). AI, Deep Learning, and Machine Learning a Prime. Retrieved fromhttp://a16z.com/2016/06/10/ai-deep- learning-machines/. Computer Visions.(2015). Deep Leariningverus Machine Learning. Retrieved from http://www.computervisionblog.com/2015/03/deep-learning-vs-machine-learning-vs.html. Grondlund,C.J..(2016). Introductionto machine learningin the cloud.Retrieved from https://docs.microsoft.com/en- us/azure/machine-learning/machine-learning-what-is-machine-learning. Khan, M. (2016). Minimal and clean Python implementations ofMachine Learningalgorithms.Great for learninghow these algorithms work! Retrieved from https://www.linkedin.com/groups/2642596/2642596-6204217888639934466 McCarthy, J. & Feigenbaum,E.(1990). In Memoriam ArthurSamuel: Pioneer in Machine Learning. AI Magazine. AAAI.11 (3). Retrieved fromhttp://www.aaai.org/ojs/index.php/aimagazine/article/view/840/758. Nvidia.(2016). What’s the Difference Between Artificial Intelligence, Machine Learning,and Deep Learning? Retrieved from https://blogs.nvidia.com/blog/2016/07/29/whats-difference-artificial-intelligence-machine-learning-deep-learning- ai/. Vidya.(2016). Machine LearningBasics. Retrieved from https://www.analyticsvidhya.com/blog/2015/06/machine- learning-basics/.
- 84. Questions? What are your thoughts, ideas, suggestions on Machine Learning?