Introduction to Machine Learning PDF

IT1815 Introduction to Machine Learning Machine Learning Models Supervised Learning: The machine is provided with an Fundamentals...

IT1815 Introduction to Machine Learning Machine Learning Models Supervised Learning: The machine is provided with an Fundamentals established set of data (labeled data). The categories of supervised Artificial intelligence is the science of training machines to learning are: perform human tasks. o Classification Subsets of artificial intelligence (Hurwitz, 2018): o Regression – predicting the values of a continuous o Reasoning - making inferences based on data variable o Natural Language Processing (NLP) – understanding Ex. Weather forecasting provides a prediction on the both written text and human speech weather based on historical weather patterns and the o Planning – acting autonomously and flexibly to construct a current conditions. sequence of actions to reach a final goal Unsupervised learning: The machine is provided with a set of data o Machine Learning but is not provided with a specific answer (unlabeled data). The Machine learning is a form of artificial intelligence that enables a machine may identify trends of similarity through clustering. system to learn from data, identify patterns, and make decisions Ex. Identifying if an e-mail is a spam or not with minimal human intervention. Semi-supervised learning: The machine learns from a dataset Common examples of machine learning: that includes both labeled and unlabeled data, usually most o When you search and view products in an online shopping unlabeled. site, you may encounter other similar products that you may Reinforcement learning: The machine learns through trial and find interesting the next time you go online. error. It is often used for robotics, gaming, and navigation. It has o Virtual assistants like Siri and Alexa, are asked over voice three (3) primary components. when you need to find information. Ex. A self-driving car that picks the best lane to go to o Transportation apps suggest faster routes and computes o Agent – the learner or decision maker (self-driving car) prices of rides based on the rider demand. o Environment – everything the agent interacts with (streets o Facebook suggests people that you may want to become other cars, human drivers, etc.) friends with and recognizes you and your friends' faces o Action – what the agent can do (drive, pick lane, etc.) when you upload photos. o Chatbots serve as customer service agents when you can't reach the company hotline. Purpose of machine learning: o Classification – classifying an item into a distinct set of categories (Ex. Recognizing the object in an image) o Clustering – grouping items with common properties (Ex. Facebook suggesting people to add in a group) o Prediction – forecasting future values (Ex. Demand for products around holiday season) A machine learning model is the output generated when you train your machine learning algorithm with data. (Hurwitz, 2018) 11 Handout 1 *Property of STI  [email protected] Page 1 of 2  IT1815 Sample Programs Python Java References: Hurwitz, J., & Kirsch, D. (2018). Machine learning for dummies, IBM limited edition. Hoboken: John Wiley & Sons, Inc. Rebala, G., Ravi, A., & Churiwala, S. (2019). An introduction to machine learning. Hyderabad: Springer Nature Switzerland AG 2019. Oracle Docs (n.d.). Citing sources. Retrieved from https://docs.oracle.com/javase/8/docs/api/java/util/package-summary.html Teach with ICT (n.d.). Citing sources. Retrieved from http://www.teachwithict.com/ Sample Output: 11 Handout 1 *Property of STI  [email protected] Page 2 of 2