The big data revolution is underway. Are you ready to embrace the machine learning innovations that can lead your company to the top? Machine Learning Foundations and Frameworks provides essential grounding in the tools and techniques that comprise the field of machine learning. You will discuss the benefits and limitations of machine learning when compared to traditional statistics; illustrate supervised, unsupervised and reinforcement learning; develop research plans for classification and regression; interpret research results from machine learning; and recommend deep learning methods for intelligent systems.
No Mandatory Login Times
4-6 Hours of Work per Module
“To be a data scientist, you have to be multilingual: You have to speak the language of business, the language of statistics and the language of information technology. Think of data science as the discipline and machine learning as a technology or group of technologies within that discipline.”
Describe the role of the data scientist. Explain how probability is a measure of uncertainty. Discuss benefits and limitations of traditional statistics versus machine learning. Contrast supervised, unsupervised and reinforced learning.
Contrast levels of measurement. Recommend methods of data transformation. Organize data for modeling. Recommend methods for converting text to numbers.
Justify methods of data selection. Design sampling and resampling plans. Recommend methods for correcting bias and variance. Interpret results from model training and testing.
Identify variable roles in models. Illustrate traditional linear regression. Design machine learning models for regression. Evaluate results from regression analysis.
Illustrate traditional classification research. Design machine learning models for classification. Interpret classification research results. Design research for addressing issues with classification models.
Illustrate methods for measuring distance between objects. Contrast partitioning and hierarchical methods of cluster analysis. Describe methods of dimension reduction. Design machine learning methods for autoencoders.
Define deep learning. Describe alternative neural network designs. Illustrate convolutional neural networks and recurrent neural networks.
Contrast logic programming and machine learning approaches to artificial intelligence. Design research for image processing. Design research for natural language processing. Evaluate methods for machine intelligence.
Learn about Turing Tests, robotic automation, knowledge-based systems and recursive neural networks. Enroll in Machine Learning Cases from EmergingEd.
Learn about data governance, database systems, API standards and more. Enroll in Machine Learning and Industry from EmergingEd.
Thomas W. Miller is faculty director of the data science program at Northwestern University. He designed distance learning training materials for the program, including courses in advanced modeling techniques, marketing analytics, data engineering and machine learning. During the 2019-20 academic year, he will be teaching artificial intelligence and deep learning, natural language processing, and knowledge engineering. Dr. Miller has published six books about data science with Pearson Education (the series "Modeling Techniques in Predictive Analytics"). He also owns a publishing and consulting firm, Research Publishers LLC, located in Manhattan Beach, California. He provides data science consulting services and is actively involved in developing chatbot and survey research applications. Earlier in his career, Dr. Miller worked as a network engineer for NCR Comten and as a field engineer and account representative for Hewlett-Packard Company. He also directed the A.C. Nielsen Center for Marketing Research and taught market research and business strategy at the University of Wisconsin-Madison.
Dr. Miller holds a doctorate in psychology and a master's degree in statistics from the University of Minnesota, as well as an MBA and a master's degree in economics from the University of Oregon.