The Expectation Maximization (EM) algorithm as implemented in Unsupervised Learning for optimizing untagged data clustering models will be explained in simple terms. An EM implementation running in Jupyter Notebook will be demoed live taking unlabeled data as input, performing the EM clustering logic, and displaying the resulting data clusters in colors.
This is a light level statistics/probability session on how Unsupervised (data with no labels) clustering algorithms may be implemented, in this case applying statistical Expectation Maximization principles. Its great to have an idea on Normal/Gaussian Distribution and how multi-modal probabilities are calculated. A simple high level description will be given during the presentation.
Amir Bahmanyari is an Advisory Engineer in the Dell Technologies’ Data-Centric Workload & Solutions team. Amir joined Dell Technologies Big Data Analytics team in late 2017. He helps Dell Technologies’ customers to build their Big Data solutions. Amir has been active in Artificial and Evolutionary Intelligence work since late 1980’s when he was a Ph.D. candidate student at Wayne State University, Detroit, MI. Prior to Dell, Amir worked for a major automotive company, a major financial company, several startups in the Silicon Valley and as a Big Data Analytics Platform Architect at a major retailer.
