Professor Carolyn Beck of the ISE department will be joining us to give a 30 minute seminar on k-means clustering, including an overview of its strengths and weaknesses, implementation, and Lloyd’s algorithm. Following a Q&A with Professor Beck, we will also be providing a tutorial on implementing the k-means algorithm on the famous Iris Flower data set. The tutorial will take place in Python, using Jupyter Notebook. It will be beneficial to bring your laptops but installing Jupyter is not required.
Where: TB 303
When: Wednesday 10/11 5:30pm
Guest Speaker: Professor Carolyn Beck, Ph.D.
k-means clustering is a standard tool in data analysis of numerical features, used to classify unlabeled data quickly. It is a heuristic but highly effective method used both in industry and academia based on the location of a k n-dimensional centroids. The k-means algorithm is both simple and intuitive, and forms the basis on which many more sophisticated methods of clustering are built. It finds use cases in cyber security classifying potentially malicious data, in online retail classifying customer types to determine what items to show, and in campaign-planning to identify what opinions voters hold on multiple, dependent issues.