Speaker
Description
Unsupervised machine learning algorithms for complex data
As part of the workshop, participants will be familiarized with the field of artificial intelligence and machine learning, unsupervised learning. In practical applications, using this learning method is much more often necessary. We do not know how to classify data (texts, images, sounds), and we analyze data in terms of, among others, similarities by creating structures that allow you to identify the group to which the recognized object belongs quickly. The topics discussed during the workshop will include methods such as cluster analysis and outlier mining. An essential part of the workshop will be a comparative analysis of various cluster analysis algorithms regarding the type of data analyzed and input parameters affecting the final result, i.e., the created group structure. The student will learn the methods of data similarity analysis and the methods of creating representatives of the created groups. Then he will learn about the methods of searching the structures of clusters of objects. The result of the work will be a cluster analysis using different algorithms and different datasets to show which algorithms are suitable for analyzing a particular data type.