Guest lecture: 100 Years of Data Mining Using Seriation and Matrix Reordering
Dr. Innar Liiv from Tallinn University of Technology will give a guest lecture on Thursday 30th of October at 13:15 in Exactum C221.
Abstract:
Seriation is an exploratory combinatorial data analysis technique to reorder objects into a sequence along a one-dimensional continuum so that it best reveals regularity and patterning among the whole series. Unsupervised learning, using seriation and matrix reordering, allows pattern discovery simultaneously at three information levels: local fragments of relationships, sets of organized local fragments of relationships and an overall structural pattern. It, therefore, combines, in a single result, and enhances the structural analysis abilities of popular unsupervised data mining methods, like clustering and association rules. The lecture will advocate that seriation should be put on a par with those standard data mining methods due to their lack of ability to analyze complex structures and to defocus from details to global relationships.
Several applications from the following disciplines are included in the retrospective review: archaeology and anthropology; cartography, graphics and information visualization; sociology and sociometry; psychology and psychometry; ecology; biology and bioinformatics; cellular manufacturing, and operations research. Recent related developments in data mining (biclustering, coclustering) will be discussed.
A unified view and an objective function will be presented for the parameter-free seriation, based on Kolmogorov complexity and data compression. The main goal of such measure is to enhance repeatability, scrutiny and rigorous benchmarking ability for seriation research.
Keywords: seriation, two-mode clustering, data mining, biclustering, coclustering, combinatorial data analysis, minimum description length principle, information visualization
About the speaker:
Innar Liiv received his Ph.D. in Computer Science in 2008 from Tallinn University of Technology. His research interests include philosophy of artificial intelligence, data mining (seriation, clustering, association rules), predictive analytics (churn, fraud, lifetime value), operations research, business intelligence in logistics, information visualization and social network analysis. He is also an editor of http://www.kmining.com, one of the TOP3 data mining webpages.
Welcome!