Tutorial Sessions/Invited Talks

The tutorial will cover recent algorithmic and statistical work on identifying and exploiting "geometric" structure in large informatics graphs such as large social and information networks. Such tools (e.g., Principal Component Analysis and related non-linear dimensionality reduction methods) are popular in many areas of machine learning and data analysis due to their relatively-nice algorithmic properties and
their connections with regularization and statistical inference. These tools are not, however, immediately-applicable in many large informatics graphs applications since graphs are more combinatorial objects; due to the noise and sparsity patterns of many real-world networks, etc. Recent theoretical and empirical work has begun to remedy this, and in doing so it has already elucidated several surprising and counterintuitive properties of very large networks. Topics include: underlying theoretical ideas; tips to bridge the theory-practice gap; empirical observations; and the usefulness of these tools for such diverse applications as community detection, routing, inference, and visualization.

Audience

This tutorial will provide an opportunity for the data analysis community, including both mathematically-oriented researchers as well as practitioners, to learn about recent algorithmic advances for dealing with
very large social and information networks. Many of these algorithmic tools have implicit geometric properties associated with them; and these geometric properties often have implicit statistical properties and consequences that indicate where these tools are more or less useful in real-world applications. As such, this tutorial should be of interest to and accessible by a large fraction of the data analysis  community - including both: established researchers who have done work in this or related areas, as well as researchers whose interests are not directly in the topic of the tutorial; and graduate students and postdocs, as well as junior and more senior researchers.

Many of the algorithmic and statistical techniques to be discussed have a strong overlap with seemingly-different problems and questions in statistics, optimization, numerical analysis, and machine learning - these connections will be highlighted throughout. Relatedly, many of these questions have been studied by researchers in theoretical computer science, scientific computing, statistics, machine learning, and data analysis; the complementary aspects of these different approaches, including their applicability to solving real-world problems from different application domains, will be emphasized. Depending on one's background, one can expect to benefit in different ways from the tutorial. In particular: Practitioners of machine learning and data analysis should gain just enough insight into the theoretical underpinnings of relevant algorithms to see how and why algorithms work well or fail to work well in real-world settings; Application-oriented theorists should gain insight into how the inner-workings of algorithms have practical implications for machine learning and data analysis on large networks, as well as learn about interesting theoretical problems raise by recent empirical findings; and Knowledgeable members of the
data analysis community should gain a broad overview of the area of large-scale graph mining and network analysis, including where data analysis methods with which they are familiar are well-suited or ill-suited.

Data mining research and practice have been experiencing an extraordinary growth over the past decade‒so have privacy concerns. Progress in data mining has been pushing the envelope of reachable depth, information and knowledge extracted from vast amounts of data‒increasingly exposing your innermost characteristics, behaviors and habits. Advanced data mining techniques and analytics have been significantly benefiting organizations in both commercial and noncommercial sectors‒yet providing an unprecedented potential for abuse. Is the interplay of data mining and privacy a conflict in making? This pertinent matter has been approached variously. Privacy preserving data mining has been tackling the issue from algorithmic and technology angles. Laws and regulations enacted by countries have been addressing the issue from legislative angles. Best practices and conducts instituted by commercial and international bodies have been exploring self-regulatory angles. Bridging data mining and privacy requires interdisciplinary endeavor. We will concisely survey the status quo and highlight selected promising directions.

Dr. Peter Geczy is a chief scientist at The National Institute of Advanced Industrial Science and Technology (AIST). He also held positions at The Institute of Physical and Chemical Research (RIKEN) and The Research Center for Future Technologies. His interdisciplinary scientific interests encompass domains of data and web mining, human interactions and behavior, social intelligence technologies, privacy, information systems, knowledge management and engineering, artificial intelligence, and adaptable systems. His recent research focus also extends to the spheres of service science, engineering, management, and computing. He received several awards in recognition of his accomplishments. Dr. Geczy has been serving on various professional boards and committees, and has been a distinguished speaker in academia and industry.

Proactive personalized medicine is expected to bring fundamental changes, offering recommendations of lifestyle adjustments and treatments to avoid diseases a patient has high risk for developing in the future. Due to common genetic, molecular, environmental, and lifestyle-based individual risk factors, most diseases do not occur in isolation. No matter how unique our medical experiences, chances are that other patients among millions have experienced genetic and environmental risk factors that closely mirror ours. In this talk, I will present our work that builds a comprehensive recommendation system, called CARE (Collaborative Assessment and Recommendation Engine), by pulling in experience of millions of patients to answer the question. I will also present our work on multi-relational representation of disease networks using both genetic knowledge, based on previously discovered gene-disease associations and phenotypic data from real patient histories.

Nitesh Chawla is an Assistant Professor in the Department of Computer Science and Engineering at the University of Notre Dame. He directs the Data Inference Analysis and Learning Lab (DIAL) and co-directs the Interdisciplinary Center of the Network Science and Applications (iCenSA) at Notre Dame. His research is primarily focused on machine learning, data mining, and social and dynamic networks. His work has led to applications in various domains including biology, medicine, finance, security, social science, fraud detection, intrusion detection, and text categorization. He is on the editorial board of IEEE Transactions on Systems, Man and Cybernetics Part B. He has received various awards and acknowledgements. He received the NAE FIE New Faculty Fellowship in 2005. His current research is supported form NSF, DOD, NWICG, NIJ, and industry sponsors.

Robert Stahlbock
General Conference Chair

E-mail: conference-chair@dmin--2011.com

Nikolaos Kourentzes

Programme Chair

E-mail: programme-chair@dmin--2011.com

Philippe Lenca, Gary M. Weiss

Tutorial Chair

E-mail: tutorial-chair@dmin--2011.com

This website is hosted by the Lancaster Centre for Forecasting at the Department of Management Science at Lancaster University Management School.