Tutorial 1: Empirical Methods for Artificial Intelligence

Tutorial 2: Writing and Presenting Scientific Papers

Tutorial 3: Agent & Data Mining: the Synergy to Empower Intelligent Information Processing and Systems



Empirical Methods for Artificial Intelligence

Paul Cohen
University of Arizona 

Abstract: The purpose of this tutorial is to introduce attendees to empirical tools including experiment and metrics design, exploratory data analysis, statistical hypothesis testing, and modeling.   Through case studies and other examples, this tutorial introduces a variety of evaluation methods, including methods for visualizing, summarizing and detecting clues in data; for designing controlled experiments; and for testing hypotheses, both with classical tests and with new, computer-intensive bootstrap and randomization methods. The case studies are organized around ten themes, each a mantra that I find helpful when conducting evaluations (e.g., “follow the variance,” and “evaluation begins with claims.”).

The tutorial does not require any prior knowledge of the subject, and specifically will not require background in statistics. The tutorial will therefore is suitable for the general AI audience, both academic and industrial.

Tutorial material will be based on three sources: my book “Empirical Methods for Artificial Intelligence” (MIT Press, 1995), lectures from my graduate-level course at the University of Massachusetts, and tutorial notes from previous tutorials. 

Biography: Paul R. Cohen is Professor and Head of Computer Science at the University of Arizona.  His PhD is from Stanford University in Computer Science and Psychology, in 1983; and his MS and BA degrees in Psychology are from UCLA and UC San Diego, respectively.  He consults to several government agencies on the design of evaluations for programs, and has published widely in methodology and in several areas of AI.

Homepage: http://www.cs.arizona.edu/~cohen/


Writing and Presenting Scientific Papers

Tu-Bao Ho
Japan Advanced Institute of Science and Technology

Abstract: The purpose of this tutorial is to introduce attendees, especially the students and young researchers, to the main issues when wring and presenting scientific papers. The tutorial starts with an emphasis on the importance of writing papers as a part of the research work then a general design of a scientific paper. Through common parts of scientific papers, this tutorial introduces the key ideas about what and how to write each of these parts. Finally, some general rules of preparing slides and giving a talk will be presented and discussed.

Biography: Tu Bao Ho is Professor of School of Knowledge Science at Japan Advanced Institute of Science and Technology. He received a bachelor degree in applied mathematics from Hanoi University of Technology (1978), M.S. and Ph.D. degrees in Computer Science from Pierre and Marie Curie University, Paris (1984, 1987), and Habilitation from Paris Dauphine University (1998). His research interest includes artificial intelligence, knowledge-based systems, machine learning and data mining, and computational science. He is with Editorial Board of several Journals and author, co-author of numerous papers in journals and refereed conferences.

Homepage: http://www.jaist.ac.jp/~bao/



Agent & Data Mining: the Synergy to Empower Intelligent Information Processing and Systems

Longbing Cao and Chengqi Zhang
University of Technology, Sydney, Australia

Abstract: Two originally separated areas, agents (includes autonomous agent and multi-agent systems) and data mining (includes knowledge discovery) are getting increasingly interrelated in the need of both parties. Such interaction and integration features a bilateral complementation and synergetic enhancement of intelligence and infrastructure for information processing and systems. This tutorial draws an overall picture of this new area in the scientific family. It covers key contents including the driving forces for the interaction, field structure, state-of-the-art, typical techniques, and case studies. It features both theoretical and practical innovation and applications of this new area. Case studies on data mining driven trading agents and agent-based financial data mining system will be illustrated. Finally, we discuss the challenges, open issues and prospects of agent-mining interaction.

Biography: Dr Longbing Cao is an IEEE Senior Member, an Associate Professor of Information Technology at the University of Technology Sydney, the Director of Data Sciences and Knowledge Discovery Lab, and the Data Mining Research Leader of Australian Capital Markets Cooperative Research Centre. His research interests include data mining and knowledge discovery, multiagent and intelligent systems, and the integration of multiagent and data mining. He has published around 100 refereed journal and conference papers in the above areas. He has made significant commitments to the research & development of real-life large data mining and complex intelligent applications in telecommunications, social security, governmental services, and financial markets. In the emergent area of Agent-Mining Interaction and Integration (AMII), he is one of the global pioneers as evidenced by organizing Agents and Data Mining Interaction and Autonomous Intelligent Systems-Agents and Data Mining workshop series, a tutorial on WI-IAT2007, a monograph, an edited book, two special issues, and a Special Interest Group on AMII (www.agentmining.org). In this particular area, his research covers (i) data mining driven trading agents, (ii) agent-driven data mining systems, and (iii) mutual issues in agent-mining interaction.

Homepage: http://www-staff.it.uts.edu.au/~lbcao/

Biography: Chengqi Zhang is a Research Professor of Information Technology at University of Technology, Sydney (UTS) since December 2001. He is currently the Director of UTS flagship Research Centre for Quantum Computation and Intelligent Systems which is one of five research flagships at UTS. In addition, he is the Leader of Data Mining program of Australian Capital Market Cooperative Research Centre and he is the Chairperson of Australian Computer Society’s National Committee for Artificial Intelligence. Chengqi Zhang had obtained his PhD degree from Queensland University in 1991 and Doctor of Science (DSc) which is the Higher Doctorate which in 2002.

Prof. Zhang’s research interests include “Multi-Agent Systems”, “Data Mining”, and their integrations. He has published more than 200 research papers. The most notable paper is the one which was published in “Artificial Intelligence” in 1992 which was the most prestigious Journal in Artificial Intelligence field. Furthermore, he has published many papers in the first class international journals, such as IEEE ad ACM Transactions. Prof. Zhang has supervised several PhD students in the area of agent and mining interactions. He has presented the invited talk in AIS-ADM 2005 and he was one of the organizers for ADMI 2006, AIS-ADM 2007.

Homepage: http://www-staff.it.uts.edu.au/~chengqi/











[ HOME ]