What does it mean to "Measure Performance"?
Professor Alistair Moffat, University of Melbourne
[abstract] [about the speaker]
Trustworthy Computing: Is it useful?
Professor Vijay Varadharajan , Macquarie University
[abstract] [about the speaker]
Towards Next Generation Web Information Retrieval
Dr Wei-Ying Ma , Microsoft Research Asia
[abstract] [about the speaker]
What does it mean to "Measure Performance"?
Professor Alistair Moffat, University of Melbourne
[Abstract] After inventing a new indexing structure or query processing algorithm, we (should) seek to implement it and measure its behavior, thereby completing the cycle of experimental science. That is, having developed what we believe to be an innovation, we need to gather evidence to convince others of its worth. Testing an algorithm tends to consist of coding up the simplest baseline we think we can get away with (or downloading someone else's implementation); coding up our improved process; running both on some test data using whatever machine is handiest; and measuring the amount of CPU or elapsed time that was consumed. Using this evidence, we draw both graphs and conclusions, often rather more of the former than the latter.
Unsurprisingly, we almost always obtain improvements, and rarely admit to having invented "inferior'' techniques. All too often, however, our work fails to earn any citations, because the evidence does not meet basic standards of rigor or persuasiveness.
In this talk the notion of "experiment'' is explored, with particular reference to the issue of distributed text query processing on a tightly-coupled cluster of computers. Pitfalls that might deceive the unwary are described, in the hope that the lessons learnt in our experiments will be applied by researchers designing experiments in other areas of computing.
[About Speaker] Alistair Moffat completed his PhD at the University of Canterbury in New Zealand in 1986. Since then he has been a member of the academic staff at The University of Melbourne, where he has worked with students and colleagues to explore a wide range of topics relating to source coding, text compression and modelling, index representations and structures, and text query evaluation heuristics. Moffat is co-author of the books Managing Gigabytes (second edition, Morgan Kaufmann, 1999) and Compression and Coding Algorithms (Kluwer, 2002), as well as a C programming textbook, and more than 100 refereed papers. Moffat is a co-program chair of the 2005 ACM SIGIR Conference on Information Retrieval, and currently serves on the ARC Expert Advisory Committee for Mathematics, Information, and Communications.
[return to WISE Keynote Speeches]
Trustworthy Computing: Is it useful?
Professor Vijay Varadharajan , Macquarie University
[Abstract] The notion of trust has been around for many decades (if not for centuries) in different disciplines in different disguises. In particular, the concept of trust has been studied extensively in various disciplines such as psychology, philosophy, economics, sociology as well as in technology. Yet we do not have a clear handle on this concept of trust which is increasingly becoming significant in our information economy and the Internet world. It is often quoted (and probably correctly) that trust is a key foundation stone of the information security world. In security, the concept of “trusted systems” has at least been around publicly for some 25 years and more recently over the last couple of years the idea of “trustworthy computing” is beginning to take a centre stage, with the launch of major initiatives in trustworthy computing by several major players in the information technology arena.
In this talk, we will take a look at the concept of trust in the secure computing world and see some of the challenges and pitfalls involved. We will try to extract some fundamental issues that could help in outlining the key elements of a broad framework for trustworthy computing. We will then reflect some of these elements within the context of the design of secure systems in a pervasive web based computing environment such as the one we are in today. I believe a serious discussion of challenges in trustworthy computing and their implications is important not only to understand better its real benefits and drawbacks but also in identifying the hard problems that require research and advances in the computing discipline.
[About Speaker] Vijay Varadharajan is the Microsoft Chair and Professor of Computing at Macquarie University. He is also the Director of Information and Networked System Security Research. He is also the Technical Board Director of Computer Science, Australian Computer Society. Previous to this, he was the Foundation Chair Professor (1995-2000) and Head of School of School of Computing and IT (1997-2000) at the University of Western Sydney, Nepean, Australia. He was also the Director of the Distributed System and Network Security Research. Prior to taking up this appointment, Professor Varadharajan was responsible for worldwide Security Research at corporate Hewlett-Packard Labs based in Europe at HP Labs., UK. He has worked also with various HP Divisions in the US, UK, Germany, France and Italy. He has been with HP since 1988. Prior to HP, he was a Research Manager at British Telecom Research Labs. U.K., from 1987 till 1988. From 1985 till 1987, he was Research Fellow and Lecturer in Computer Science at Plymouth and Reading Universities. He did his Ph.D in Computer and Communication Security in the U.K in 1984, which was sponsored by BT Research Labs. He did his Electronic Engineering degree from Sussex University, UK (1978-1981).
Vijay has technically led, managed and successfully delivered on numerous projects in Secure Systems funded by European Commission, Hewlett-Packard Labs., SERC/EPSRC and British Telecom Labs. in the UK and USA, and Australian Research Council Grants and University Grants as well as sponsored projects from Fujitsu, Microsoft and other industries. He has also acted as consultant and has been architect for several projects in computing, financial and telecom organizations in the UK, US and more recently in Australia. He has also acted as an Expert for European Union and worked with the UK Dept. of Trade and Industry, and has worked several European Research Projects such as COST, EUREKA and ESPIRIT. Vijay has also been a member of several International Standards activities such as ISO, ECMA, OSF, CCITT/ITU, IEEE, British and Australian Standards and the European Union. He is currently the Australian Delegate on IFIP Technical Committee TC 11 on Information Security.
Vijay has published more than 200 papers in International Journals and Conferences, has co-authored and edited 8 books on Security, Networks and Distributed Systems and holds 4 patents. His research work over the years has contributed to the development of several secure commercial systems in the areas of Secure Distributed Applications, Secure Network Systems, Security Tools, Secure Mobile Systems as well as Cryptographic and Smart Card based Systems. His current areas of research interest include security in high speed networks, security for large distributed systems, security policies and management in distributed applications, Internet security, secure electronic commerce and payment models, secure mobile agents, wireless security, security models and architectures and security protocols.
He is a member of the Editorial Board of the Journal of Information Security, Springer-Verlag (Germany), ACM Transactions on Information Systems Security (USA) and in the Board of Computer and Communication Security Abstracts (UK). He has been a member of the Board of Advisors in Trusted Computing Platform Association (TCPA) and is on the Microsoft Trustworthy Computing Academic Advisory Board (TCAAB). He has been an invited Visiting Professor at INRIA Research Labs in France, Institute of Mathematical Sciences, National University of Singapore, Indian Institute of Science and Visiting Scientist at Microsoft Research Cambridge, UK. He is a Fellow of the British Computer Society (FBCS), a Fellow of the IEE (FIEE), a Fellow of the IMA (FIMA), a Fellow of the Australian Institute of Engineers (FIEAust), a Fellow of the Australian Computer Society and a Senior Member of IEEE.
[return to WISE Keynote Speeches]
Towards Next Generation Web Information Retrieval
Dr Wei-Ying Ma, Microsoft Research Asia
[Abstract] Today search engines have become one of the most critical applications on the Web, driving many important online businesses that connect people to information. As the Web continues to grow its size with a variety of new data and penetrate into every aspect of people’s life, the need for developing a more intelligent search engine is increasing. In this talk, we will briefly review the current status of search engines, and then present some of our recent works on building next generation web search technologies. Specifically, we will talk about how to extract data records from web pages using vision-based approach, and introduce new research opportunities in exploring the complementary properties between the surface Web and the deep Web to mutually facilitate the processes of web information extraction and deep web crawling. We will also present a search prototype that data-mines deep web structure to enable one-stop search of multiple online web databases. In contrast with current web search that is essentially document-level ranking and retrieval, an old paradigm in IR for more than 25 years, we will introduce our works in building a new paradigm called object-level web search that aims to automatically discover sub-topics (or taxonomy) for any given query and put retrieved web documents into a meaningful organization. We are developing techniques to provide object-level ranking, trend analysis, and business intelligence when the search is intended to find web objects such as people, papers, conferences, and interest groups. We will also talk about vertical search opportunities in some emerging new areas such as mobile search and media search. In addition to providing information adaptation on mobile devices, we believe location-based and context-aware search is going to be important for mobile search. We also think that by bridging physical world search to digital world search, many new user scenarios that do not yet exist on desktop search can potentially make a huge impact on the mobile Internet. For media search, we will present those new opportunities in analyzing the multi-typed interrelationship between media objects and other content such as text, hyperlinks, deep web structure, and user interactions for better semantic understanding and indexing of media objects. We will also discuss our goal of continually advancing web search to next level by applying data mining, machine learning, and knowledge discovery techniques into the process of information analysis, organization, retrieval, and visualization.
[About Speaker] Dr. Wei-Ying Ma is a Research Manager at Microsoft Research Asia (MSRA). He received the B.S. degree in electrical engineering from the National Tsing Hua University in Taiwan in 1990, and the M.S. and Ph.D. degrees in electrical and computer engineering from the University of California at Santa Barbara (UCSB) in 1994 and 1997, respectively. From 1994 to 1997 he was engaged in the Alexandria Digital Library (ADL) project in UCSB while completing his Ph.D. He developed a web-based image retrieval system called Netra which has been frequently cited by other researchers and is regarded as one of the most representative image retrieval systems. This system is the first large-scale image retrieval system which uses image segmentation to enable region-based search. From 1997 to 2001, he was with HP Labs where he worked in the field of multimedia adaptation and distributed media services infrastructure for mobile Internet. He initiated the works on multimedia adaptation services network that later became a major research focus for the lab. He joined Microsoft Research Asia in April 2001 as the Research Manager of the Web Search and Mining Group, leading the research in the areas of information retrieval, text mining, search, multimedia management, and mobile browsing. In the 2003 and 2004, his research group has published 9 full papers in SIGIR, 5 full papers in WWW, and 8 full papers in ACM Multimedia conference, which account for 5-10% of the total number of accepted papers in these conferences. He currently serves as an Editor for the ACM/Springer Multimedia Systems Journal and Associate Editor for the Journal of Multimedia Tools and Applications published by Kluwer Academic Publishers. He has served on the organizing and program committees of many international conferences including ACM Multimedia, ACM SIGIR, ACM CIKM, WWW, ICME, CVPR, SPIE Multimedia Storage and Archiving Systems, SPIE Multimedia Communication and Networking, etc. He is also the general co-chair of International Multimedia Modeling (MMM) Conference 2005 and International Conference on Image and Video Retrieval (CIVR) 2005.