|
|
|
|
|
|
|
|
|
 |
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Subscribe]
DM: AAAI Fall Symposium on AI and Link AnalysisFrom: David Jensen Date: Fri, 30 Jan 1998 16:29:27 -0500 (EST) <html><BR> <DIV> 1998 AAAI Fall Symposium on</DIV> <DIV> Artificial Intelligence and Link Analysis</DIV> <DIV> Orlando, Florida - October 23-25, 1998</DIV> <DIV> <A href="http://eksl-www.cs.umass.edu/aila/" EUDORA = AUTOURL>http://eksl-www.cs.umass.edu/aila/</A></DIV><BR> <DIV>Computer-based link analysis is increasingly used in law </DIV> <DIV>enforcement investigations, fraud detection, telecommunications </DIV> <DIV>network analysis, pharmaceuticals research, epidemiology, and </DIV> <DIV>many other specialized applications. Link analysis explores </DIV> <DIV>associations among large numbers of objects of different types. </DIV> <DIV>For example, a law enforcement application might examine familial </DIV> <DIV>relationships among suspects and victims, the addresses at which </DIV> <DIV>those persons reside, and the telephone numbers that they called </DIV> <DIV>during a specified period. The ability of link analysis to </DIV> <DIV>represent relationships and associations among objects of </DIV> <DIV>different types has proven crucial in assisting human </DIV> <DIV>investigators to comprehend complex webs of evidence and draw </DIV> <DIV>conclusions that are not apparent from any single piece of </DIV> <DIV>information.</DIV><BR> <DIV>However, there is both a need and opportunity to apply new </DIV> <DIV>technologies. Much of the current software for link analysis is </DIV> <DIV>little more than a graphical display tool. While visualizing </DIV> <DIV>networks has proven useful, many advanced applications of link </DIV> <DIV>analysis involve thousands of objects and links as well as a rich </DIV> <DIV>array of possible data models. Manual construction and analysis </DIV> <DIV>of such networks has proven difficult. In addition, a large </DIV> <DIV>number of related techniques in artificial intelligence and </DIV> <DIV>several other fields have the potential to assist human reasoning </DIV> <DIV>about complex networks of relationships. These techniques draw </DIV> <DIV>on work from search, semantic networks, ontological engineering, </DIV> <DIV>autonomous agents, inductive logic programming, graph theory, </DIV> <DIV>social network analysis, knowledge discovery in databases, </DIV> <DIV>entity-relationship modeling, information extraction, information </DIV> <DIV>retrieval, and metaphor.</DIV><BR> <DIV>This two-and-a-half day symposium will bring two communities into </DIV> <DIV>contact: 1) Members of the research community who currently have </DIV> <DIV>(or could soon develop) useful technology; and 2) Users of link </DIV> <DIV>analysis techniques whose needs go beyond the capabilities of </DIV> <DIV>current software. Note that the focus of the symposium is new </DIV> <DIV>technologies, not capabilities and applications embodied in </DIV> <DIV>current software such as Netmap (Alta Analytics, Inc.), Watson </DIV> <DIV>(Harlequin, Inc.), and the Analyst's Notebook (i2 Ltd.). These </DIV> <DIV>products have enabled current applications and may eventually </DIV> <DIV>incorporate new technologies. However, the focus of the workshop </DIV> <DIV>is on techniques that can be developed and deployed within 3-5 </DIV> <DIV>years.</DIV><BR> <DIV>The Symposium will be held Friday through Sunday, October 23-25, </DIV> <DIV>1998 at the Omni Rosen Hotel in Orlando, Florida. The Symposium </DIV> <DIV>is part of the AAAI Fall Symposium Series, which includes seven </DIV> <DIV>other symposia on a variety of topics. An informal reception </DIV> <DIV>will be held on Friday, October 23. A general plenary session </DIV> <DIV>featuring the highlights of each symposium will be presented on </DIV> <DIV>Saturday, October 24.</DIV><BR> <DIV>Important dates:</DIV> <DIV> April 15 Submissions due</DIV> <DIV> May 15 Notification of acceptance</DIV> <DIV> August 21 Material for working notes due</DIV> <DIV> October 23-25 Symposium held</DIV><BR> <DIV>Potential participants should submit either a 5-8 page extended </DIV> <DIV>abstract describing a relevant AI technology or application for </DIV> <DIV>advanced link analysis, or a 3-5 page statement of interest </DIV> <DIV>outlining their expertise and potential contributions to the </DIV> <DIV>symposium.</DIV><BR> <DIV>The symposium website <<A href="http://eksl-www.cs.umass.edu/aila/" EUDORA = AUTOURL>http://eksl-www.cs.umass.edu/aila/</A>> </DIV> <DIV>provides information about the symposium, background materials on </DIV> <DIV>link analysis, discussions about relevant AI technologies, links </DIV> <DIV>to existing software and datasets, and references to articles on </DIV> <DIV>link analysis theory and applications. In addition, website </DIV> <DIV>visitors can subscribe to an electronic mailing list for </DIV> <DIV>announcements about the symposium.</DIV><BR> <DIV>Organizing Committee: </DIV><BR> <DIV>David Jensen, co-chair -- CS Dept., Univ. of Massachusetts</DIV> <DIV>Henry Goldberg, co-chair -- NASD Regulation, Inc.</DIV> <DIV>William Mills -- Office of R&D, Central Intelligence Agency</DIV> <DIV>Malcolm Sparrow -- JFK School of Government, Harvard</DIV> <DIV>Katia Sycara -- The Robotics Institute, CMU</DIV> <DIV>Chris Westphal -- United Information Systems, Inc.</DIV> <DIV>Raphael Wong -- FinCEN, U.S. Treasury Department</DIV></html>
|
Copyright © 1998 Nautilus Systems, Inc. All Rights Reserved.MHonArc 2.2.0