ICDM 2005 Workshop Proposal
Workshop on Knowledge Acquisition from Distributed, Autonomous, Semantically
Heterogeneous Data and Knowledge Sources
Description of the workshop topic and the associated research issues
Recent advances in high performance computing, high speed and high bandwidth communication, massive storage,
and software (e.g., web services) that can be remotely invoked on the Internet present unprecedented opportunities in
data-driven knowledge acquisition in a broad range of applications in virtually all areas of human endeavor including
collaborative cross-disciplinary discovery in e-science, bioinformatics, e-government, environmental informatics,
health informatics, security informatics, e-business, education, social informatics, among others. Given the explosive
growth in the number and diversity of potentially useful information sources in many domains, there is an urgent need
for sound approaches to integrative and collaborative analysis and interpretation of distributed, autonomous (and
hence, inevitably semantically heterogeneous) data sources.
Machine learning offers some of the most cost-effective approaches to automated or semi-automated knowledge
acquisition (discovery of features, correlations, and other complex relationships and hypotheses that describe
potentially interesting regularities from large data sets) in many data rich application domains. However, the
applicability of current approaches to machine learning in emerging data rich application domains presents several
challenges in practice:
(a) Centralized access to data (assumed by most machine learning algorithms) is infeasible because of the large
size and/or access restrictions imposed by the autonomous data sources. Hence, there is a need for
knowledge acquisition systems that can perform the necessary analysis of data at the locations where the
data and the computational resources are available and transmit the results of analysis (knowledge acquired
from the data) to the locations where they are needed.
(b) Ontological commitments associated with a data source (that is, assumptions concerning the objects that
exist in the world, the properties or attributes of the objects, the possible values of attributes, and their
intended meaning) are determined by the intended use of the data repository (at design time). In addition,
data sources that are created for use in one context often find use in other contexts or applications.
Therefore, semantic differences among autonomous data sources are simply unavoidable. Because users
often need to analyze data in different contexts from different perspectives, there is no single privileged
ontology that can serve all users, or for that matter, even a single user, in every context. Effective use of
multiple sources of data in a given context requires reconciliation of such semantic differences from the
user’s point of view.
(c) Explicitly associating ontologies with data repositories results in partially specified data, i.e., data that are
described in terms of attribute values at different levels of abstraction. For example, the program of a
student in a data source can be specified as Graduate, while the program of a different student in the same
data source (or even a different data source) can be specified as Doctoral.
Against this background, the proposed workshop seeks to bring together researchers in relevant areas of artificial
intelligence (machine learning, data mining, knowledge representation, ontologies), information systems
(information integration, databases, semantic web) distributed computing, and selected application areas (e.g.,
bioinformatics, security informatics, environmental informatics) to address several questions such as:
1) What are some of the research challenges presented by emerging data-rich application domains such as
bioinformatics, health informatics, security informatics, social informatics, environmental informatics?
2) How can we perform knowledge discovery from distributed data (assuming different types of data
fragmentation, e.g., horizontal or vertical data fragmentation; different hypothesis classes, e.g., naïve Bayes,
decision tree, support vector machine classifiers; different performance criteria, e.g., accuracy versus
complexity versus reliability of the model generated, etc.)?
3) How can we make semantically heterogeneous data sources self-describing (e.g., by explicitly associating
ontologies with data sources and mappings between them) in order to help collaborative science from
autonomous information sources?
4) How can we represent, manipulate, and reason with ontologies and mappings between ontologies?
5) How can we learn ontologies from data (e.g., attribute value taxonomies)?
6) How can we learn mappings between semantically heterogeneous data source schemas and between their
7) How can we perform knowledge discovery in the presence of ontologies (e.g., attribute value taxonomies)
and partially specified data (data that are described at different levels of abstraction within an ontology)?
8) How can we achieve online query relaxation when an initial query posed to the data sources fails (i.e., returns
no tuples)? That is, how do we perform a query-driven mining of the individual sources that will result in
knowledge that can be used for query relaxation?
Reasons why an ICDM workshop on this topic should take place
As noted above, the explosive growth in the number and diversity of potentially useful information sources in many
domains, there is an urgent need for sound approaches to integrative and collaborative analysis and interpretation of
distributed, autonomous (and hence, inevitably semantically heterogeneous) data sources. At present, while there are
several research conferences focus on well-established research areas (e.g., machine learning, data mining, knowledge
representation, databases), there is relatively little interaction among the different research communities. For example,
machine learning researchers working on algorithms for learning predictive models from distributed data, are isolated
from the large community of database researchers working on data integration, and the community of artificial
intelligence researchers focused on knowledge representation and inference. Researchers in this area can also benefit
from a better understanding of specific challenges posed by emerging informatics-enabled application domains such as
bioinformatics, health informatics, security informatics, environmental informatics.
Fundamental advances in collaborative approaches to knowledge acquisition and data-driven decision making from
distributed, autonomous, semantically heterogeneous data and knowledge sources require synergistic synthesis of
research advances, insights, algorithms, and results in multiple areas of:
• artificial intelligence – especially machine learning, data mining, knowledge representation and inference,
intelligent agents and multi-agent systems;
• information systems – especially databases, information integration, semantic web;
• distributed computing (e.g., service-oriented computing).
The proposed workshop aims to bring them together in order to enable discussion of research problems, approaches,
insights, and results drawn from multiple, and at present, largely disparate areas of artificial intelligence, computer
science, and emerging informatics-enabled disciples. At present, there is no annual conference or workshop dedicated
to this topic. It is hoped that the resulting exchanges will stimulate further interaction between these communities and
result in the development of new approaches that would advance the current state of the art in collaborative systems
for collaborative analysis, interpretation, and decision making from distributed, autonomous, semantically
heterogeneous data and knowledge sources.
The workshop will consist of:
• An opening session for introducing the workshop topics, goals, participants, and expected outcomes
• A small number of invited talks carefully intermixed with presentation of contributed papers. The invited
talks will give overviews of the key topics (learning from distributed data, semantic Web, ontology-based
information integration, distributed description logics, selected applications, etc.). A possible list of invited
• Alex Borgida (ontologies and databases) - Rutgers University
• Katy Borner (information visualization) -- Indiana University
• Foster Provost (machine learning and data mining) – New York University
• James Hendler (semantic web) - University of Maryland at College Park
• Alon Halevy (information integration) – University of Washington
• Dieter Fensel (ontologies) - University of Innsbruck
• Tom Dietterich (machine learning and environmental informatics) – Oregon State University
• H. Jagadish (biological data management) – University of Michigan
• Daphne Koller (probabilistic models) – Stanford University
• Munindar Singh (service-oriented computing) – North Carolina State University
• Michael Pazzani (intelligent information systems) – National Science Foundation
• Presentations of contributed papers that represent completed work.
• Breaks between sessions, meant to encourage informal discussions related to the topics discussed in the
sessions and to create opportunities for collaborations.
• A panel discussion on challenges and future research directions
• A wrap-up session summarizing the workshop (including formal or informal discussions).
Description of the anticipated target group(s) of attendees
The workshop is of interest to researchers, students, and practitioners in a number of areas of artificial intelligence,
information systems, and related areas including: machine learning and data mining, information extraction,
information integration, knowledge representation, semantic web, software agents and multi-agent systems, and
service-oriented computing. The workshop is also of interest to researchers and practitioners in emerging informatics-
enabled application domains such as bioinformatics, environmental informatics, health informatics, security
informatics, e-business, social informatics.
The organizers will make an effort to ensure a good mix of established researchers as well as graduate students and
junior researchers on the one hand and academic and industrial participants on the other.
Potential authors and attendees
A number of researchers who were informally contacted have expressed an interest in the proposed workshop. We put
together a short list of potential participants. (The list below does not include members of the program committee or
participants named on the list of potential invited speakers). We expect the target size of the workshop to be around 40
participants to allow for fruitful interactions and discussion in an informal setting among the workshop participants.
AnHai Doan -- University of Illinois at Urbana-Champaign
Lise Getoor – University of Maryland
Barbara Eckman -- IBM Life Sciences Solution Development
George Forman – Hewlett Packard Labs
Simon Kasif – Boston University
Zoe Lacroix -- Arizona State University
Pat Langley – Stanford University
Bertram Ludaescher -- University of California, Davis and San Diego Supercomputer Center
Sanjay Madria -- University of Missouri-Rolla
Nina Mishra – Stanford University and IBM
Vibhu Mittal – Google
Joyce Mitchell – University of Utah
Katia Sycara – Carnegie Mellon University
Lee Giles – Pennsylvania State University
Peter Tarczy-Hornoch – University of Washington
Workshop Organizing Committee – Contact Information
Dr. Doina Caragea (Contact Person)
226 Atanasoff Hall
Department of Computer Science
Iowa State University
Ames, IA 50011-1040 USA
Professor Vasant Honavar
226 Atanasoff Hall
Department of Computer Science
Iowa State University
Ames, IA 50011-1040 USA
Dr. Ion Muslea
Language Weaver, Inc.
4640 Admiralty Way
Marina del Rey, CA 90292
Professor Raghu Ramakrishnan
Department of Computer Sciences
University of Wisconsin-Madison
1210 West Dayton Street
Madison, WI 53706-1685 USA
Preliminary Program Committee
Naoki Abe - IBM
Liviu Badea – ICI, Romania
Marie desJardins - University of Maryland, Baltimore County
Tim Finin -- University of Maryland, Baltimore County
Joydeep Ghosh -- University of Texas
Hillol Kargupta – University of Maryland, Baltimore County
Sally McClean -- University of Ulster at Coleraine
Dragos Margineantu – Boeing
Bamshad Mobasher – DePaul University
Jay Modi – Carnegie Melon University
C. David Page Jr. – University of Wisconsin, Madison
Alexandrin Popescul - Ask Jeeves, Inc.
Adrian Silvescu – Iowa State University
Steffen Staab -- University of Koblenz
Previously Organized Related Workshops
• IJCAI-2001 Workshop on “Knowledge Discovery from Heterogeneous, Distributed, Autonomous, Dynamic
Data and Knowledge Sources”.Vasant Honavar, Chair.
• AAAI-2004 workshop on "Adaptive Text Extraction and Mining". Ion Muslea, Chair.
• IJCAI-2001 workshop on "Adaptive Text Extraction and Mining". Ion Muslea, Co-chair.
• AAAI-99 workshop on "Machine Learning for Information Extraction". Ion Muslea, Co-Chair.
CALL FOR PAPERS
ICDM 2005 WORKSHOP ON KNOWLEDGE ACQUISITION FROM DISTRIBUTED, AUTONOMOUS, SEMANTICALLY
HETEROGENEOUS DATA AND KNOWLEDGE SOURCES
NOVEMBER 27 , NEW ORLEANS, LOUISIANA, USA
Steffen Staab – Univ. of Koblenz
Aug. 12th: Paper Due
Sept. 4th: Notification Workshop Goals
Sept. 26th: Camera Ready
The workshop aims to bring together researchers in relevant areas of
Nov. 27th: Workshop artificial intelligence (machine learning, data mining, knowledge
representation, ontologies), information systems (information
Organizing Committee integration, databases, semantic web) distributed computing, and
selected application areas (e.g., bioinformatics, security informatics,
Doina Caragea environmental informatics) to address several questions that arise in the
Iowa State University process of knowledge acquisition from distributed, autonomous,
email@example.com semantically heterogeneous data and knowledge sources.
Vasant Honavar Topics of Interest
Iowa State University
firstname.lastname@example.org Topics of interest include, but are not restricted to:
• Challenges presented by emerging data-rich application domains
Ion Muslea such as bioinformatics, health informatics, security informatics,
Language Weaver, Inc. social informatics, environmental informatics.
email@example.com • Knowledge discovery from distributed data (assuming different
types of data fragmentation, e.g., horizontal or vertical data
Raghu Ramakrishnan fragmentation; different hypothesis classes, e.g., naïve Bayes,
University of Wisconsin-Madison decision tree; different performance criteria, e.g., accuracy versus
firstname.lastname@example.org complexity versus reliability of the model generated, etc.).
• Making semantically heterogeneous data sources self-describing
Program Committee (e.g., by explicitly associating ontologies with data sources and
mappings between them) in order to help collaborative science .
Naoki Abe, IBM • Representation, manipulation, and reasoning with ontologies and
Liviu Badea, ICI, Romania mappings between ontologies.
Doina Caragea, Iowa State Univ. • Learning ontologies from data (e.g., attribute value taxonomies).
AnHai Doan, UIUC • Learning mappings between semantically heterogeneous data source
Marie desJardins, UMBC schemas and between their associated ontologies.
Joydeep Ghosh, Univ. of Texas
• Knowledge discovery in the presence of ontologies (e.g., attribute
C. Lee Giles, Penn State Univ.
value taxonomies) and partially specified data (data described at
Vasant Honavar, Iowa State Univ.
different levels of abstraction within an ontology)?
Hillol Kargupta, UMBC
• Online query relaxation when an initial query posed to the data
Sally McClean, U. of Ulster, UK
sources fails (i.e., returns no tuples), or equivalently, query-driven
Bamshad Mobasher – DePaul U.
mining of the individual sources that will result in knowledge that
Jay Modi – Carnegie Mellon Univ.
can be used for query relaxation.
C. David Page, Univ. of Wisconsin
Alexandrin Popescul - Ask Jeeves Submission Instructions
Raghu Ramakrishnan, Univ. of
Postscript or PDF versions of papers, no more than 10 pages long
Zbigniew Ras, UNC-Charlotte
(including figures, tables, and references) in the ICDM camera-ready
format (IEEE 2-column format), should be submitted electronically to email@example.com by August
12th. Each paper will be rigorously refereed by at least 2 reviewers for technical soundness, originality, and clarity of
presentation. Accepted papers will be included in informal workshop proceedings published by ICDM and distributed
at the workshop. More details about the workshop can be found at www.cs.iastate.edu/~dcaragea/ICDM-KA.