First International Workshop on
Mining Graphs, Trees and Sequences (MGTS-2003)
Call For Participation (CFP)
In conjunction with ECML/PKDD-2003
14th European Conference on Machine Learning (ECML'03) and
7th European Conference on Principles and Practice of Knowledge Discovery in Databases (PKDD'03)
Cavtat-Dubrovnik, Croatia, September 22 and 23, 2003
Single File Version of Online Workshop Proceedings (Zipped PDF)
Below are attendance information, important dates and organization details.
Description of the workshop topic and goals
General scope
Ever since the early days of machine learning and data mining, it has been realized that the traditional attribute-value and item-set representations are too limited for many practically applications in domains such as chemistry, biology, network analysis and text mining. This has triggered a lot of research on mining and learning within alternative and more expressive representation formalisms such as computational logic, relational algebra, graphs, trees and sequences. The state-of-the-art is that attribute-value and item-set representations lie at one extreme end of the spectrum, and multi-relational data mining and inductive logic programming at the other end. The middle is occupied by traditional data structures employed throughout the field of computer science. These include graphs, trees and sequences (or strings). The motivation for using such representations is that they are 1) more expressive (and therefore more widely applicable) than flat representations, and 2) potentially more efficient than multi-relational learning and mining techniques. At the same time, the data structures of graphs, trees and sequences are among the best understood and most widely applied representations within computer science. Thus these representations offer ideal opportunities for developing interesting contributions in data mining and machine learning that are both theoretically well-founded and widely applicable.
Motivation
Whereas there have been a large number of workshops devoted to multi-relational data mining and inductive logic programming as well as applications of intermediate representations in e.g. ontologies, bioinformatics, XML-data, text-mining, there has ? to the best of our knowledge ? not been any workshops specifically devoted to foundational issues in intermediate representations. It is precisely the goal of this workshop to bring together researchers interested in mining and learning within graphs, trees and sequences. We believe it is the right time to organize such a workshop because of the increasing interest in the role of such data structures, which is in turn motivated by the many interesting application domains. Indeed, the total number of papers related to graph and tree mining in SIGMOD, SIGKDD, IJCAI/AAAI, ICML, ECML/PKDD and IEEE ICDM was 10 in 2001 and their number inceased to 18 in 2002. At the same time, we believe ECML/PKDD 2003 is the right forum for organizing this workshop due to the traditional European interests in rich representations for mining and learning. In addition, various Japanese and American groups have contributed important results to these areas. It is therefore to be expected that this workshop will not only attract European researchers to attend ECML/PKDD but also Asian and American ones. Another reason for choosing ECML/PKDD as the workshop location is the close proximity in time and space to the ILP 2003 conference. The focus of the present workshop is closely related to that of inductive logic programming but yet sufficiently different to hope that researchers will attend both events.
We are looking for contributions related to graph, tree and sequence structure mining and learning. More specifically, the workshop will focus on the following topics:
- Efficiency issues in graph, tree and sequence mining and learning
- Identifying interesting subclasses that can efficiently be mined or learned
- Basic princinciples of graph, tree and sequence mining
- Analysis of the complexity of graph, tree and sequence mining
- Applications to real world problems in e.g., biology, chemistry, XML, etc.
- Relationship of graph, tree and sequence mining to other techniques
- Any other result relevant to graph, tree and sequence mining.
- Research works presenting theoretical results, basic research, perspective solutions and practical developments are welcome, provided that they address the topic of the workshop.
Two types of basic submissions will be considered:
- Technical contributions (an extended abstract with up to 10 pages in the Lecture Notes in Computer Science (LNCS) format)
- Position papers (up to 4-page abstracts in the Lecture Notes in Computer Science (LNCS) format)
- Technical papers should be original and not previously published elsewhere. An exception can be made for high quality papers that have not yet been presented in Europe. Multiple submissions are acceptable(as long as this clearly indicated).
- Position papers will be presented in a special "Discussion" section.
- Electronic submission in PDF format is preferred, and should be sent by June 20, 2003 (DEADLINE has been EXTENDED!!) to papersubmit@ar.sanken.osaka-u.ac.jp and subo@informatik.uni-freiburg.de, Subject: MGTS-2003 workshop submission paper. Please note that hard copy and fax submissions will not be accepted.
- Submitted papers will be reviewed by referees from the Program Committee. The authors will be notified about the acceptance or rejection of their papers by July 4, 2003. Camera-ready versions of the papers are due July 11, 2003.
Final Version
Accepted papers will be published in the working notes provided by ECML/PKDD-2003 and available on our website. Depending on the interest in the theme and the number and quality of submissions, further publications (such as a book, or a special issue of a journal) may be considered.
Attendance is not limited to the paper authors. We strongly encourage interested researchers from related areas to attend the workshop. The intended audience for the workshop involves researchers and practitioners from all over the world interested in graph, tree and sequence mining, and we hopes to attract interest from a wide range of possible fields, including: inductive logic programing, multi-relational data mining, tree mining, symbolic sequence mining, semi-structured data mining, chemistry, biology, computer networking, and etc.
Important dates
Paper Submission Deadline June 20, 2003 (DEADLINE has been EXTENDED!!) Notification of Acceptance July 4, 2003 Camera-Ready Copy Due July 15, 2003 (DEADLINE has been EXTENDED!!) Proceedings (camera/web-ready) July 18, 2003 Workshop September 23, 2003
Program Committee Chairs
Luc De Raedt, Professor, Univ. of Freiburg, Germany deraedt@informatik.uni-freiburg.de Takashi Washio, Assoc. Prof., ISIR, Osaka Univ., Japan washio@ar.sanken.osaka-u.ac.jp
Program Committee
Hiroki Arimura Kyushu University, Japan Diane J. Cook University of Texas at Arlington, United States of America Luc Dehaspe Katholieke Universiteit Leuven, Belgium Saso Dzeroski Jozef Stefan Institute, Slovenia Peter Flach University of Bristol, United Kingdom Thomas Gaertner GMD - German National Research Center for Information Technology, Germany Jiawei Han Univ. of Illinois at Urbana-Champaign, United States of America Colin de la Higuera University Jean Monnet at Saint-Etienne, France Lawrence B. Holder University of Texas at Arlington, United States of America Tamas Horvath GMD - German National Research Center for Information Technology , Germany Akihiro Inokuchi Tokyo Research Laboratory, IBM, Japan Stefan Kramer Albert-Ludwigs-Universitat Freiburg, Germany Michihiro Kuramochi University of Minnesota, United States of America Satoru Miyano University of Tokyo, Japan Hiroshi Motoda Osaka University, Japan Richard Nock Universite Antilles-Guyane, France Jan Ramon Katholieke Universiteit Leuven , Belgium Michele Sebag Universite Paris-Sud, France Hannu TT Toivonen University of Helsinki, Finland Brijnesh-Johannes Jain Technische Universitat, Berlin Germany Mohammed J. Zaki Rensselaer Polytechnic Institute, United States of America