ACM KDD CUP Data Mining Contest 2004
Joint Chinese Academy of Science
has won the Championship of the 2004 Bioinformatics
Prof. Yang Qiang, Associate Professor of the Department of Computer Science at HKUST, with joint effort of the researchers from the Chinese Academy of Sciences and Insistute of Computing, has won the championship of the ACM KDD CUP Data Mining Conest 2004.
The ACM KDD CUP is the most rigorous annual competition in the field of predictive technology and data mining. The competition is opened to all industries and hundreds of individuals from around the world participated. ACM KDD CUP Contest 2004 was held between May and July 2004 by the ACM Special Interest Group on Knowledge Discovery and Data Mining. The contest consisted of 59 teams worldwide solving a difficult and practical problem in data mining from biological databases, for knowledge discovery that is of interest to biologists. HKUST's Professor Qiang Yang is the coach of the team.
The Association for Computing Machinery's Knowledge Discovery and Data Mining (ACM SIGKDD) officially presented the team with the award on 24 Aug 2004 at the ACM KDD 2004 Conference in Seattle, Washington, USA. This success marked a major milestone for the newly established HKUST-CAS joint lab, which aims to promote joint research between researchers from Hong Kong UST and Chinese Academy of Sciences on topics ranging from data mining and multimedia to sensor networks.
Data mining is a multi-disciplinary research and application area that aims to discover novel and useful knowledge from vast databases, using methods ranging from artificial intelligence, statistics and databases. This year's KDD CUP focused on finding patterns from a large biological dataset. The competition consists of two phases. The first phase lasted for over a month in which a training dataset is provided for constructing a predictive model. In the second phase the test data are provided and the predicted results are submitted. The final results are judged by an independent panel on a number of success criteria.
This year's competition focuses on data-mining for a variety of performance criteria such as Accuracy, Squared Error, Cross Entropy, and ROC Area. The joint team won number one in two of the criteria and tied on overall scores of all criteria, with two other teams from New Zealand and USA, respectively.
Result of the event is available on ACM KDD CUP 2004 - Results.