Research Areas
Artificial intelligence, machine learning, data mining, disease outbreak surveillance

Research Description
Weng-Keen Wong's research interests in data mining lie primarily in the area of anomaly detection. While much of data mining is currently concerned with discovering patterns in the data, there is also a growing interest in finding anomalies. These anomalies play a significant role in scientific discovery and also in surveillance systems. Surveillance systems have traditionally played an important role in domains such as fraud detection and computer security. An emerging field for the application of surveillance algorithms is syndromic surveillance, which has the goal of detecting disease outbreaks as early as possible by monitoring pre-diagnosis health-care data. Present challenges for anomaly detection algorithms include detecting anomalies in spatial and spatio-temporal domains, finding meaningful anomalies, and dealing with massive data sets. Dr. Wong is also interested in Bayesian network structure learning, hierarchical Bayesian approaches and clustering.