IBM Research - Haifa

The Machine Learning Technologies group specializes in developing algorithms for automatic pattern recognition, prediction, classification, and learning of structures. We provide both core technologies and machine learning services. Our core technologies include massive scale, distributed and on-line machine learning, Bayesian networks, learning and classifying structures, anomaly detection, feature selection, and rule learning. We are currently active in domains including anomaly detection and prediction in time series data, with a focus on anomalies in machine logs; in the domain of financial services we have a special focus on fraud detection. We work in close collaboration with IBM teams around the globe, and externally in academic collaborations and industrial companies.

• Detecting Fraud in Financial Transactions (DRIFT)
• Problem Determination, System Management and Monitoring (MELODY)