Pejman Khadivi, Ph.D.
Ph.D., Computer Science, Virginia Tech
Building/Room: SINE 290-09
- Design & Analysis of Algorithm
- Comp Systems Principles
- Artificial Intelligence
My primary research interests are in the field of artificial intelligence, machine learning, and data analytics using large scale datasets, with emphasis on time series analytics, which is considered as a critical component in various domains including forecasting, cyber-physical systems, anomaly detection, and reliable system design.
In cyber-physical systems, beside forecasting and prediction tasks, machine learning techniques can be used in other applications such as anomaly detection, smart controller design, and surveillance systems. Furthermore, one of the important issues with machine learning and time series analytics is the existence of noise in datasets. Machine learning algorithms are sensitive to noise and hence, denosing is a crucial step to perform before using any machine learning algorithm. Current research problems that I am working on them are as follows:
- Real time techniques for time series denoising using information theory and deep learning
- Anomaly detection in cyber-physical systems in order to perform fault detection and fault location with the aim of reliability improvement
- Using open source datasets for forecasting applications in health science and tourism industry
- Wikipedia usage behavior modeling