2022 International Conference on Image, Signal Processing and Pattern Recognition(ISPP2022)
Assoc. Prof. Guoqing Chao



Assoc. Prof. Guoqing Chao
Harbin Institute of Technology, China

Multi-View Learning

Multi-View indicates multiple feature sets of the same object. In real world, multi-view data is ubiquitous. For example, a merchandise online can be described by the images, text and even speech, the image, text and speech descriptions construct three views of the merchandise. There are consensus and complementary information in these multi-view information. How to combine these multi-view information to improve learning performance is the key to multi-view learning. According to whether the label available, multi-view learning can be split into multi-view supervised learning, multi-view semi-supervised learning and multi-view unsupervised learning. In addition, it's quite related to multi-modal learning, multi-task learning, ensemble learning, and so on. We will introduce them comprehensively and point out some open problems.

Research Area:
Machine Learning, Data Mining, Pattern Recognition, Bioinformatics, Medical Informatics with a focus on multi-view supervised and semi-supervised learning, multi-view clustering.

Research Experience:
Now I'm an associate professor in the School of Computer Science and Technology of Harbin Institute of Technology (HIT)at Weihai. My main research interests include machine learning, data mining, pattern recognition, and so on. Before I joined HIT, I worked as a research scientist in Singapore Management University for one and half a year and worked as a research fellow in Northwestern University and University of Connecticut for two and half a year. My research focus on multi-view learning including multi-view classification, multi-view semi-supervised learning and multi-view clustering.