Learning Subspaces from Proximities

Add to your list(s) Download to your calendar using vCal

• Dr. Tingting Mu, Department of Electrical Engineering and Electronics, University of Liverpool
• Monday 17 November 2014, 16:00-17:00
• Learning Centre, UG05.

If you have a question about this talk, please contact Lars Kunze.

Host: Prof. Xin Yao

The modelling of information as data points in multi-dimensional spaces is the fundamental stage in data analysis, reasoning, prediction and detection tasks. When facing raw information, distorted by noise, obscured by redundancy and often presented through varying formats, a compact and refined low-dimensional subspace can be learned, where the hidden information content of interest is extracted, preserved and ideally enhanced. In this talk, I will present our recent research on learning low-dimensional subspaces, where the core idea is to extract latent data characteristics by formulating and processing pattern proximities. Application examples on sentiment analysis, descriptive document clustering and biometric identity recognition will be demonstrated.

Speaker’s homepage: http://pcwww.liv.ac.uk/~tmu/Site/About_Me.html

This talk is part of the Artificial Intelligence and Natural Computation seminars series.

This talk is included in these lists:

• Artificial Intelligence and Natural Computation seminars
• Computer Science Departmental Series
• Computer Science Distinguished Seminars
• Learning Centre, UG05
• computer sience

Note that ex-directory lists are not shown.