Uncertain Data Processing and Applications

PhD Thesis Proposal Defence


Title: "Uncertain Data Processing and Applications"

by

Mr. Zhou ZHAO


Abstract:

Data uncertainty is inherent in many real-world applications like mobile 
tracking, due to physical device limitations and different kinds of 
environmental noise. Therefore, it is important to reduce the data uncertainty 
for these applications. In this proposal, we mainly introduce several uncertain 
data processing methods for RFID data.

We first study the problem of cleansing RFID data streams. Existing work mainly 
focuses on RIFD data cleansing in a static environment. We propose a 
probabilistic model for object tracking in mobile environment. We next study 
the problem of RFID localization and prsent a gradient descent method based on 
the decayed read rate. The possible applications are inventory checking and 
object checking.

We then study the problem of mining sequential patterns from mobile RFID data. 
We propose to measure pattern frequentness based on the possible world 
semantics. Inspired by the famous PrefixSpan algorithm, we develop a new 
algorithm called U-PrefixSpan for this problem.

Finally, we discuss some future work about other possible applications for
uncertain data processing.


Date:  			Friday, 28 March 2014

Time:                   1:30pm - 3:30pm

Venue:                  Room 5501
                         lifts 25/26

Committee Members:	Dr. Wilfred Ng (Supervisor)
 			Prof. Dik-Lun Lee (Chairperson)
 			Prof. Shing-Chi Cheung
 			Prof. Frederick Lochovsky


**** ALL are Welcome ****