Hao Li (Ph.D. student from CS) and I conducted a big-data analysis project using the MapReduce framework (Hadoop) for the final project of INFM718G (Data-Intensive Computing with MapReduce, by Dr. Jimmy Lin). Targeting all the news images in April 2013, we tried to rank news images based on the imporance and popularity level of each news image. In order to do that, we extracted image features using SIFT (Scale-invariant feature transform) and constructred a graph of images using LSH (Locality-sensitive Hashing) as a means to approximate the simliarity of images.
This prototype was an intermediate result of a robot-human interaction project conducted at Torooc Inc., a start-up company where I was a co-founder and Director of Software Development, in 2011. I first designed microphone amplifier circuits using transistors and amplifier chips, respectively. Then, I connected 3 microphones to Texas Instruments' Stellais LM4F embedded board. On top of that, I implemented a Time-Difference of Arrival (TDOA) algorithm in C language.
This research project had been conducted for my master's thesis (Master of Information Management degree). I conceptualized cultual background with Hall's high- and low-context culture (1976) and tried to see whether people's perceptions of urban places vary between physical addresses and symbolic representations of spaces (landmarks), when their cultural backgrounds were different. A survey questionnaire was used to measure cultural background, and a web-based online game was used to measure people's perceptions of places.