About me

Hi, I am Suzana, a third-year student in the Ph.D. program in computer science at the University of Tsukuba, with graduation expected in March 2024.

Now, let’s go back in time for a second. In 2016, I graduated as a computer engineer at the Federal University of Amazonas. After that, I started to work as a software developer in Méliuz, an e-commerce company. My job tasks were related to search engines, crawlers, and optical character recognition. During that time, I learned a lot about machine learning and how we can use that to solve real-world problems.

In 2018, I was selected to study in Japan by the MEXT scholarship. I was a research student in my first year here, but I could extend the scholarship to the master’s and doctoral programs. Since then, I’ve been developing my research under the supervision of Prof. Kazuhiro Fukui at the Computer Vision Laboratory (CVLab) at the University of Tsukuba.

During my Ph.D, I have been researching about action recognition and the representation of temporal sequences. For this aim, I developed the slow feature subspace that can represent temporal structures while keeping the benefits of traditional subspaces, as their ability to learn from few data and dimensionality reduction. Our findings indicate that the slow feature subspace is highly effective to represent temporal sequences, often producing competitive results to deep learning methods.

To summarize, my research interests include human action recognition, multimodal learning, and change detection. Given my background, I am also interested in data mining, information retrieval, recommendation systems, and search engines.

Thank you for taking the time to read about me, and please do not hesitate to contact me if you need any information.