About
I have recently completed my PhD degree at the School of Computer Science, University College Dublin (UCD), under the guidance of Dr. David Lillis. My research focused on addressing the categorization of crisis messages in low-data environments. Currently, I am working as a research assistant, concentrating on hierarchical multi-label classification for lengthy documents. During my free time, I enjoy coding challenging projects that are difficult to find online, such as open-sourced ones. Furthermore, I have a keen interest in cooking, tutoring or casual writing. For more information about my background and expertise, please refer to my CV.
Recent News
- 7/02/2023: I passed my PhD Viva and my thesis is titled Coping with low data availability for social media crisis message categorisation (final version will be available soon).
- 24/03/2020: I passed the stage transfer examination. Here is the Transfer report.
Misc.
QAs
For anyone interested, below are recommended tools for mantaining a blog site. These are just so far personally thinking good choices. Let me know if you have better alternatives.
- What framework do you use for this blog site?
I use jekyll. A quick start guide for building a Jekyll blog can be found here jekyll-now. Jekyll also provides easy ways for tracking the data flow of the site via Google Analytics and highlighting programming codes (with one or several lines of codes). If I feel I need to use more front-end GUI components in my blogs, I can easily intergrate the bootstrap package with the site as well.
- What drawing tool you use for blogging?
I draw via draw.io and export graphs to svg files so that I can render the graphs as tensor images in my blogs. I also consider Google Slides some times.
- What tool do you use for equation annotation?
I use both Mathpix and MathJax. The former helps me convert equations in screenshots to editable latex format. The latter helps display the equations correctly in all browsers.
- What do you do if want to cite graphs or tables from papers in the literature?
I usually first crop the table/graph portions from the papers with some pdf tools such as pdfresizer. With the cropped pdf portions, I then convert them to svgs files after adding references so that I can use them as tensor images in my blogs. There are many online pdf2svg convertors such as ZAMZAR. I know this sounds a bit troublesome, so I am looking for better ways for citing tensor images from pdf files.