Dissertation Diary

It is May 14th when I first set up this dissertation diary. On this day, it has been ten days into my MSc dissertation research proper. So I have had a fair amount of material - background readings and data - that I have collected for my dissertation. It is just that I must acknowledge that I RELY ON LARGE LANGUAGE MODELS for this research and I WILL RELY ON THEM THROUGHOUT THE ENTIRE PROCESS. So this will allow me to disclose how I have been using it in a more comprehensive manner that I will never be able to do in a formal dissertation report, and explain how AI is augments my work and not doing the whole dissertation project for me.

You can click on the "Context" tab below to find out more about this dissertation in particular, but it was on May 12th where I asked Claude to guide me in building a Python script to scrape real-time bus location data from the Bus Open Data Service. Claude returned me this curious link. It turned out that there's a non-profit organisation, Open Innovations, which has already been scraping real-time bus location data from BODS since June 18th, 2025 - so that means I don't have to scrape for data anymore! All I had to do was just download data from whichever day(s) that I needed and convert them into retrospective bus timetable data!

On their main page, they have blog posts, through which they share about their other open data projects but also about why they bothered to archive real-time bus locations. It was also through the blog that I learnt that Open Innovations had only recently closed down, and that their real-time bus location archive would soon terminate in about two months when their server finally runs out of memory. This bus location data archive exists just in time for my dissertation. Thus, I thought it would be fitting to record my dissertation journey in a frank, public-facing manner as a tribute to Open Innovations because I feel indebted to their hard work, upon which I am building.

I intend to structure my ramblings recording entries into sections - "Context", "Literature", "Data and Methods", "Results" - so that it can also double up as a resource that I can fall back on when I have to start writing the formal report in July (submission is due on August 20th). Killing two birds with one stone, or maybe taking two journeys on one fare 🤣?

I hope to be as clear as possible here, but if you have any questions, feel free to message me on LinkedIn!