All articles

Machine Learning with Linked Data

Epimorphics attracted me as they had proposed an interesting set of projects which the internship could cover, while also offering some opportunity to shape the direction of the project I was assigned. I was also looking to further improve my programming skills during the experience, something I believed the position would offer. 

As I had expressed my interest in the field of Machine Learning during my interview, I was assigned the task of using machine learning techniques to develop a database matching algorithm. The goal was to match items from a database containing food business addresses, which often varied in how they were formatted, to another database, which had more robust standardisation of its fields while giving unique identifiers for each address. 

The bulk of the project was built in Python and utilised the SciKit-Learn library for the machine learning elements and I was also able to find some opportunities to learn about the Kotlin programming language. Alongside producing the machine learning model, I developed a tool that allowed users to manually match up addresses from either database via an interface. This was a great learning experience as it allowed me to get better acquainted with the object-oriented side of Python.

You can read about one of the elements of the project I found most interesting over here.

Epimorphics are an incredibly approachable team, which leads to the creation of a wonderful environment in which to learn new skills. Working for them has been a valuable experience for me and I hope to stay in contact with them in the future. I will now be returning to my master’s degree feeling more confident in my ability to apply the material I am studying in the workplace and with a new set of skills under my belt.