Student Profile: Tobin South
Tobin is a second-year student at the University of Adelaide studying a Bachelor of Mathematical and Computer Sciences. While at University, he has developed a passion for Data Science, Big Data Problems, Networks, and Machine Learning. Tobin enjoys spending his time working on projects, usually building robots, with his mates and running the Mathematics student society at University of Adelaide.
Collaborative Network Based Categorisation Models
As a species, humans love to group things together. We group songs into genres, folklores into categories and mathematical topics into disciplines. However, songs often break genre conventions, folklores diverge from myth conventions, and mathematical challenges draw from many different disciplines.
By using data sets to create a network of connections, the behaviour of these networks can be examined. This project proposes the analysis of these networks, and it explores if the current classifications are reflective of the actual clustering of the networks that are created.
To achieve this, a variety of available data sets will be examined to construct networks and develop new data-driven categorisation schemes based on observed network properties, rather than arbitrarily assigned labels. An example of this process would be to collect data from Spotify using an Application Programming Interface (API) to build a collaboration network between artists, which will enable the graph properties of this previously-unexplored network to be measured (e.g. degree distribution, clustering, span, etc), and then use the community structure of the network to develop new genre classifications for musicians.
This project will explore the relationships between subjective classifications and the community structure of the underlying networks. The research will attempt to answer the questions, “How diverse are the people that contribute to ‘subreddits’ on Reddit?”, or “Do genres mean anything in the modern age of musical collaboration?”, and more.