Developing and evaluating new and emerging artificial intelligence (AI) methods to extract useful information from unstructured financial text datasets. In particular, I am applying B.E.R.T to Earnings Call Transcripts to unearth sentiment within the calls and evalaute any relationships said sentiment has with financial market movements. To add an extra element to my thesis I have also been researching multimodal sentiment analysis techniques - particularly the combination of text + audio. If successful, I will be combining PRAAT and B.E.R.T to extensively analyse the characteristics produced in earnings calls.
Drawing earnings call data from FinnHub (https://finnhub.io/), I will be attempting to add to the literature the first classified dataset of its kind. If successful, this dataset could be used by a plethora of academics to continue bringing the area of research forward.
Master of Science, Financial Technology, Strathclyde Business School
1 Sept 2019 → 28 Aug 2020
Award Date: 28 Aug 2020
Bachelor of Mathematics and Statistics, Mathematics, Statistics and Finance, University Of Strathclyde
1 Sept 2015 → 20 Jun 2019
Award Date: 20 Jun 2019
Doctor of Science, Classification of Unstructured Financial Text in Earnings Call Transcripts, Strathclyde Business School
1 Apr 2021 → …
1 Jun 2019 → 1 Aug 2019
1 Jun 2018 → 1 Aug 2018