This year we have also invested in ProQuest's TDM Studio, a workspace in which researchers have been able to explore and analyse ProQuest data. The service consists of a cloud-based platform on Amazon Web Services with examples of Jupyter notebooks that can be used to process the collections of papers. By making services like this available to our community, we aim to ensure they have all the tools they need to adopt data-led methods.
One of our first projects using TDM Studio is led by Dr Galina Andreeva in the Business School, who is using this service to automate some parts of literature review writing. She is interested in the new form of credit/funding that has emerged recently – peer-to-peer (P2P) lending - and using the query search she has extracted the academic papers that contain this term (and related ones). ProQuest also gives the possibility to extract business reports and business news articles which she is analysing separately. The objective is to compare the narratives between academics and practitioners, in order to identify areas of (dis)agreement and potential gaps/research questions. ProQuest sample notebooks allow you to produce counts of papers by year, most frequent words and their contexts/ embeddings, counts on named entities, topic modelling. Together with Rayna Andreeva, a PhD student from Informatics, Galina is working on creating a workflow of automated process. They hope to use this work to investigate which parts/tasks can be automated, and which parts of a literature review will still require subjective/qualitative input.