Research and investment projects rely on data, which comes in different forms
Structured data is neatly arranged in rows and columns, like a spreadsheet.
Unstructured data is messy and disorganised. It is found in PDFs, Word documents, presentations, open-ended survey responses, transcripts of virtual meetings, e-mails, webpages, and elsewhere.
Extract the wisdom from unstructured data
Unstructured data such as text is harder to manage than structured data, but the effort is worthwhile. This data can provide rich insights and competitive advantage.
We use natural language processing (NLP) to unlock the hidden value in text data. Natural language means human language and NLP is a field of artificial intelligence that enables computers to analyse and understand human language.
NLP puts shape on unruly data, speeds up the research process, and ensures that every nugget of wisdom has been uncovered.
Summarise long documents
Identify key themes and patterns of repetition
Calculate sentiment (positive, neutral, or negative)
Spot trends in a time series of reports or surveys
Extract the most relevant quotations
Since 2019, Stephen Ryan has used NLP extensively within research and investment projects.
In 2021, leading academics in Trinity College Dublin sought his expertise to unlock the value in a database comprising over 300 million words. The collaboration was a resounding success.
Building on this, Stephen Ryan is now available for similar projects, helping data-owners to convert their unstructured text data into an asset.
Creating unstructured data
NLP can work in the other direction too, generating phrases from structured inputs.
Used correctly, NLP can turn rows and columns of numerical data into the outline of a narrative...and without the 'surprises' that come with ChatGPT.
Click here for a detailed example of how one type of narrative, the investment market update, could be created more efficiently using NLP.