Keywords: NLP, STM, TextData
Need: The government’s approach to the pandemic relies on a test, trace and isolate strategy, mainly implemented via the digital Contact Tracing and Advice Service (CTAS). Feedback on user experience is central to the successful development of public-facing services. The aim of this project is to evaluate the use of machine learning techniques as tools to understand the issues with the service as expressed in the free-text responses of the users, in order to rapidly generate actionable recommendations that can be used to increase user satisfaction of the CTAS.
See outputs from project - Structural Topic Modelling for NHS survey data
Current Knowledge/Examples & Possible Techniques/Approaches: The CTAS team have explored the free-text data and have conducted some preliminary analysis using Structural Topic Modelling (STM). Based on this work, the team generated insights around the most prominent issues with the service and want to continue to develop the approach further.
Based on this preliminary work, the team hope to extend the capabilities of the system by:
Related Previous Internship Projects: n/a
Enables Future Work: Establishing a methodology for evaluation of services and extending public health emergency response capability. Exploring the synergy between behavioural and data science to develop innovative tools for analysis of qualitative data.
Outcome/Learning Objectives: A practical system that enables us to rapidly analyse and present topics/summaries from live user feedback.
Datasets: CTAS feedback free-text responses, possibly alternative public health datasets.
Desired skill set: When applying please highlight any experience around natural language processing, machine learning, user experience data, responsive interfaces for data presentation, coding experience, or any other data science experience you feel relevant.
Return to list of all available projects.