Martin Wood is a Data Scientist at the Data Science Campus. He has a Geology degree and too many sci-fi books.
His particular interests are data/text mining/web scraping, data linkage and statistical and machine learning methods for text analytics and summarisation, because it’s tangentially related to books.
We have been exploring how NLP techniques and LLMs could be used in the future to improve website search experience for end-users.
Read more on Using large language models (LLMs) to improve website search experience with StatsChat
We are planning to follow up on previous work we published earlier this year looking at public transport availability across the UK, by providing metrics for urban centres across the UK, as well as their international counterparts.
Read more on Comparing international transport performance in urban centres: upcoming work
We have published data to help researchers and local planners understand how public transport access varies across the UK, using open data and open-source software.
Read more on Using open data to understand hyperlocal differences in UK public transport availability
The Department for Business, Energy and Industrial Strategy (BEIS) asked the Campus to see if machine learning techniques could be used to predict the energy efficiency of properties that do not have an official score. Read more about our findings on this blog post.
Read more on Predicting the Energy Performance Certificates (EPC) of properties
Rail services can be affected by temporary factors such as unseasonably hot weather, industrial action and engineering works. We produced maps from open data showing service levels at every station in Great Britain over a 21-day period.
Read more on Visualising rail schedules using open data
This report summarises our work to meet widespread demand for better automated and higher quality solutions to this problem, by using machine learning (ML) methods to improve the match rate and accuracy of automatic classification of SIC and SOC classifications.
Read more on Automated coding of Standard Industrial and Occupational Classifications (SIC/SOC)