Since we built Harmony, a common complaint has been that it frequently identifies the wrong questions in PDFs. The original algorithm for finding questions in PDFs was a mixture of rule based heuristics and some hand coded logic to look for e.g. lines in the document which begin with numbers. This was very fragile and worked fine on short questionnaires such as the GAD-7, but failed on larger documents. We decided to run a competition with our partner DOXA AI where members of the public could train their own model to extract questions from PDFs.

Harmony at MQ and Datamind Data Science workshop On 2 May 2025, Dr Eoin McElroy demonstrated Harmony at the MQ and Datamind Data Science workshop in Deutsche Bank. Eoin’s presentation focused on “Maximising the use of existing survey data: facilitating cross-study research using retrospective harmonization.” The workshop brought together researchers interested in applying novel harmonisation techniques to existing datasets. Eoin explained traditional harmonisation processes and presented a user-friendly guide to the Harmony tool, demonstrating how natural language processing can streamline the harmonisation process.