-
Notifications
You must be signed in to change notification settings - Fork 23
Home
Welcome to the Harmony wiki!
Harmony is an open source tool which is for psychologists to use to help them harmonise text data. That means questionnaires often in PDF format, where multiple PDFs could be used to ask similar questions.
- Psychologists and social scientists often have to match items in different questionnaires, such as "I often feel anxious" and "Feeling nervous, anxious or afraid".
- This is called harmonisation.
- Harmonisation is a time consuming and subjective process.
- Going through long PDFs of questionnaires and putting the questions into Excel is no fun.
- Enter Harmony, a tool that uses natural language processing and generative AI models to help researchers harmonise questionnaire items, even in different languages.
Here's a video demo of Harmony: https://www.youtube.com/watch?v=cEZppTBj1NI
Our source code is at https://github.com/harmonydata/harmony and our website is at https://harmonydata.ac.uk/
You can try the tool itself at https://harmonydata.ac.uk/app
You can see that, at present, the tool has the functionality that it can indicate the similarity between questionnaire items. However, the PDF data extraction is very tricky and can definitely be improved, and there are a number of open issues in the Github. Also any new ideas are welcome.
Here is a video about contributing to Harmony: https://www.youtube.com/watch?v=WAWRgCHSV5I
We have a Kaggle competition here: https://harmonydata.ac.uk/harmony-on-kaggle/
And here's a guide on contributing to Harmony: https://harmonydata.ac.uk/contributing-to-harmony/
Here's our ideas list: https://harmonydata.ac.uk/ideas
For transparency, we are not a monetised project and there is no paid work on the Harmony project at present.