reach
reach copied to clipboard
wellcometrust
Wellcome tool to parse references scraped from policy documents using machine learning
We currently report that there can be multiple citations of the same reference in one policy document. This can happen when there are multiple references sections in a policy document....
Text is found to match short reference titles in the main text of the policy document, and thus included as a match in the exact text match. Increase the title...
I know of 3 types of badly identified references sections problems: 1. Text in between references is included 2. Text after a references sections is included 3. The references section...
If there are multiple matches with the same similarity the code picks the first one as the match. Should we pick this randomly or is it fine to always pick...
Currently `fuzzymatch` gives the first of the highest matches for each reference. So if there are two matches with the same cosine similarity, e.g. (made up example): ``` Reference id...
Issue #41 occurred when a Title field was never created. We fixed this so that it'll always be created, but we should perform a test to check that this is...
We find that policy document year minus reference year is sometimes negative, this shouldn't be the case. Investigate the circumstances in which this happens to understand whether there is a...
As a **user** I need to ** filter and narrow results** So that I can **get a smaller, more relevant set of data** This issue will need further ideation and...
Users wanted better 'linking' of results to pubs etc and there is some evidence to suggest that they want a richer set of data, eg DOI, This needs further research...
### Results rows to include checkbox Add checkbox to each row to allow users to include or exclude results from a .csv download ### Results rows to include 'actions' icon...
Metadata
Owner
Metadata
Wellcome tool to parse references scraped from policy documents using machine learning