Cross-Border ART Guidelines Version Control
I would like to pitch the concept of tracking down ART guideline mismatches across governments. The idea would be to learn to recognize authoritative ART guidelines and flag deviating ones in various types of publication. By doing so, outdated guidelines can be detected and updated. As an extension, this could be connected to drug resistance research papers. Policy makers could be notified early about advances in the state-of-the-art.
What would be needed in the team:
- Natural Language Processing (NLP) expertise (I can take care of that)
- Some subject matter expertise would be very welcome as well
- Scraping skills
Feel free to join or to come up with more/better ideas.
ricardo last edited by
Awesome Scrapping tools might help https://github.com/lorien/awesome-web-scraping/blob/master/python.md#web-scraping-frameworks
Thanks for the pointer. I am familiar with scrapy, which is my preferred option, because it allows you to define pipelines for scraped objects, while it also features loop detection and a decent queueing mechanism.
Irene del Carmen last edited by Irene del Carmen
@Joachim is it possible for you to hang out around Digityser wednesday-thursday or friday? Maybe you can tell us more about your idea match with some of the data camp bootcampers.
@Irene-del-Carmen : on Monday the 12th, I am in Brussels. Probably I can make some time in the afternoon. Would that be OK?
Irene del Carmen last edited by
@Joachim Hey Joachim, we have a lunch break on Monday, an ideal opportunity to hear more about your project. However, we have a workshop till 5pm. I already now there is a bootcamper interested in your project. It this is the only day you could pass by, yes, do it. We will be at the 5th floor (access through the stair near the entrance).
Hi, I cannot make it for lunch. But I will drop by this afternoon.
@Irene-del-Carmen Hi, I am in the kitchen at DigitYser right now.