ContentMine is thinking about proposing a session at Mozfest 2015.
The format is an interactive hacking session with ContentMine tools, the content is going to be at the intersection of science and political action, in order to be attractive to a wide range of participants.
The current idea is to review drug papers regarding company contributions. Our technology is ready to extract relevant entities from specific parts of texts. Building on that we are able to create relations between entities, e.g. authors, drugs, and companies.
A session is only going to last about two hours, so we need to focus on working with results. We would prepare the data and tools, so that participants can focus on asking questions to the data. A dataset could include metadata of a publication (e.g. authors, institutions, publication dates,...) and data we extract from the content (e.g. mentions of drugs, testing procedures, company names,...). Participants would learn how to visualize data in e.g. a network graph or how to create a timeseries.
We welcome your input to the session, especially regarding the question of long term outcomes and learning effects.
- How does this activity benefit you and your community?
- How would you take this activity forward after the hack?
- What questions would you ask the data?
- What are the relevant data sources?