Skip to content

OVERVIEW.md

Latest commit

 

History

History
62 lines (43 loc) · 3.23 KB

OVERVIEW.md

File metadata and controls

62 lines (43 loc) · 3.23 KB

OVERVIEW of ContentMining Workshop

A 1-day hands-on workshop to explore how new knowledge in Plant Sciences can be rapidly extracted from the scientific literature.

Overview

The literature contains millions of articles and reports with detailed knowledge about plants. You will

  • search EuropePMC which contains Open Access articles.
  • download them automatically (getpapers from ContentMine).
  • transform them into semantic form (AMI from ContentMine)
  • search them with multiple dictionaries (created from Wikipedia)
  • analyze the results (AMI)

Style

All activities are hands-on. Some is online, but the much of the work is on your own machine. We have created a copy (memory stick) of the software, documentation, dictionaries and corpora so you can do most of the tasks offline. Everything is also is on http://github.com/petermr/tigr2ess.

The morning will have formal presentations with delegates "click-along" ("karaoke-style"), with pauses for questions and feedback. We'll use the example of "Holy Basil" (Ocimum tenuiflorum) as it links to many fields (cooking, medicine, religion, and plant science).

In the afternoon most delegates will form small free-form groups:

  • multidisciplinary ("what can I learn about my plant?" - pests, climate, stress, invasive)
  • hackathon style - small groups collaborating to create knowledge
  • informal communication (Etherpad);
  • dictionary-based

Support staff

Team:

  • Ambarish Kumar (NIPGR)
  • Gitanjali Yadav (NIPGR-Cambridge)
  • Peter Murray-Rust (Cambridge-ContentMine)
  • Vinita Lamba (NIPGR)

Ambarish, Amit Yadav and Vinita have all worked very hard to make this workshop work.

Thanks

Rik Smith-Unna (ex Plant Sciences Cambridge) wrote getpapers.

Program

morning

These are the modules with owners

  • installation and housekeeping. A brief review of any technical issues and re-programming. (Peter, Amit, Ambarish)
  • Searching EuropePMC (online) (Vinita)
  • download papers (online). Might be staggered due to bandwidth at either end. (Ambarish)
  • Wikipedia, Wikidata, WikiFactMine (online) PeterMR will present these resources online, with complete instructions for self-paced work later. (Peter, Vinita)
  • creating dictionaries (online). Wikipedia pages will be used to generate dictionaries for searching. (Ambarish)
  • searching with dictionaries (local). A wide variety of Wiki-enhanced dictionaries will be used to search local corpora. (Amit)

afternoon

Using the model of Ocimum choose your own plant (e.g. Millet, Rice, Wheat) to do a free-form project and report back. Partial resources for all these are supplied.

(Optionally some delegates may wish to re-run the Ocimum material privately or explore the technology.) All material is licensed CC BY or Apache 2 and can be used without permission for any purpose (teaching, research, software, as long as attributed).

directories

List of directories/resources in tigr2ess distribution and at http://github.com/petermr/tigr2ess. We shall refer to these throught the day. The definitive version will always be Github. Read the following to find where the resources are.

ContentMine Directories in tigr2ess.