4
edits
Peter Murray-Rust: Difference between revisions
Semantic tools and content for climate science and policy
(Created page with "{{Participant |Timezone=Europe/London (GMT+00:00/GMT+01:00) |Affiliation=University of Cambridge |Projects=https://semanticclimate.github.io/p/en/ |Interests=ontologies textMining Wikidata |Discord Handle=handle#1941 }} {{Workshop Submission}}") |
(Semantic tools and content for climate science and policy) |
||
| Line 5: | Line 5: | ||
|Interests=ontologies textMining Wikidata | |Interests=ontologies textMining Wikidata | ||
|Discord Handle=handle#1941 | |Discord Handle=handle#1941 | ||
|ORCID=0000-0003-3386-3972 | |||
|Twitter=petermurrayrust | |||
|Github=petermr | |||
}} | }} | ||
{{Workshop Submission}} | {{Workshop Submission | ||
|Interest=Have been running Open Free Collaborative systems for 25 years (BBK-PPS, BlueObelisk, ContentMine, and now #semanticClimate. This is a global Open community, centered in India, where we are making climate documents (such as the IPCC's recent 10,000 pages) semantic. | |||
|Frame=Tool-builder, Researcher | |||
|Materials=HTML versions of the IPCC report. | |||
A suite of Python tools for searching, processing, cleaning and analysing, | |||
Semantic climate dictionaries linked to Wikidata and hence into the Linked Open Data cloud | |||
}} | |||
Was invited to join (2022-11-04). | |||
I enjoy sparking off Open community action in creating interoperable systems for science. Worked with W3C in the creation of XML, which provides a framework for modelling the world. Models are driven by discourse (i.e. what people write) rather than god-given classifications. Henry Rzepa and I created Chemical Markup Language (CML) which is capable of modelling much of published chemistry and interoperates with MathML, HTML and SVG. I have always believed in "Rough consensus and running code" so I have written code (Java) that supports the design, The problems are now not conceptual but sociopolitical (scientists are conservative and prefer PDF which in not machine processable). | |||
The advent of Wikidata (2012-) now means we can map much of our discourse into the Linked Open Data cloud. | |||
Because we model discourse there is a major requirement for text-mining so I and colleagues have developed a suite of tools. pygetpapers (Ayush Garg) is a rapid search and download for Open Access science, docanalysis applies NLP tools and AMI is a framework for analysing and recombining scientific publications. We have recently concentrated on applying this to climate literature and making it semantic | |||