Content mining hackathon: Unterschied zwischen den Versionen
Pk (Diskussion | Beiträge) (foormatting) |
Pk (Diskussion | Beiträge) (+adr) |
||
Zeile 14: | Zeile 14: | ||
* Start 10:00 | * Start 10:00 | ||
* End 18:00 | * End 18:00 | ||
* main room | * at Metalab, Vienna, [[Lage|Rathausstraße 6]], main room | ||
The scientific literature (papers, theses, reports) is an untapped mine of hugely valuable content. Traditionally data has been published as tables, figures or running text (recipes, accounts of observations, etc.). Although there is a movement to publish data explicitly most scientific results are still presented as text. Artificial Intelligence techniques are now robust and accessible to everyone and can give very good recall and precision for extracting this data. | The scientific literature (papers, theses, reports) is an untapped mine of hugely valuable content. Traditionally data has been published as tables, figures or running text (recipes, accounts of observations, etc.). Although there is a movement to publish data explicitly most scientific results are still presented as text. Artificial Intelligence techniques are now robust and accessible to everyone and can give very good recall and precision for extracting this data. |
Aktuelle Version vom 17. Mai 2014, 00:57 Uhr
Content Mining Hackathon & Open Science MeetUp | |
Thu 2014-06-05, 11:00 | |
socialhack | |
Workshop | |
0€ | |
active | |
The scientific literature (papers, theses, reports) is an untapped mine of hugely valuable content. Traditionally data has been published as tables, figures or running text (recipes, accounts of observations, etc.). Although there is a movement to publish data explicitly most scientific results are still presented as text. Artificial Intelligence techniques are now robust and accessible to everyone and can give very good recall and precision for extracting this data. | |
Zuletzt aktualisiert: | 17.05.2014 |
Content Mining Hackathon
- Start 10:00
- End 18:00
- at Metalab, Vienna, Rathausstraße 6, main room
The scientific literature (papers, theses, reports) is an untapped mine of hugely valuable content. Traditionally data has been published as tables, figures or running text (recipes, accounts of observations, etc.). Although there is a movement to publish data explicitly most scientific results are still presented as text. Artificial Intelligence techniques are now robust and accessible to everyone and can give very good recall and precision for extracting this data.
We shall concentrate on applied Computer Vision (CV) for scientific diagrams, coupled with Natural Language Processing for text. We'll take bioscience and chemistry as examples. The examples we shall take can be understood be everyone - species, dates, licences, granting bodies, identifier systems. We'll show how to use the Open Source AMI2 system and also show how YOU can contribute. AMI2 has been developed with a "plugin" architecture so you can create your now - in an hour or two.
Whether you are a geek or a scientist or a wonk or an informed citizen everyone has something to bring to a hackathon and take away. You don't have to be a geek - we form groups where different skills and experience work together. For example if you understand Wikipedia that's a great resource to contribute. And the hackathon philosophy is a great tool to take to other events and get everyone hacking together.
The hackathon will be held in English, but there will be mostly German speaking people.
OKFN Open Science MeetUp: Content Mining
- Start 19:00
- End 22:00
- main room
Meetup of the Open Science Working Group of the Austrian Open Knowledge Foundation around the topic of content mining. We are happy to welcome the actual Shuttleworth Fellow and Open Science enthusiast Peter Murray Rust, who will tell about his recent activities around content mining in context of scientific publications. There will be several lightning talks around this and also the results of the [link/zu/hackathon hackathon] will be presendet and discussed. The MeetUp will be hold in English.