iHub seems to be carrying on a lot of activity of late. Just days following the biggest and baddest Barcamp in Africa yet, the iHub hosted the SwiftRiver team for a day of introducing the community to the SwiftRiver initiative.
Jon Gosier is the team leader for the SwiftRiver project which is an initiative under the Ushahidi umbrella. The whole point of SwiftRiver (which we have covered before) is to literally make sense out of a lot of ‘unstructured’ or semi-structured textual data such as tweets or SMS. The usefulness of this could be clearly seen in the case of disaster situations where there is a lot of data on such media as Twitter and one might want to make some sense of that data, verify that data and possibly making some good use of it such as perhaps informing some decision making. There are other situations where this could be applied, of course, say brand monitoring.
SwiftRiver 101 covered the basics of the technology which is basically Machine Language and Natural Language Processing stuff as well as algorithms for verifying data; as well as some of the key technologies that have been produced through the project:
- SiLCC – semantic tag extraction
- SULSa – location extraction
- SiCDS – content duplication filtering (retweets, xml, etc.)
For more info visit:
There may be an ‘Ushahidi 101′ coming up soon as well so keep your ear on the ground!