I have conflicting feelings about mixing family time and work time. I consider myself a workaholic but on the other hand, would like to remain sane and love my family. Have to find time for both, but sometimes do not know how. So, here I go again in mixing my family time, when I'm supposed to be fixing dinner with writing a blog and learing to Tweet.
Nowadays, I get all excited to see articles that related to my work like this one about pictures, fliker, metadata and maps.
The process developed by the team did not rely on geo-tagged photos, but used various clues to interpret location from metadata and the images themselves. The project was part funded by Google, Yahoo and the MacArthur Foundation.We have tried to do some analysis for WorldCat Registry and flikr. Took us some tries to better understand how pictures are tagged and if it's worth the trouble to connect the metadata in both of these applications.
We developed classification methods for characterizing these locations from visual, textual and temporal features," explained Daniel Huttenlocher, professor of computing, information science and business. These methods reveal that both visual and temporal features improve the ability to estimate the location of a photo compared to using just textual tags.
For now, we're staying put. Will see what comes out of it.
