Assessing the quality of socially created metadata to image indexing
Corinne Jörgensen, Besiki Stvilia, Shuheng Wu · Proceedings of the American Society for Information Science and Technology · 2011
Abstract Deploying photograph collections in existing social tagging communities (e.g., Flickr) to collect large sets of socially created metadata to extend existing controlled vocabularies is a recent activity in building Knowledge Organization Systems. This study evaluates the reuse value of social metadata for image indexing by using a controlled end‐user experimental design to elicit a set of quality criteria for index terms from users. The findings indicate that a number of quality criteria of image indexing are different from those of textual document indexing, such as moods, importance, generality, level of detail, and ease of use. This suggests that specific criteria need to be taken into account when integrating socially created metadata with image indexing, and that the two processes can be complementary