Roman Kern, M. Granitzer, Viktoria Pammer-Schindler
{"title":"扩展大众分类法用于图像标记","authors":"Roman Kern, M. Granitzer, Viktoria Pammer-Schindler","doi":"10.1109/WIAMIS.2008.43","DOIUrl":null,"url":null,"abstract":"Due to the unsatisfactory results of content based image retrieval methods, organisation and retrieval of multimedia data strongly relies on metadata and free text description. Folksonomies, collaboratively created sets of metadata, merged recently and help organising multimedia information on the Web. Our contribution addresses the question how to extend a classical folksonomy with additional metadata. We also discuss the quality of the extended folksonomy and its application for tag recommendation. We show that some relations of the original folksonomies can be replaced while others are unique. In addition our analysis shows, that for 40% the correct tag is in the first 10% of the set of tag recommendations.","PeriodicalId":325635,"journal":{"name":"2008 Ninth International Workshop on Image Analysis for Multimedia Interactive Services","volume":"1 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2008-05-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"20","resultStr":"{\"title\":\"Extending Folksonomies for Image Tagging\",\"authors\":\"Roman Kern, M. Granitzer, Viktoria Pammer-Schindler\",\"doi\":\"10.1109/WIAMIS.2008.43\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Due to the unsatisfactory results of content based image retrieval methods, organisation and retrieval of multimedia data strongly relies on metadata and free text description. Folksonomies, collaboratively created sets of metadata, merged recently and help organising multimedia information on the Web. Our contribution addresses the question how to extend a classical folksonomy with additional metadata. We also discuss the quality of the extended folksonomy and its application for tag recommendation. We show that some relations of the original folksonomies can be replaced while others are unique. In addition our analysis shows, that for 40% the correct tag is in the first 10% of the set of tag recommendations.\",\"PeriodicalId\":325635,\"journal\":{\"name\":\"2008 Ninth International Workshop on Image Analysis for Multimedia Interactive Services\",\"volume\":\"1 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2008-05-07\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"20\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2008 Ninth International Workshop on Image Analysis for Multimedia Interactive Services\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/WIAMIS.2008.43\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2008 Ninth International Workshop on Image Analysis for Multimedia Interactive Services","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/WIAMIS.2008.43","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Due to the unsatisfactory results of content based image retrieval methods, organisation and retrieval of multimedia data strongly relies on metadata and free text description. Folksonomies, collaboratively created sets of metadata, merged recently and help organising multimedia information on the Web. Our contribution addresses the question how to extend a classical folksonomy with additional metadata. We also discuss the quality of the extended folksonomy and its application for tag recommendation. We show that some relations of the original folksonomies can be replaced while others are unique. In addition our analysis shows, that for 40% the correct tag is in the first 10% of the set of tag recommendations.