A comparative study of Flickr tags and index terms in a general image collection

نویسنده

  • Abebe Rorissa
چکیده

concepts (I) 218 (5.24%) 92.39 0.22 (0.64) 25 (0.67%) 96.84 0.03 (0.16) Art historical information (I) 66 (1.59%) 93.98 0.07 (0.29) 88 (2.37%) 99.21 0.09 (0.32) People-related attributes (I) 111 (2.67%) 96.65 0.11 (0.37) 29 (0.78%) 100 0.03 (0.19) Visual elements (P) 75 (1.80%) 98.45 0.08 (0.35) 0 (0.00%) 100 0.00 (0.00) Color (P) 64 (1.54%) 100 0.07 (0.35) 0 (0.00%) 100 0.00 (0.00) Total 4159 (100%) 4.27 (4.04) 3709 (100%) 3.73 (1.48) Note. P=Perceptual; I= interpretive. TABLE 3. Distribution of Flickr tags and index terms according to two broad categories of Jörgensen’s (1998) categories of image attributes. Flickr tags (n= 975) Index terms (n= 996) Category Frequency (%) M (SD) Frequency (%) M (SD) Perceptual 3051(73.36%) 3.13(3.30) 2622(70.69%) 2.63(1.24) Interpretive 1108(26.64%) 1.14(1.43) 1087(29.31%) 1.09(0.81) Total 4159(100%) 4.27(4.04) 3709(100%) 3.73(1.48) Photographic Archive (76.87%), even though indexers at the Archive used refiners with more nonunique index terms than unique ones. Flickr users assigned more than 100 times as many unique tags than unique index terms assigned to images in the St. Andrews Library Photographic Archive. On the other hand, there were only 4.3 times as many unique index terms with refiners as Flickr tags with refiners. Table 4 presents the frequencies, percentages, means, and standard deviations for numbers of Flickr tags and St. Andrews Library Photographic Archive index terms within Enser and McGregor’s four categories. A closer look at the frequencies and the mean numbers of tags and index terms that are either unique or nonunique and are with or without refiners reveals an even more apparent difference according to Enser and McGregor’s (1992) framework of image attributes. These differences were confirmed by results of a chi-square analysis and subsequent t-tests. The chi-square analysis resulted in a significant chi-square value, χ2 (3, N = 7,868)= 4261, p< 0.05. An independent group’s t-test (with unequal variances) was also conducted to see if the mean number of unique and nonunique Flickr tags and index terms were comparable. The t-tests resulted in significant differences (p< 0.05). These findings are in contrast with Enser and McGregor’s (1992) analysis of 2,700 requests submitted to the Hulton Deutsch collection (a general-purpose picture archive that is part of Getty Images). Their analysis found that the queries varied in their focus of topics. They showed that the bulk of the requests were for either specific objects or events and often refined by a number of facets such as spatial and temporal facets. In terms of the uniqueness and refiners used, the two samples (tags and

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

An Architecture for Real-time Textual Query Term Extraction from Images

This paper presents an approach that generates a set of search queries from a user’s query image. The approach provides the ability to query one’s environment in real-time, effectively allowing a user to ask ”What is that?”. This is achieved by capturing the user’s context through a geotagged photograph, and using it to filter a community image collection, for example Flickr, to retrieve a set ...

متن کامل

QUOTE: "Querying" Users as Oracles in Tag Engines a Semi-Supervised Learning Approach to Personalized Image Tagging

One common trend in image tagging research is to focus on visually relevant tags, and this tends to ignore the personal and social aspect of tags, especially on photoblogging websites such as Flickr. Previous work has correctly identified that many of the tags that users provide on images are not visually relevant (i.e. representative of the salient content in the image) and they go on to treat...

متن کامل

Tags Re-ranking Using Multi-level Features in Automatic Image Annotation

Automatic image annotation is a process in which computer systems automatically assign the textual tags related with visual content to a query image. In most cases, inappropriate tags generated by the users as well as the images without any tags among the challenges available in this field have a negative effect on the query's result. In this paper, a new method is presented for automatic image...

متن کامل

Facets of user-assigned tags and their effectiveness in image retrieval

1 Acknowledgements I would like to thank my supervisor, Dr. Pauline Rafferty, for her valuable support and encouraging feedback during the process of writing this dissertation, and also Marianne Taylor for her support during this time. I would also like to thank my son, Adam Ransom, who wrote and amended the computer program to download images and data from Flickr, without whose assistance I co...

متن کامل

Member activities and quality of tags in a collection of historical photographs in Flickr

There is growing interest in, and an increasing number of attempts by, traditional information providers to engage social content creation and sharing communities in creating and enhancing the metadata of their photo collections to make the collections more accessible and visible. To enable and guide effective metadata creation, however, it is essential to understand the structure and patterns ...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

عنوان ژورنال:
  • JASIST

دوره 61  شماره 

صفحات  -

تاریخ انتشار 2010