Implementing an image search system with integrating social tags and DBpedia

Authors:
Chie Iijima;Makito Kimura;Takahira Yamaguchi
Affiliations:
Keio University, Yokohama, Japan;Keio University, Yokohama, Japan;Keio University, Yokohama, Japan
Venue:
KES'10 Proceedings of the 14th international conference on Knowledge-based and intelligent information and engineering systems: Part III
Year:
2010

Citing 4
Cited 0

Recommender systems

Communications of the ACM
Being accurate is not enough: how accuracy metrics have hurt recommender systems

CHI '06 Extended Abstracts on Human Factors in Computing Systems
YAGO: A Large Ontology from Wikipedia and WordNet

Web Semantics: Science, Services and Agents on the World Wide Web
Metrics for evaluating the serendipity of recommendation lists

JSAI'07 Proceedings of the 2007 conference on New frontiers in artificial intelligence

Quantified Score

Hi-index	0.00

Visualization

Abstract

Although the number of recommending system has increased, many of the existing recommending systems often only offer general-purpose information. In the case of multimedia searches, novelty and unexpectedness are seen as particularly important. In this paper, we propose an image search method with a high degree of unexpectedness by integrating the social tag of Flickr and DBpedia, and using preference data from search logs. We also propose an image search system named Linked Flickr Search, which implemented the proposed method. By evaluation with an unexpectedness index, and by comparing the basic Flickr search functions and flickr wrappr, which is related research, we confirmed that particularly in the initial stages of the search, our proposed system was possible to recommend highly unexpected images.