Using semantic resources in an image retrieval system
Journal Title: Romanian Journal of Human - Computer Interaction - Year 2014, Vol 7, Issue 4
Abstract
In the recent years, with the development of the intelligent devices, users are sharing more and more multimedia data across the Internet using social networks. This is the reason why, web search has also become the main method for people to fulfill their information needs. Nevertheless, searching for diverse data is not that straightforward, since most information retrieval systems base their algorithms on popularity, relevance, rather than dissimilarity. In the CHIST-ERA project MUCKE (Multimedia and User Credibility Knowledge Extraction), we created a multimedia retrieval system that takes into consideration both relevance and diversity. To address the problem of diversification, we explore two approaches: (1) using semantic resources (like YAGO) in the processing of the initial query, and (2) the operations of clustering the results provided by us, based on the images titles or based on their content. In this paper, we present the first approach and we compare the results of our solution with the results provided by the Google search engine.
Authors and Affiliations
Alexandra-Mihaela SiriĊ£eanu, Adrian Iftene
Keywords
Related Articles
- EP ID EP28957
- DOI -
- Views 373
- Downloads 8
Online Communities: Approaches, Principles, Dilemmas
Online communities mediated by computer facilitate the gathering and the group interaction throughout social support, information and the feeling of belonging. The social interaction areas created by technologies that al...
Paronym Generation Algorithms for Malapropism Correction
The Web pages have been intensively used lately for automatic or semiautomatic extraction of useful information. Because of the open nature of the Web, the texts that have no spelling errors are very rare exceptions. One...
Improving parsing using morpho-syntactic and semantic information
In this paper we present the efforts of creating a syntactic parser with a very good performance on Romanian sentences. Instead of creating a parser from scratch, we decided to test the freely available existing ones and...
The Pedagogical Usability - Methodological and Practical Considerations
The paper presents some specific aspects of the usability of web-based e-learning systems, focusing on the concept of pedagogical usability as a quality feature distinct from general usability. There are briefly analyzed...
A Corpus of Multi-participant Conversations and an Editor for its Annotation
The paper presents the details and the history of building a corpus of instant messenger (chat) conversations with multiple participants used for Computer-Supported Collaborative Learning. Two XML annotation schemas are...