Advanced methods for multimedia content selection

• Grant number: 11/17366-2
• Support Opportunities: Regular Research Grants
• Start date: January 01, 2012
• End date: December 31, 2013
• Field of knowledge: Physical Sciences and Mathematics - Computer Science
• Principal Investigator: Marcelo Garcia Manzato
• Grantee: Marcelo Garcia Manzato
• Host Institution: Instituto de Ciências Matemáticas e de Computação (ICMC). Universidade de São Paulo (USP). São Carlos , SP, Brazil

Abstract

The knowledge of semantic information about the content and users' interests is an important factor to provide multimedia selection and recommendation services, which filter the data according to each user's intentions and preferences. On the other hand, the lack of generic and efficient techniques to extract such high-level information contributes to the well-known semantic gap problem. The use of peer-level annotations can reduce this problem, because semantically richer metadata can be gathered from the analysis of users interaction. However, in order to make possible this strategy, some problems must be overcome, such as the presence of noise and irrelevant data, the way of constructing and representing such information, the lack of mechanisms to transform annotations into semantically structured representations, and how to apply annotation-based metadata in real applications. This plan of research has the objective of studying such challenges in order to make possible the development of content selection services. It is expected semantic metadata about content and user's preferences to be extracted collaboratively by end users in a unrestrictive way. (AU)