In conjunction with the VIE 2008
Special Issue on Dependable Semantic Inference
Call for Papers
Many exciting achievements in the multimedia information retrieval (MIR) field in the past years have made the field mature enough to enter a new development phase --- the phase in which the MIR technology should start to be adopted in practical solutions and realistic application scenarios. As users' expectations are generally high, one of the most important remaining research challenges in the field is to improve dependability of MIR systems. For example, in the case of consumer-oriented multimedia retrieval solutions (e.g. a personal video recorder, a mobile video retrieval system, a music search framework, a web search engine), the service they provide in terms of the paradigm getting the content I like, anytime and anyplace will have to be at least as dependable as the button turning a TV set or a mobile device on and off. In other domains, such as automated surveillance, where MIR theory and algorithms may be employed to automatically analyze surveillance data and alert the authorities in case of a threatening situation, the dependability plays even a more critical role.
This special issue addresses the dependability of those critical parts of MIR systems dealing with semantic inference. Semantic inference refers to a set of theories and algorithms designed to relate multimedia data to semantic-level descriptors allowing content-based search, retrieval and management of data. An increase in semantic inference dependability could be achieved in several ways. For instance, a better understanding of the processes underlying semantic concept detection could help forecast, prevent or correct possible semantic inference errors. Furthermore, next to optimizing the performance of individual semantic inference systems, the theory of using redundancy for building reliable structures from less reliable components could be applied to integrate "isolated" semantic inference algorithms into a network characterized by distributed and collaborative intelligence. This could be either a social network of users tagging the content and seamlessly recommending the content to each other through collaborative filtering, and/or a network of semantic inference devices that learn from each other and complement each other. Finally, mechanisms need to be provided for evaluating the developed solutions according to standard dependability criteria such as fault detection, fault diagnosis and recovery (service restoration), outage, and fault forecasting.
The goal of this special issue is to gather high --- quality and original contributions that reach beyond conventional ideas and approaches and make substantial steps towards dependable, practically deployable semantic inference theories and algorithms.
Topics of interest include (but are not limited to):
- Theory and algorithms of generic and scalable semantic inference
- Exploration of applicability scope and theoretical performance limits of semantic inference algorithms
- Modeling of system confidence in its semantic inference performance
- Dependability evaluation and characterization of a semantic inference system
- Matching user requirements to dependability criteria (e.g. mobile user, user at home, etc.)
- Context/user adaptation and self-improvement of an inference system through self-learning
- Interactive learning for online adaptable semantic inference
- Modeling synergies between different semantic inference mechanisms (e.g. content analysis, indexing through user interaction, collaborative filtering)
- Semantic inference based on synergetic integration of content analysis, user actions (e.g. tagging) and user/device collaboration (e.g. in social/P2P networks)
Alan Hanjalic, TU Delft, Netherlands (Lead Guest Editor)
Tat-Seng Chua, National University of Singapore (to be confirmed)
Edward Chang, Google, China/UC Santa Barbara, USA (to be confirmed)
Ramesh Jain, UC Irvine, USA (to be confirmed)