The MUCKE Information Retrieval and Evaluation System
This project presents a multimedia retrieval system framework that incorporates components for processing multimedia content in different modes and languages. The framework provides concept-based information retrieval facilities that applies credibility information for result re-ranking. The architecture combines both a direct user interface and a batched evaluation interface for reproducible research in multimedia IR. At the current state, the software package represernts a demo as a first, preliminary version of the system framework that demonstrates a use case based on an image test collection that provides user crediblity information.