Evaluating Fusion Techniques at Different Domains in ImageCLEF Subtasks.
Sergio Muñoz, Rafael Muñoz, Fernando Llopis · 2009
In our participation in the 2009 edition of the ImageCLEF task we pursued two objectives, first to expand the number of subtasks in which we evaluate our proposed multimodal fusion presented in previous works, MultiModal Local Context Analy-sis (MMLCA). Furthermore, we evaluated three new proposals: a subquery genera-tion technique based on clustering, a new variant of Multimodal Re-ranking TF-IDF (MMRR TF-IDF), finally we evaluated a term filtering technique for the medical do-main which is based on WordNet Domains. [4]. From the experiments conducted: On the one hand, we have confirmed that MMLCA performs better than the other local expansion techniques evaluated. On the other hand, the results show that our proposal of subquery generation based on clustering combined with PRF obtains es-pecially good results (the 5th best textual run of the Photo Retrieval subtask in terms of F-Meassure).