RMIT at TREC 2010 Blog Track: Faceted Blog Distillation Task

This paper reports RMIT’s participation in the TREC Blog Track 2010. For the baseline task, we adopted the BM25 model implemented in the Zettair search engine to establish a retrieval system of blog posts based on topic relevance. We then experimented with a number of different approaches to aggregate the post similarity scores to retrieve the most relevant blogs. Similarly, for the faceted distillation task we built a system at the post level first. After that, scores are aggregated for blogs to re-rank the most relevant blogs for the facet inclinations. A SVM classifier has been trained on Blog 06 collection to produce the opinion scores for each post. The cross entropy is used to evaluate posts for the in-depth versus shallow facet. For the personal versus official facet, we assumed blogs which are opinionated are also personal.