Sentence-Level Opinion-Topic Association for Opinion Detection in Blogs

The Opinion Detection from blogs has always been a challenge for researchers. One of the challenges faced is to find such documents that specifically contain opinion on users´ information need. This requires text processing on sentence level rather than on document level. In this paper, we have proposed an opinion detection approach. The proposed approach tries to tackle opinion detection problem by using some document level heuristics and processing documents on sentence level using different semantic similarity relations of WordNet between sentence words and list of weighted query terms expanded through encyclopedia Wikipedia. According to initial results, our approach performs well with MAP of 0.2177 with improvement of 28.89% over baseline results obtained through BM25 matching formula. TREC Blog 2006 data is used as test data collection.