An Information Retrieval-Based System for Multi-domain Sentiment Analysis

摘要: This paper describes the SHELLFBK system that participated in ESWC 2015 Sentiment Analysis challenge. Our takes a supervised approach builds on techniques from information retrieval. The algorithm populates an inverted index with pseudo-documents encode dependency parse relationships extracted sentences training set. Each record stored is annotated polarity and domain of sentence it represents; this way, possible to have more fine-grained representation learnt sentiment information. When new has be computed, converted query two-steps computation performed: firstly, assigned by comparing content contextual during phase, and, secondly, once sentence, computed sentence. Preliminary results in-vitro test case demonstrated promising results.