Predicting Subjectivity in Multimodal Conversations

In this research we aim to detect subjective sentences in multimodal conversations. We introduce a novel technique wherein subjective patterns are learned from both labeled and unlabeled data, using n-gram word sequences with varying levels of lexical instantiation. Applying this technique to meeting speech and email conversations, we gain significant improvement over state-of-the-art approaches. Furthermore, we show that coupling the pattern-based approach with features that capture characteristics of general conversation structure yields additional improvement.