Contextual and Position-Aware Factorization Machines for Sentiment Classification
This work addresses sentiment classification for fine-grained analysis, offering incremental improvements for snippet-level tasks.
The paper tackles the problem of fine-grained sentiment analysis at the snippet level by developing Position-aware Factorization Machines that capture sentiment-oriented word interaction, context, and position information, resulting in improved performance for snippet/sentence-level classification while matching state-of-the-art methods at the document level.
While existing machine learning models have achieved great success for sentiment classification, they typically do not explicitly capture sentiment-oriented word interaction, which can lead to poor results for fine-grained analysis at the snippet level (a phrase or sentence). Factorization Machine provides a possible approach to learning element-wise interaction for recommender systems, but they are not directly applicable to our task due to the inability to model contexts and word sequences. In this work, we develop two Position-aware Factorization Machines which consider word interaction, context and position information. Such information is jointly encoded in a set of sentiment-oriented word interaction vectors. Compared to traditional word embeddings, SWI vectors explicitly capture sentiment-oriented word interaction and simplify the parameter learning. Experimental results show that while they have comparable performance with state-of-the-art methods for document-level classification, they benefit the snippet/sentence-level sentiment analysis.