Hierarchical Attention Transfer Network for Cross-Domain Sentiment Classification

Cross-domain sentiment classification aims to leverage useful information in a source domain to help do sentiment classifi- cation in a target domain that has no or little supervised infor- mation. Existing cross-domain sentiment classification meth- ods cannot automatically capture non-pivots, i.e., the domain- specific sentiment words, and pivots, i.e., the domain-shared sentiment words, simultaneously. In order to solve this prob- lem, we propose a Hierarchical Attention Transfer Network (HATN) for cross-domain sentiment classification. The pro- posed HATN provides a hierarchical attention transfer mech- anism which can transfer attentions for emotions across do- mains by automatically capturing pivots and non-pivots. Be- sides, the hierarchy of the attention mechanism mirrors the hierarchical structure of documents, which can help locate the pivots and non-pivots better. The proposed HATN consists of two hierarchical attention networks, with one named P-net aiming to find the pivots and the other named NP-net align- ing the non-pivots by using the pivots as a bridge. Specif- ically, P-net firstly conducts individual attention learning to provide positive and negative pivots for NP-net. Then, P- net and NP-net conduct joint attention learning such that the HATN can simultaneously capture pivots and non-pivots and realize transferring attentions for emotions across domains. Experiments on the Amazon review dataset demonstrate the effectiveness of HATN.