Explaining a neural attention model for aspect-based sentiment classification using diagnostic classification

Abstract

Many high performance machine learning models for Aspect-Based Sentiment Classification (ABSC) produce black box models, and therefore barely explain how they classify a certain sentiment value towards an aspect. In this paper, we propose explanation models, that inspect the internal dynamics of a state-of-the-art neural attention model, the LCR-Rot-hop, by using a technique called Diagnostic Classification. Our diagnostic classifier is a simple neural network, which evaluates whether the internal layers of the LCR-Rot-hop model encode useful word information for classification, i.e., the part of speech, the sentiment value, the presence of aspect relation, and the aspect-related sentiment value of words. We conclude that the lower layers in the LCR-Rot-hop model encode the part of speech and the sentiment value, whereas the higher layers represent the presence of a relation with the aspect and the aspect-related sentiment value of words.

Graph Adaptive Semantic Transfer for Cross-domain Sentiment Classification

“…We optimize the model with an integrated adaptive strategy, which deeply explores the impacts of the adaptive graph structures for cross-domain sentiment classification. Although there have been some studies (e.g., in the ABSA task [14,31,40,46]) to utilize syntactic graph structures to enhance semantic representation, the transferability exploration for adaptive graph structure information is still limited. That is, our proposals are the first solution to learn domain adaptive graph semantics for CDSC, and we encourage more effective studies to be explored and further improve our graph adaptive framework.…”