Graph Capsule Aggregation for Unaligned Multimodal Sequences

TL;DR

This paper introduces GraphCAGE, a graph-based neural model with Capsule Networks, to effectively analyze unaligned multimodal sequences for sentiment analysis, overcoming RNN limitations and improving interpretability.

Contribution

It proposes a novel graph capsule aggregation method that models unaligned multimodal data without sequence alignment, enhancing long-range dependency learning and interpretability.

Findings

Achieves state-of-the-art results on benchmark datasets.

Effectively models unaligned multimodal sequences.

Provides interpretable multimodal sentiment analysis.

Abstract

Humans express their opinions and emotions through multiple modalities which mainly consist of textual, acoustic and visual modalities. Prior works on multimodal sentiment analysis mostly apply Recurrent Neural Network (RNN) to model aligned multimodal sequences. However, it is unpractical to align multimodal sequences due to different sample rates for different modalities. Moreover, RNN is prone to the issues of gradient vanishing or exploding and it has limited capacity of learning long-range dependency which is the major obstacle to model unaligned multimodal sequences. In this paper, we introduce Graph Capsule Aggregation (GraphCAGE) to model unaligned multimodal sequences with graph-based neural model and Capsule Network. By converting sequence data into graph, the previously mentioned problems of RNN are avoided. In addition, the aggregation capability of Capsule Network and the graph-based structure enable our model to be interpretable and better solve the problem of long-range dependency. Experimental results suggest that GraphCAGE achieves state-of-the-art performance on two benchmark datasets with representations refined by Capsule Network and interpretation provided.