Deep multimodal fusion for persuasiveness prediction

TLDR

Persuasiveness measures a speaker’s influence on audience beliefs and behaviors, and its analysis is increasingly important as social multimedia spreads ideas, yet the field is hampered by limited annotated data. The study aims to predict persuasiveness using the POM dataset by developing a deep multimodal fusion architecture that integrates complementary modality information. The authors employ the POM dataset and design a deep multimodal fusion model that combines complementary modality signals to predict persuasiveness. The proposed model achieves significant performance gains over prior methods.

Abstract

Persuasiveness is a high-level personality trait that quantifies the influence a speaker has on the beliefs, attitudes, intentions, motivations, and behavior of the audience. With social multimedia becoming an important channel in propagating ideas and opinions, analyzing persuasiveness is very important. In this work, we use the publicly available Persuasive Opinion Multimedia (POM) dataset to study persuasion. One of the challenges associated with this problem is the limited amount of annotated data. To tackle this challenge, we present a deep multimodal fusion architecture which is able to leverage complementary information from individual modalities for predicting persuasiveness. Our methods show significant improvement in performance over previous approaches.