Leveraging Artificial Intelligence to Analyze Consumer Sentiments within Their Context: A Case Study of Always #LikeAGirl Campaign

Abstract

Abstract Because practitioners and scholars are increasingly using artificial intelligence (AI) to analyze consumer sentiments toward social media–based campaigns, we compared various supervised machine-learning (ML) algorithms (four traditional ML-based algorithms and two proprietary deep-learning-based models from Amazon and Google) in terms of their performances in classifying user comments into a sentiment category. We adopted the #LikeAGirl campaign by Always feminine hygiene brand as a case and analyzed the sentiments of 19,198 YouTube comments on the campaign using different supervised ML algorithms. Results indicate that the two proprietary models from Amazon and Google performed better than traditional supervised ML algorithms. We then used unsupervised ML to reveal the hidden topics from positive and negative consumer comments. The study has important methodological implications for advertising practitioners and scholars.