Deep Learning Applications in Digital Marketing of Luxury Brands: Consumer Sentiment Analysis Using YSL as a Case Study

Abstract

The world has gone digital; therefore, luxury fashion brands have had to adapt their brand-building methods to some extent. Apply various deep learning methods, such as Long Short-Term Memory (LSTM) networks, Bidirectional Encoder Representations from Transformers (BERT), and Convolutional Neural Networks (CNN), to conduct consumer sentiment analysis for Yves Saint Laurent (YSL) in the luxury fashion industry in this paper. Based on a corpus of over 120,000 user-generated content (UGC) items collected from Weibo, Instagram and other brand-owned platforms between 2020 and 2024, multiple deep learning models for multi-class sentiment classification and aspect-level analysis have been constructed and evaluated in this study. According to the above experiments, the finely tuned BERT model has achieved a classification accuracy of 91.3% and exceeded the results of the traditional machine-learning baseline by 14.7%. Based on the above results, the main reasons for changes in public opinion about YSL are quality issues, brand history and influencer endorsements. Based on the above analysis, some practical suggestions are put forward for managers of luxury brands, and at the same time, a data-driven digital marketing system for real-time sentiment analysis and campaign optimization is proposed.