³ÉÈË´óƬ

This session will explore a novel approach that integrates deep learning and social media analytics to enhance reverse logistics decision-making. The primary challenge lies in the need to reduce waste and optimize the reuse and recycling of returned goods by effectively incorporating customer feedback—often overlooked in traditional reverse logistics frameworks.

This approach addresses the complexity of analyzing vast, unstructured social media data, from X, Meta, and Reddit, which is crucial for understanding the underlying reasons for product returns. We have developed a hybrid deep learning model, combining Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) networks, to process and classify sentiment data from these platforms. The session will present real-world case studies demonstrating how integrating this data into reverse logistics decisions can optimize business operations, reduce waste, and enhance customer satisfaction, ultimately driving strategic decision-making and promoting sustainability.