Abstract
PurposeThis paper seeks to demonstrate the value of using social media to capture fashion trends, including the popularity of specific features of clothing, in order to improve the speed and accuracy of supply chain response in the era of fast fashion.Design/methodology/approachThis study examines the role that text mining can play to improve trend recognition in the fashion industry. Researchers used n-gram analysis to design a social media trend detection tool referred to here as the Twitter Trend Tool (3Ts). This tool was applied to a Twitter dataset to identify trends whose validity was then checked against Google Trends.FindingsThe results suggest that Twitter data are trend representative and can be used to identify the apparel features that are most in demand in near real time.Originality/valueThe 3Ts introduced in this research contributes to the field of fashion analytics by offering a novel method for employing big data from social media to identify consumer preferences in fashion elements and analyzes consumer preferences to improve demand planning.Practical implicationsThe 3Ts improves forecasting models and helps inform marketing campaigns in the apparel retail industry, especially in fast fashion.
Subject
Marketing,Business and International Management
Reference66 articles.
1. Tweetluenza: predicting flu trends from Twitter data;Big Data Mining and Analytics,2019
2. Predicting the future with social media,2010
3. Fashion supply chains and social media: Examining the potential of data analysis of social-media texts for decision making-processes in fashion supply chains.;In Business Models and ICT Technologies for the Fashion Supply Chain: Proceedings of IT4Fashion 2016,2017
4. A survey on retail sales forecasting and prediction in fashion markets;Systems Science & Control Engineering,2015