Research on Market Demand Forecasting and Informing Marketing Strategy for Sustainable Functional Textiles Based on New Media Data Analysis

Jing Chen

Article
2026 / Volume 9 / Pages 385-402
Received 15 October 2025; Accepted 4 November 2025; Published 28 February 2026
https://doi.org/10.31881/TLR.2026.385

Abstract
Accurate demand forecasting in the textile industry, especially for innovative functional textile products, is a significant challenge due to dynamic market trends. This study proposes a framework for forecasting market demand for sustainable textiles by analyzing new media data, directly linking consumer discourse on fabric properties to sales patterns. The methodology involves analyzing a specialized dataset from Instagram focusing on sustainable functional textiles, such as those made from recycled fibers or organic materials. A Long Short-Term Memory (LSTM) network is employed, integrating conventional sales data with metrics derived from consumer discussions about specific textile attributes. These attributes were identified using topic modeling and include material sourcing, certifications (e.g., GOTS, Oeko-Tex), and in-use performance characteristics like durability and breathability. The results show that this integrated model significantly improves forecasting accuracy, reducing the Mean Absolute Percentage Error (MAPE) by 18.5%. The analysis confirms that market demand is strongly influenced by technical textile properties and sustainability credentials rather than purely aesthetic factors. This research provides the sustainable functional textile sector with a robust, data-driven methodology to better anticipate market needs, thereby optimizing textile product inventory and informing future material development.

Keywords
sustainable textiles, functional textiles, demand forecasting, marketing strategy, new media data