社交平台的时尚流行偏好与机构预测结果的差异性分析  

Discrepancy between fashion preferences on social platforms and institutional forecasting results

作　　者：刘丽娴[1,2] 陈明 李浩[1,2] 向忠[4] LIU Lixian;CHEN Ming;LI Hao;XIANG Zhong(Silk and Fashion Culture Center of Zhejiang,Zhejiang Sci-Tech University,Hangzhou,Zhejiang 311199,China;Zhejiang-France Digital|Fashion Collaboratine Laboratory Zhengjiang Sci-Tech University,Hangzhou,Zhejiang 310018,China;School of Fashion Design&Engineering,Zhejiang Sci-Tech University,Hangzhou,Zhejiang 311199,China;School of Mechanical Engineering,Zhejiang Sci-Tech University,Hangzhou,Zhejiang 310018,China)

机构地区：[1]浙江理工大学浙江省丝绸与时尚文化研究中心,浙江杭州311199 [2]浙江理工大学浙江-法国数字时尚联合实验室,浙江杭州310018 [3]浙江理工大学服装学院,浙江杭州311199 [4]浙江理工大学机械工程学院,浙江杭州310018

出　　处：《毛纺科技》2025年第1期90-96,共7页Wool Textile Journal

基　　金：浙江省哲学社会科学领军人才培养项目(22YJRC03ZD-1YB);中央引导地方科技发展资金项目(2023ZY1029);杭州市重大科技创新项目(2022AIZD0153)。

摘　　要：为了分析流行趋势机构预测结果与消费者时尚流行偏好的差异,以WGSN、亦服科技、蝶讯网这3家在时尚趋势预测领域颇具影响力的机构作为研究对象,梳理目前国内外消费者时尚偏好、时尚流行趋势预测的研究现状,以小红书、微博作为消费者偏好数据源,利用文本挖掘、关键词频统计、相似性分析等方法,将3组基于专家转述的流行趋势关键词分别与平台趋势关键词进行对比。结果表明:3组基于专家转述的流行趋势关键词与平台趋势关键词具有显著性差异,其中蝶讯网的预测结果与消费者偏好相似度较高,其次是亦服科技,WGSN的预测与消费者偏好的差异性较大。可为时尚趋势预测行业更好地满足消费者需求提供参考。In order to analyze the differences between organizations′prediction results and consumers′fashion trend preferences,WGSN,IF Technology,DICTION,the three influential agencies in the field of fashion trend prediction were selected as the research object,by combing the current research status of consumers′fashion preferences and fashion trend prediction at home and abroad,and by taking Xiaohongshu and Weibo as the data source of consumers′preferences and utilizes the methodologies of text mining,keywords frequency statistics,and similarity using text mining,keywords frequency statistics,similarity analysis and other methods,the three groups of expert-based fashion trend keywords were compared with the platform trend keywords respectively.The results show that the three groups of popular trend keywords based on experts′retellings have significant differences with platform trend keywords,in which DICTION prediction results have a higher similarity with consumer preferences,followed by IF Technology,and WGSN′s prediction has a higher discrepancy with consumer preferences.This study helps to match trend forecasting results with actual consumer preferences,and provide a reference for the fashion trend forecasting industry to better meet consumer needs.

关 键 词：消费者时尚偏好 趋势预测 社交平台 文本数据 差异性分析 

分 类 号：TS941[轻工技术与工程—服装设计与工程]