基于文本挖掘的蒙古族"时装"消费影响因素研究OA
Research on the influencing factors of Mongolian"fashion"consumption based on text mining
为探究消费者在线购买蒙古族"时装"的关注因素,文章以抖音商城蒙古族"时装"购买的在线评论为依据,采用文本挖掘方法分析其消费影响因素.通过情感分析、聚类分析与主题建模等方法识别出评论数据中产品质量、外观版型、舒适度、价格感知、服务体验等关键维度.研究结果表明:当下消费者对蒙古族"时装"整体的视觉美感与工艺质量等方面的满意度相对较高,但价格适配度、尺码准确性与日常场景适用性等方面的缺失也显著影响着消费者满意度.文章从消费反馈视角厘清蒙古族"时装"消费过程中影响因素、感知价值与满意度之间的递进关系,为品牌在产品创新、定价策略、服务优化与文化转译等方面提供决策参考与有益启示.
To explore the key factors consumers focus on when purchasing Mongolian"fashion"online,this study utilizes user reviews of related products from Douyin Mall as data and applies text mining methods for analysis.Through sentiment analysis,clustering,and topic modeling,critical dimensions such as product quality,appearance and cut,comfort,price perception,and service experience were identified.The results indicate that while consumers are relatively satisfied with the visual appeal and craftsmanship,shortcomings in price appropriateness,sizing accuracy,and suitability for daily wear significantly impact overall satisfaction.This reveals the structure of value perception for Mongolian"fashion"within the e-commerce context from the perspective of consumer feedback,providing insights for product innovation,pricing,service optimization,and cultural translation. The research obtained 10 363 valid reviews.Topic identification was performed using the LDA model,consumer perceived value was parsed through multimodal text mining,and price stratification was identified via visualization analysis.As for topic modeling,a three-layer"document-topic-term"probabilistic mode was constructed,and the distribution of themes in positive versus negative reviews was compared through sentiment analysis.Sales analysis shows products are mainly distributed across three price ranges:99-199 yuan,200-499 yuan,and 500-999 yuan,with various promotional forms used.A chi-square test indicates a significant association between sales volume and labeling strategy(χ2=28.41,p<0.001). With a threshold of 0.5 for sentiment analysis,it is found that most reviews are positive,reflecting a generally high satisfaction level.Analysis of positive reviews(Set A)shows that perceived value concentrates on five aspects:craftsmanship quality,appearance and cut,practical wearing experience,price acceptance,and recommendation willingness.Negative reviews(Set B)focus on five dimensions:craftsmanship flaws,aesthetic dissonance,functional inadequacy,value imbalance,and cultural alienation.On the whole,online purchase satisfaction is primarily influenced by four factors:product quality,pricing and promotions,packaging and logistics experience,and customer service quality. The study demonstrates that:consumers value both the cultural and aesthetic appeal signified by terms like"good-looking"and"Mongolian,"and practical attributes such as"comfort"and"everyday wear;"while basic quality like"fabric"and"workmanship"is acknowledged,issues like"size deviation"and"physical discomfort"remain constraints on the experience;pricing strategy significantly impacts decisions,with the correlation between promotional labels and sales volume reflecting consumer price sensitivity alongside cultural value appreciation;logistics efficiency and customer service quality profoundly affect brand reputation,particularly in non-core regions where delays or poor responsiveness can amplify negative experiences. This study is limited by the nature of review content and technical scope,preventing a more granular deconstruction of quality indicators or a cross-platform comparison of sales and feedback across different e-commerce environments.Furthermore,platform privacy mechanisms make it difficult to establish correspondences between reviews and consumer profiles.Future research could employ interviews and questionnaires to conduct a more in-depth analysis across dimensions such as gender,age,and region,so as to further enrich studies in this field.
木斯;雷青;王思琴
内蒙古师范大学 设计学院,呼和浩特 010022内蒙古师范大学 设计学院,呼和浩特 010022内蒙古师范大学 设计学院,呼和浩特 010022
轻工纺织
蒙古族"时装"文本挖掘抖音商城在线评论情感分析聚类分析主题建模
Mongolian"fashion"text miningDouyin Mallonline commentssentiment analysiscluster analysistopic modeling
《丝绸》 2026 (3)
30-40,11
内蒙古师范大学引进高层次人才科研启动金专项资金资助项目(2025YJRC026)内蒙古自治区教育科学规划项目(JG2024016C)
评论