STEMM Institute Press
Science, Technology, Engineering, Management and Medicine
Research on Campus Beauty Consumption Characteristics and Precision Recommendation Based on Multi-source Data and Machine Learning
DOI: https://doi.org/10.62517/jbm.202609509
Author(s)
Jingyuan Yang*, Xiang Li, Xiuyun Zhang, Yawen Zhao, Xinru Guo, Zhen Hao
Affiliation(s)
School of Artificial Intelligence and Big Data, Henan University of Technology, Zhengzhou, Henan, China *Corresponding Author
Abstract
Driven by the growing appearance economy and self-pleasing consumption, China’s beauty sector continues expanding, despite major challenges, including slowing traffic growth, intensifying product homogenization and weak consumer brand loyalty. University students are significant potential consumers, but many of them are unaware of their skin condition and face the high trial expense and unsuitable products. This study combines the sales record data of e-commerce with online reviews and questionnaire feedback from Zhengzhou university students. It aims at investigating the campus beauty consumption, student segmentation and product-campus demand matching by utilizing text mining, sentiment analysis, descriptive statistics, K-Means clustering and association rule mining. A mobile app for cosmetic recommendations is also created. The results indicate that cost-effective items outsell other items on campus. There are three groups of students, namely those who use basic skincare; those who are cautious about buying skincare as they have concerns about their skin’s reaction; and those who use products often and have preferences. There is a strong correlation between product reputation and sales. Main grievances involve irritation, allergies, greasiness and effectiveness issues, with demand also fluctuating according to seasons and campus activities. These findings shed light on beauty consumer groups. This application facilitates rational purchases for students, targeted marketing to campus and personalized services. The approach and results can also be applied to other groups of consumers.
Keywords
Beauty Consumption; Text Mining; Sentiment Analysis; K-Means Clustering; Precision Recommendation Application; User Profiling
References
[1] Zheng Y, Li C. Body socialization: the practical logic of female college students’ cosmetics consumption. China Youth Study, 2025(5): 110-119. (in Chinese) [2] Yu Q. Investigation and analysis of female college students’ makeup consumption behavior. China Business (Theory Research), 2019(2): 11-12. (in Chinese) [3] Xia B, Han M, Zhu Y. Experiential marketing of L’Oréal in China: a case study based on consumer behavior of Chinese female university students//Proceedings of the International Conference on Economic Management and Cultural Industry. Paris: Atlantis Press, 2021. [4] Li B. Research on marketing strategy of Chinese domestic cosmetics based on female college students’ consumption characteristics//Proceedings of the 2021 6th International Conference on Social Sciences and Economic Development (ICSSSED 2021). Paris: Atlantis Press, 2021. [5] Kaur G, Sharma A. A deep learning-based model using hybrid feature extraction approach for consumer sentiment analysis. Journal of Big Data, 2023, 10: 61. [6] Mutinda J, Mwangi W, Okeyo G. Sentiment analysis of text reviews using lexicon-enhanced BERT embedding (LeBERT) model with convolutional neural network. Applied Sciences, 2023, 13(3): 1445. [7] Khan R H, Dofadar D F, Alam M G R, et al. LRFS: online shoppers’ behavior-based efficient customer segmentation model. IEEE Access, 2024, 12: 96462-96480. [8] Song H, Zhang H, Xing Z. Research on personalized recommendation system based on association rules. Journal of Physics: Conference Series, 2021, 1961(1): 012027. [9] Zhang M. Consumer behavior analysis based on Internet of Things platform and the development of precision marketing strategy for fresh food e-commerce. PeerJ Computer Science, 2023, 9: e1531. [10] Latha Y M, Rao B S. Amazon product recommendation system based on a modified convolutional neural network. ETRI Journal, 2024, 46(4): 633-647.
Copyright @ 2020-2035 STEMM Institute Press All Rights Reserved