Scholarlink

<p>Mass marketing techniques are no longer meeting the demands of the commercial arena. This change has resulted largely from the rise of e-commerce and the mounting demand for personalization and real-time relevance. Businesses, therefore, must transition these days toward highly targeted marketing by using advanced customer segmentations. This study attempts to apply K-Means and Fuzzy C-Means clustering algorithms on a large-scale e-commerce transaction dataset to harvest actionable consumer insights. K-Means clustering provides crisp and distinct clusters to aid in the development of stable strategies, whereas the FCM clustering algorithm gives overlapping clusters, thereby facilitating the understanding of more nuanced consumer behaviors. Our analyses suggest that while K-Means is marginally better in terms of computational efficiency and interpretability, FCM sheds light on deeper, multidimensional customer traits essential for hyper-personalized engagements. This finding reinforces the importance of selecting clustering methods suited to specific business objectives. Hence, the study confirms that both approaches can powerfully contribute to data-driven marketing, customer retention, and behavioral analysis in digitally competitive marketplaces.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

The Growing Complexity of Consumer Choices: Unravelling Consumer Patterns with K-Means and Fuzzy Logic

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年11月20日

查看全文
  • Yashodhan Karulkar,
  • Shagun Srivastava,
  • Rakshit Nandwana,
  • Stacia Stanley

摘要

Mass marketing techniques are no longer meeting the demands of the commercial arena. This change has resulted largely from the rise of e-commerce and the mounting demand for personalization and real-time relevance. Businesses, therefore, must transition these days toward highly targeted marketing by using advanced customer segmentations. This study attempts to apply K-Means and Fuzzy C-Means clustering algorithms on a large-scale e-commerce transaction dataset to harvest actionable consumer insights. K-Means clustering provides crisp and distinct clusters to aid in the development of stable strategies, whereas the FCM clustering algorithm gives overlapping clusters, thereby facilitating the understanding of more nuanced consumer behaviors. Our analyses suggest that while K-Means is marginally better in terms of computational efficiency and interpretability, FCM sheds light on deeper, multidimensional customer traits essential for hyper-personalized engagements. This finding reinforces the importance of selecting clustering methods suited to specific business objectives. Hence, the study confirms that both approaches can powerfully contribute to data-driven marketing, customer retention, and behavioral analysis in digitally competitive marketplaces.