Scholarlink

Enhancing retail decisions requires knowledge of how customers buy their products. Determining how customers group their purchases and anticipating more items they would buy constitutes a complex problem for retail companies. The research investigated ways to solve this problem by implementing a dynamic association rule mining algorithm which discovered valuable item relationships along with frequent itemsets. The interpretation of extracted patterns received enhancement through the utilization of a custom synergy score combined with zhang’s score. The developed metrics generated more sophisticated understanding of item association insights which performed better than standard evaluation approaches. The pattern discovery results produced a significant enhancement because the highest synergy score reached 1.097 and the average score exceeded 1 which indicated more accurate and relevant mined rules. The proposed approach demonstrated superior performance than deep learning models which were used for comparison purposes. Analyzed findings enable businesses to use them for personalizing marketing strategies and inventory control and retail strategy development to support their data-driven decision-making while better serving client needs. The research presents an improved method which analyzes buying patterns for customers while deriving meaningful business insights from retail information.

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

Enhancing Retail Insights: Introducing Dynamic Association Rule Mining over Deep Learning and Machine Learning

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Abhay Nath,
  • Aakanksha Kumari,
  • Ruma Pal,
  • Sachin Patel,
  • Amit Nayak

摘要

Enhancing retail decisions requires knowledge of how customers buy their products. Determining how customers group their purchases and anticipating more items they would buy constitutes a complex problem for retail companies. The research investigated ways to solve this problem by implementing a dynamic association rule mining algorithm which discovered valuable item relationships along with frequent itemsets. The interpretation of extracted patterns received enhancement through the utilization of a custom synergy score combined with zhang’s score. The developed metrics generated more sophisticated understanding of item association insights which performed better than standard evaluation approaches. The pattern discovery results produced a significant enhancement because the highest synergy score reached 1.097 and the average score exceeded 1 which indicated more accurate and relevant mined rules. The proposed approach demonstrated superior performance than deep learning models which were used for comparison purposes. Analyzed findings enable businesses to use them for personalizing marketing strategies and inventory control and retail strategy development to support their data-driven decision-making while better serving client needs. The research presents an improved method which analyzes buying patterns for customers while deriving meaningful business insights from retail information.