Scholarlink

<p>User purchase behaviour (UPB) analysis is essential in enhancing marketing strategies on e-commerce platforms. The user purchasing pattern is determined from the product analysis to procurement evaluation, directly influencing business growth. User behaviour is challenging to predict because user requirements frequently change in a dynamic environment. Traditional methods utilize various learning strategies to understand consumer purchasing patterns. However, the existing techniques fail to learn the purchasing patterns from high-dimensionality and overfitting data. The research difficulties are addressed using Deep Generative Adversarial Networks (DGAN), which use the generator and discriminator to observe the purchase pattern. The DGAN model uses the nearest neighbouring integrated data formulation procedure to eliminate the missing data. Then, the normalization and standardization procedure that format and simplify the computation procedure. Finally, consumer profiles are generated to predict and differentiate the false data, improving the overall UPB efficiency with a minimum error rate. The system uses the E-commerce Customer Behaviour Dataset to evaluate system efficiency, and DGAN attains 98.23% accuracy. Thus, the introduced system achieves robustness and scalability and improves user purchase behaviour patterns.</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 Application of Deep Learning in User Purchase Behaviour Analysis and Marketing Decision Support

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年10月7日

查看全文
  • Zhi Wang

摘要

User purchase behaviour (UPB) analysis is essential in enhancing marketing strategies on e-commerce platforms. The user purchasing pattern is determined from the product analysis to procurement evaluation, directly influencing business growth. User behaviour is challenging to predict because user requirements frequently change in a dynamic environment. Traditional methods utilize various learning strategies to understand consumer purchasing patterns. However, the existing techniques fail to learn the purchasing patterns from high-dimensionality and overfitting data. The research difficulties are addressed using Deep Generative Adversarial Networks (DGAN), which use the generator and discriminator to observe the purchase pattern. The DGAN model uses the nearest neighbouring integrated data formulation procedure to eliminate the missing data. Then, the normalization and standardization procedure that format and simplify the computation procedure. Finally, consumer profiles are generated to predict and differentiate the false data, improving the overall UPB efficiency with a minimum error rate. The system uses the E-commerce Customer Behaviour Dataset to evaluate system efficiency, and DGAN attains 98.23% accuracy. Thus, the introduced system achieves robustness and scalability and improves user purchase behaviour patterns.