Scholarlink

Visualisations of sales offers have a significant impact on their reception. Given the difficulty in assessing the impact of visualizations on consumer decisions, this study attempted to determine whether a consumer's decision can be recognized by how they look at offers. Eye-tracking data obtained from viewing house sales offers was used to develop a classifier model for predicting decisions. The research results show that the use of appropriate classification algorithms (e.g., C4.5, DT, kNN, ANN), methods to reduce the impact of data imbalance on results (e.g., SMOTE), or attribute filtering methods (e.g., CFS) provide effective methodological approaches to develop such a classifier.

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

Machine Learning-Based Exploration of Eye-Tracking Data to Predict Offer Selection

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Mateusz Piwowarski,
  • Paweł Ziemba,
  • Jacek Cypryjański

摘要

Visualisations of sales offers have a significant impact on their reception. Given the difficulty in assessing the impact of visualizations on consumer decisions, this study attempted to determine whether a consumer's decision can be recognized by how they look at offers. Eye-tracking data obtained from viewing house sales offers was used to develop a classifier model for predicting decisions. The research results show that the use of appropriate classification algorithms (e.g., C4.5, DT, kNN, ANN), methods to reduce the impact of data imbalance on results (e.g., SMOTE), or attribute filtering methods (e.g., CFS) provide effective methodological approaches to develop such a classifier.