Scholarlink

The Online Rating Visualization (ORV) is a widely used tool along with text reviews that offer business intelligence about product performance to prospective e-buyers. The rapid growth in e-business environment has led to an increasing trend to host products and related analytics for prospective consumers in a manner that is informative, credible, and fair. Surprisingly, there is not much prior research done in making the ORV productive and informative. In this paper, we formulate a deep-time resolved view of buyer sentiment analysis called as DBRV (Deep-Bin Rating-Vote Visualization) based on equi-length time slices of ratings-votes. We also propose a novel performance measure called Popularity index ( \(\mathcal{P}\) ) which combines ratings-votes meaningfully in time differential manner to depict discriminant power of the DBRV. Simulations performed using real data of ratings-votes for iPhone 12 and Earbuds with 4139 and 8497 reviews respectively, shows that DBRV offers more informative insights about consumer experience than average ratings-votes. The proposed DBRV-1 depicts time-binning of ratings-votes distribution thereby illustrating its information resolved into time-bins. Likewise, DBRV-2 powered by \(\mathcal{P}\) conveys higher intelligence about perceived product performance than existing ORV. Finally, we also formulate a more effective and novel user-interactive dashboard, shown for iPhone 12 and Earbuds, that offers a promising frontend analytics for prospective customers.

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

Deep Multi-time Visualization and Analytics for E-commerce Platform

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Jay R. Bhatnagar,
  • Ravi Prakash Iyer,
  • Krishna Rawat,
  • Ankit Tripathi,
  • Akhil Bhatnagar,
  • Dhruv Goyal

摘要

The Online Rating Visualization (ORV) is a widely used tool along with text reviews that offer business intelligence about product performance to prospective e-buyers. The rapid growth in e-business environment has led to an increasing trend to host products and related analytics for prospective consumers in a manner that is informative, credible, and fair. Surprisingly, there is not much prior research done in making the ORV productive and informative. In this paper, we formulate a deep-time resolved view of buyer sentiment analysis called as DBRV (Deep-Bin Rating-Vote Visualization) based on equi-length time slices of ratings-votes. We also propose a novel performance measure called Popularity index ( \(\mathcal{P}\) ) which combines ratings-votes meaningfully in time differential manner to depict discriminant power of the DBRV. Simulations performed using real data of ratings-votes for iPhone 12 and Earbuds with 4139 and 8497 reviews respectively, shows that DBRV offers more informative insights about consumer experience than average ratings-votes. The proposed DBRV-1 depicts time-binning of ratings-votes distribution thereby illustrating its information resolved into time-bins. Likewise, DBRV-2 powered by \(\mathcal{P}\) conveys higher intelligence about perceived product performance than existing ORV. Finally, we also formulate a more effective and novel user-interactive dashboard, shown for iPhone 12 and Earbuds, that offers a promising frontend analytics for prospective customers.