Merchant Relative Dwell Time Ranking for Electric Vehicle Charger Merchant Site Using Semi-Supervised Learning

カテゴリ: 研究会(論文単位)

論文No: CMN24028

グループ名: 【C】電子・情報・システム部門 通信研究会

発行日: 2024/03/25

タイトル(英語): Merchant Relative Dwell Time Ranking for Electric Vehicle Charger Merchant Site Using Semi-Supervised Learning

著者名: Kumar Sheetal(Hitachi India Private Limited),Kumar Sharath (Hitachi India Private Limited),Sharma Ankit(Hitachi India Private Limited),Kumar Vinoth (Hitachi India Private Limited)

著者名(英語): Sheetal Kumar(Hitachi India Private Limited),Sharath Kumar(Hitachi India Private Limited),Ankit Sharma (Hitachi India Private Limited),Vinoth Kumar (Hitachi India Private Limited)

キーワード: Dwell time |EV Charging |site selection|Dwell time |EV Charging |site selection

要約(日本語): Dwell time is the amount of time spent by customer at a merchant location for making a purchase. It is required to assess customers’ behavior, shopping habits, merchant placement and to evaluate existing locations for value-added services like Electric Vehicles (EVs) charging stations. We propose a novel method to estimate dwell time at merchant locations by using payment transaction data to identify profitable locations for EV charging stations. We use a combination of supervised and unsupervised learning to train a set of classifiers and establish relative dwell time rank. The results indicate the method exhibited an accuracy of 74%.

要約(英語): Dwell time is the amount of time spent by customer at a merchant location for making a purchase. It is required to assess customers’ behavior, shopping habits, merchant placement and to evaluate existing locations for value-added services like Electric Vehicles (EVs) charging stations. We propose a novel method to estimate dwell time at merchant locations by using payment transaction data to identify profitable locations for EV charging stations. We use a combination of supervised and unsupervised learning to train a set of classifiers and establish relative dwell time rank. The results indicate the method exhibited an accuracy of 74%.

本誌: 2024年3月28日-2024年3月29日通信研究会

本誌掲載ページ: 61-66 p

原稿種別: 英語

PDFファイルサイズ: 1,296 Kバイト