English  |  正體中文  |  简体中文  |  Items with full text/Total items : 78852/78852 (100%)
Visitors : 35335800      Online Users : 516
RC Version 7.0 © Powered By DSPACE, MIT. Enhanced by NTU Library IR team.
Scope Tips:
  • please add "double quotation mark" for query phrases to get precise results
  • please goto advance search for comprehansive author search
  • Adv. Search
    HomeLoginUploadHelpAboutAdminister Goto mobile version

    Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/92678

    Title: 利用圖卷積網路的時間感知客戶購買意圖預測;Time-Aware Customer Purchase Intention Prediction with Graph Convolutional Networks
    Authors: 周亭妏;Chou, Ting-Wen
    Contributors: 資訊管理學系
    Keywords: 序列推薦;深度學習;神經網路;圖卷積網路;Transformer;Sequential Recommendation;Deep Learning;Neural Network;Graph Convolutional Networks;Transformer
    Date: 2023-08-17
    Issue Date: 2023-10-04 16:08:09 (UTC+8)
    Publisher: 國立中央大學
    Abstract: 深度學習模型被廣泛應用於處理序列資料,例如用戶購買歷史,以預測可能的購買項目。這些模型可以考慮複雜的非線性和動態關係,並且能夠捕捉用戶的長期序列行為。然而,對企業而言,現有的深度學習推薦模型對企業來說是不夠直觀且方便的;他們通常只是希望知道用戶是否會購買某項商品以及何時會購買。
    本文是第一個提出決策模型以回答商家關心的資訊的研究:在特定時間內,使用者是否會購買某個商品。該模型基於消費者過去按時間順序購買商品的序列資料,以及商品被消費者依序購買的序列資料,採用圖形神經網路來捕捉使用者和商品之間的高階相關性,使用Transformer來處理序列間的依賴關係,並使用GRU來處理序列中不同時間元素之間的動態關係。相較於過去的序列推薦模型,此模式可以讓商家了解對於它所關注的重點商品和客戶,是否有可能產生有效的購買。;Deep learning models have been widely applied to handle sequential data, such as customer purchase history, to predict potential purchases. These models can consider complex non-linear and dynamic relationships and capture users′ long-term sequential behaviors. However, for businesses, existing deep learning recommendation models may not be intuitive and convenient enough. They often simply want to know whether a user will make a purchase on specific items and when it is likely to occur.
    This paper presents the first decision model to address the information of interest to businesses: answering whether a user will purchase a specific item within a given time frame. The model is based on sequential data of users′ past purchases and the sequential data of items’ past purchases by users. It utilizes graph convolutional networks to capture high-order correlations between users and items, uses Transformers to handle dependencies among sequences, and utilizes GRU (Gated Recurrent Unit) to capture dynamic relationships between different time elements in the sequences. Compared to previous sequential recommendation models, this approach allows businesses to gain insights into the potential effectiveness of purchases for their key items and customers of interest.
    Appears in Collections:[資訊管理研究所] 博碩士論文

    Files in This Item:

    File Description SizeFormat

    All items in NCUIR are protected by copyright, with all rights reserved.

    社群 sharing

    ::: Copyright National Central University. | 國立中央大學圖書館版權所有 | 收藏本站 | 設為首頁 | 最佳瀏覽畫面: 1024*768 | 建站日期:8-24-2009 :::
    DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library IR team Copyright ©   - 隱私權政策聲明