English  |  正體中文  |  简体中文  |  Items with full text/Total items : 889/889 (100%)
Visitors : 14785520      Online Users : 35
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://ccur.lib.ccu.edu.tw/handle/A095B0000Q/351

    Title: 資料導向非平衡型分類於推薦系統之研究;A Data-Driven Imbalanced Classifying Method for Recommendation Systems
    Authors: 曾思婷;Tseng, Sih-Ting
    Contributors: 電機工程研究所
    Keywords: 重採樣;非平衡型資料;稀疏問題;推薦系統;Synthetic Minority Over-sampling Technique;Imbalanced dataset;Sparsity problem;Recommendation system
    Date: 2016
    Issue Date: 2019-07-17
    Publisher: 電機工程研究所
    Abstract: 推薦系統之主要目的是提供使用者可能喜好的項目或服務。但是要進行推薦時,因受限使用者歷史評價紀錄中,每位使用者曾經評價過的項目,實際僅占可選擇之全部項目的極少比例,導致使用者評價矩陣呈現稀疏狀態。當系統能用於分析的評價資料過少,就會使推薦的難度提升。也因為稀疏問題會有推薦成效不佳的影響,所以過去學者們也一直提出新的方法及架構,嘗試提升推薦準確度。本論文是以參考使用者的歷史評價紀錄做為推薦的依據,故採取評價計算的協同式過濾方法,而MovieLens 電影資料庫會做為我們的實驗資料。在改善推薦系統典型的稀疏問題,則是採用重採樣技術(SMOTE)做為應對方針,它主要是用於看待非不平衡型數據資料的一種處理方式,我們將它放在推薦系統處理評價資料的過程中,藉由增加虛擬評價樣本到訓練集合,並建立成預測模型來找出使用者感興趣之電影。最後,本論文所提供的方法可以提升電影推薦系統的預測準確率。
    The main goal of a recommendation system is to suggest services or items that users like. However, the system suffers from the sparsity problem if the percentage of user feedbacks is very small in the historical database. The sparsity problem will have the effect of poor recommendation since a small number of the contributed user ratings are too difficult to analyze. Therefore, in order to improve the accuracy of recommendations, many researchers have referred to different methods and frameworks. In this thesis, we try to incorporate an oversampling method to increase the number of useful data. The MovieLens database is used as the experimental data in our study, and we add the resampling technique, the Synthetic Minority Over-sampling Technique (SMOTE), which is a way for processing imbalanced classification problem in our recommendation system. While training dataset is preprocessed by the sampling method, the new training dataset will be built as a prediction model to find movies that the user is interested in. Finally, we show in our experiments that the proposed method in this thesis can improve the prediction accuracy of the movie recommendation system.
    Appears in Collections:[電機工程研究所] 學位論文

    Files in This Item:

    File Description SizeFormat

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

    版權聲明 © 國立中正大學圖書館網頁內容著作權屬國立中正大學圖書館


    DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library IR team Copyright ©   - Feedback