English  |  正體中文  |  简体中文  |  Items with full text/Total items : 49378/84106 (59%)
Visitors : 7383109      Online Users : 74
RC Version 7.0 © Powered By DSPACE, MIT. Enhanced by NTU Library & TKU 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://tkuir.lib.tku.edu.tw:8080/dspace/handle/987654321/110957


    Title: 以圖像資料預測美食圖片點擊數
    Other Titles: Using image features to forecast clicks of gourmet photos
    Authors: 周雅愉;Chou, Ya-Yu
    Contributors: 淡江大學管理科學學系碩士班
    陳怡妃;Chen, I-Fei
    Keywords: 點擊數預測;圖片特徵;多元適應性雲形迴歸;線性迴歸;click prediction;image features;Multivariate adaptive regression splines;Linear Regression
    Date: 2016
    Issue Date: 2017-08-24 23:40:17 (UTC+8)
    Abstract: 網際網路在近十年來趨於成熟,加上智慧型裝置普及,使大眾改變原本的購物習慣,且參考部落客意見消費的人數趨漲,故如何使自己經營之部落格有更多追蹤者即為一新的營利指標,而在智慧型裝置中,圖片為最先吸引觀看者注意的關鍵。
    在本文中將針對圖片特徵探討其與點擊數之間的關係,分別以線性迴歸模型及多元適應性雲形迴歸模型(MARS)作為研究方法,再以RMSE、MAPE、MAD三項誤差指標檢視此兩種方法之預測能力,結果顯示以MARS模型在圖片的點擊數預測上有較好的預測能力,且針對不同類別的食品有不同之影響其點擊數的重要變數,其結果可提供往後業者在這些類別產品於網路上架時,作為參考依據。
    The Internet has been growing extremely fast in the last decade, coupled with the popularity of smart devices has a huge impact on the shopping habits of the general public. Moreover, the amount of people who make decisions of purchasing based on bloggers’ opinions is gradually increasing. Thus, it has become a new beneficial index that how to manage the blogs with more and more followers. Among the smart devices, the product picture is a crucial key that will raise the buyers’ awareness.
    This paper will be exploring the relationship between image features and web hits. Using linear regression model and multivariate adaptive regression splines model (MARS) as the research methods, and RMSE, MAPE, MAD to evaluate the error index of the predictions. The results show that the MARS model has better click predictions, and has different impacts on clicks according to the different variables. The results may be used for future industries reference.
    Appears in Collections:[管理科學學系暨研究所] 學位論文

    Files in This Item:

    File Description SizeFormat
    index.html0KbHTML2View/Open

    All items in 機構典藏 are protected by copyright, with all rights reserved.


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