English  |  正體中文  |  简体中文  |  全文筆數/總筆數 : 18278/19583 (93%)
造訪人次 : 1023413      線上人數 : 436
RC Version 7.0 © Powered By DSPACE, MIT. Enhanced by NTU Library IR team.
搜尋範圍 查詢小技巧:
  • 您可在西文檢索詞彙前後加上"雙引號",以獲取較精準的檢索結果
  • 若欲以作者姓名搜尋,建議至進階搜尋限定作者欄位,可獲得較完整資料
  • 進階搜尋
    請使用永久網址來引用或連結此文件: http://nhuir.nhu.edu.tw/handle/987654321/28995


    題名: 運用文字探勘探討美容議題-以PTT美容版為例
    其他題名: Applying Text Mining Techniques to Salon Issues on PTT Beauty Salon
    作者: 楊嬿庭
    YANG, YEN-TING
    貢獻者: 資訊管理學系
    洪銘建
    HUNG, MING-CHIEN
    關鍵詞: 網路爬蟲;文字探勘;主題模型
    Web Crawler;Text Mining;LDA
    日期: 2021
    上傳時間: 2022-08-23 15:25:31 (UTC+8)
    摘要:   科技網路的發達下,人與人之間的交流不再受時間和空間的限制,可以立即的透過網路分享資訊,讓資訊的傳遞更為快速,在新冠疫情影響之下群眾更加依賴網路帶來的生活便利,為了配合防疫措施在無法外出的情形下,民眾開始大量使用網路購買所需用品,但因為無法看到實體物品,消費者為了降低購買前的不確定感,因此形成消費者在購買前會先看網路產品評論的習慣。從PTT各大版中有數不清的產品評論,大量來自於消費者使用心得,潛在消費者會因為評論而影響原先的購買意願,因此本研究針對PTT美容版文章,利用文字探勘和爬蟲技術針對使用者對美妝產品的購後心得進行網路口碑分析。研究結果發現在PTT美容版中最常被提及的品牌為雅漾、蘭蔻、克蘭詩以及碧兒泉,並從各品牌之中以主題模型分析出關於品牌產品的「保濕」和「吸收」成效是消費者較為著重,以及在行銷上消費者會以紙本式「DM」當作購買前的參考指標,因此本研究建議品牌可以在產品目錄設計上更符合消費者的設計需求。
      With the intelligence of the technology network, the communication between people is no longer limited by time and space. Immediately share information through the Internet, and under the influence of the Covid-19, the people rely more on the convenience in life brought by the Internet. Cooperating with epidemic prevention measures, people began to use the Internet to buy supplies in large quantities under activities that were unable to go out. Due to being unable to see the products physically, consumers will read online product reviews before buying in order to reduce their sense of uncertainty. There are countless comments in the major editions of PTT, and a large number of them come from consumer experience. Potential consumers’ initial purchase intentions will be affected due to the influence of the reviews on the product. because of product reviews. In order to extract the most frequently mentioned keywords from a large amount of text, this research aims at PTT beauty articles, using text mining and Web crawler to explore users’ post-purchase experience of beauty products and conduct online word-of-mouth analysis. The results of the study found that the most frequently mentioned brands in the PTT beauty edition were “Avene”, “Lancome”, “Clarins” and “Biotherm”. By analyzing the products from each brand using the LDA model, the results found out that consumers pay more attention to the effects of “moisturizing” and “absorption”, and consumers will use the paper-based “DM” as a reference index before purchase in marketing. Therefore, this study suggests the brands product catalog designs can made based on the needs of the consumers.
    顯示於類別:[資訊管理學系] 博碩士論文

    文件中的檔案:

    檔案 描述 大小格式瀏覽次數
    110NHU00396002-001.pdf4462KbAdobe PDF350檢視/開啟
    index.html0KbHTML158檢視/開啟


    在NHUIR中所有的資料項目都受到原著作權保護.

    TAIR相關文章

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