完整後設資料紀錄
DC 欄位語言
dc.contributor.author郭佩菁en_US
dc.contributor.authorKuo, Pei-Chingen_US
dc.contributor.author劉敦仁en_US
dc.contributor.authorLiu, Duen-Renen_US
dc.date.accessioned2014-12-12T02:36:08Z-
dc.date.available2014-12-12T02:36:08Z-
dc.date.issued2012en_US
dc.identifier.urihttp://140.113.39.130/cdrfb3/record/nctu/#GT070063406en_US
dc.identifier.urihttp://hdl.handle.net/11536/72827-
dc.description.abstract預測是銷售其中的一個重要環節,準確的預測不儘能降低存貨成本,還能提升利潤,因此如何準確的預測則成為銷售業者重要的課題。隨著網際網路的普遍化,現今人們的活動訊息也被留存在網路上,如此龐大且繁雜的資訊,透過分析可發掘有助於銷售預測的有用資訊。 吸引消費者購買雜誌,主要為其內容引起消費者興趣,通常也是由於雜誌文章中的標題字高度引人注意,而在網際網路上,則可表示為熱門的搜尋。本研究將以雜誌銷售為探討,將流通於網際網路上的資訊,結合雜誌本身,以二者來做為其銷售預測。 本研究運用迴歸分析、分類迴歸樹、類神經網路模型以加入雜誌標題熱門指數的方法進行雜誌銷售預測。實驗結果顯示,銷售預測加入使用標題熱門指數比基礎使用過往銷售量做預測更能改善預測的準確性。zh_TW
dc.description.abstractSales prediction is a vital issue for publish industry, so increasing the accuracy of the prediction can not only reduce the cost of inventory but also raise the profits. Nowadays more and more daily activities, including shopping and reading, occur online and the related information is also stored online. Process this mass information is difficult but may discovers useful information for sales prediction. Since the titles of articles and popular topics on the Internet are triggers for readers’ magazine purchasing, this work focuses on predicting the sales of magazines and proposes prediction models combing the information extracted from the Internet and the article titles of magazines. Regression, decision tree, and artificial neural network are applied in this study. The experimental results show that our proposed model can improve the accuracy of sales prediction.en_US
dc.language.isozh_TWen_US
dc.subject銷售量預測zh_TW
dc.subject標題熱門zh_TW
dc.subjectGoogle 搜尋趨勢zh_TW
dc.subject迴歸zh_TW
dc.subject分類迴歸樹zh_TW
dc.subject類神經網路zh_TW
dc.subjectForecasten_US
dc.subjectPopularity indexen_US
dc.subjectGoogle Trenden_US
dc.subjectRegressionen_US
dc.subjectCARTen_US
dc.subjectArtificial Neural Networken_US
dc.title考量標題熱門程度之雜誌銷售量預測方法比較zh_TW
dc.titleComparisons of Forecasting Methods for Predicting Magazine Sales based on Title Popularityen_US
dc.typeThesisen_US
dc.contributor.department管理學院資訊管理學程zh_TW
顯示於類別:畢業論文