根據電影屬性預測所獲評分之研究

NCUIR > School of Management at National Central University > Graduate Institute of Business Administration > Electronic Thesis & Dissertation > Item 987654321/64428

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/64428

Title:	根據電影屬性預測所獲評分之研究
Authors:	謝翔安;Xie,Xiang-An
Contributors:	企業管理學系
Keywords:	電影評價預測;多元線性迴歸分析;凸組合;類神經網路;forecast film evaluation;multiple linear regression;convex combination;artificial neural network
Date:	2014-06-27
Issue Date:	2014-08-11 18:19:30 (UTC+8)
Publisher:	國立中央大學
Abstract:	由於電影工業的蓬勃發展，劇院聲光效果持續不斷進步，電影已是現代人不可或缺的重要休閒娛樂方式。但製作一部商業電影需要花費高額成本及面臨高風險，有許多電影上映後票房收入遠低於成本而造成鉅額虧損，是否可以賺取利潤儼然成為電影產業的重要考量因素。但由於電影上映之前很難評估該電影的品質及觀眾的接受程度，而電影製作公司在電影拍攝前就必須決定成本及各方面考量的情況下，分析及預測未上映的電影變得格外重要。本研究透過蒐集過去十年電影資料，包含電影各種屬性及觀眾評價，使用多元線性迴歸分析(multiple linear regression analysis)、凸組合(convex combination)以及類神經網路(artificial neural network)預測未上映電影觀眾評價，讓投資者及電影製作公司在拍攝電影前作為決策的參考依據。;Due to the rise of the film industry, movies have been an essential and important recreation to human beings. But making a commercial film need high production costs and face high risks, the gross of many movies are usually far less than the costs, which causes huge losses. Therefore, the engagement of a forecast in box office income and cost is deemed to be an important issue to many scholars and the industry members. However, it’s difficult to evaluate the quality of the movie and acceptability of moviegoers before the movie is released, the film production company must determine the cost and various aspects before filming. This research collected movie data from over the past decade, which includes various attributes of movies and moviegoer’s evaluations, and uses multiple linear regression analysis method, convex combination and artificial neural network method to forecast unreleased moviegoer’s evaluations, so that investors and film production companies can have a reference basis whilst making decisions before filming.
Appears in Collections:	[Graduate Institute of Business Administration] Electronic Thesis & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	487	View/Open

社群 sharing

Loading...