影视大数据的处理及其应用研究
首发时间:2015-12-11
摘要:随着影视行业信息化系统的不断完善,行业中所拥有的信息量越来越庞大。对海量影视数据进行分析从而获取其中有价值的信息,为影视相关人员提供决策方向是影视大数据研究当中的热点。运用机器学习的相关技术对影视大数据进行处理分析,通过降维方法和预测方法对影视数据进行挖掘和探讨是研究人员所常用的方法。这些方法的性能往往取决于降维及预测方法的准确性和适用性。本文提出了一个基于因子分析的影视大数据降维方法,并使用支持向量回归机对电视剧收视率进行预测。实验结果表明该方法可以有效的对影视数据进行收视率的预测,从而为制片商和放映商对影视收视率进行更好的预测提供了决策支持。
For information in English, please click here
The Processing And Application of Big Data for the Film
Abstract:With the constant improvement of the film information technology system, the system increasingly has a large amount of information, which can be called big data of film. The hotspot of film big data research is to gain the valuable information via analyzing those data, which can provide the policy direction for the film and television workers. The researchers commonly utilized the machine learning techniques to process and analysis film big data via dimension reduction and forecasting methods. The performance of those algorithms are related to the effectiveness and applicability of machine learning techniques in big data research. In this paper, a factor analysis-based dimension reduction method is proposed for the film big data, and the support vector regressionis utilized to predict the audience ratings. Experimental results show that this method can effectively predict audience ratings, which means that the producers and exhibitors can be provided the decision support with a more accurate forecast.
Keywords: Film Data Machine Learning Dimension Reduction Prediction Method
基金:
论文图表:
引用

No.4667279111122914****
同行评议
共计0人参与
勘误表
影视大数据的处理及其应用研究
评论
全部评论