CSA  >> Vol. 7 No. 8 (August 2017)

    基于用户评论的手机特征挖掘应用研究
    Research and Mine on Mobile Phone Feature Based on User Reviews

  • 全文下载: PDF(753KB) HTML   XML   PP.738-746   DOI: 10.12677/CSA.2017.78085  
  • 下载量: 113  浏览量: 183   国家自然科学基金支持

作者:  

高 威,傅湘玲:北京邮电大学,北京

关键词:
数据挖掘特征提取短语匹配能量模型Data Mining Feature Extraction Phrase Match Energy Model

摘要:

用户在线评论逐渐成为互联网环境下企业获取用户需求的重要数据资源。然而如何能够准确有效的在浩如烟海的评论中提炼出产品的特征及其对特征的描述,是一个理论和实践界的难题。本文对手机评论数据的预处理和特征提取的基础上,引入了LinLog能量模型,对产品特征及其描述信息进行聚类分析,从而获得该特征的准确评价。本文对采集自京东商城的4款手机评论应用能量模型,通过分析聚类结果,最终获得了这四款手机特征的优劣,结果经过验证,表明该方法能够直观有效的从用户产生内容中提炼出产品特征的优劣。

User online comment is becoming the important data resource for enterprise to get user's re-quirement in Internet environment. However, how to accurately and effectively extract the char-acteristics of the product and its description of the characteristics in the vast sea of comments is a difficult problem of the theory and practice. Based on the data preprocessing and feature extraction, the LinLog energy model is introduced to cluster and analyze the features of products and their descriptive information, so as to obtain the accurate evaluation of the feature. This paper applies the energy model to the evaluation of 4 mobile phones collected from Jingdong Mall, and then analyzes the clustering results, finally obtains the advantages and disadvantages of the four mobile phones. The results show that this method can extract the features of the product intuitively and effectively from the content generated by user.

文章引用:
高威, 傅湘玲. 基于用户评论的手机特征挖掘应用研究[J]. 计算机科学与应用, 2017, 7(8): 738-746. https://doi.org/10.12677/CSA.2017.78085

参考文献

[1] Berger, J., Sorensen, A.T. and Rasmussen. S.J. (2010) Positive Effects of Negative Publicity: When Negative Reviews Increase Sales. Marketing Science, 29, 815-827.
https://doi.org/10.1287/mksc.1090.0557
[2] 李升林, 乌兰木其. 基于数据挖掘的需求分析研究[J]. 中国机械工程, 2003, 14(5): 392-395.
[3] Vignoli, F., Gulik, R.V. and Wetering, H.V.D. (2004) Mapping Music In The Palm Of Your Hand, Explore and Discover Your Collection. International Conference on Music Information Retrieval.
[4] 石晶, 李万龙. 基于LDA模型的主题词抽取方法[J]. 计算机工程, 2010, 36(19): 81-83.
[5] 黄德才, 戚华春. PageRank算法研究[J]. 计算机工程, 2006, 32(4): 145-146.
[6] Kim, S.M. and Hovy, E. (2005) Automatic Detection of Opinion Bearing Words and Sentences. Proceedings of Ijcnlp.
[7] Noack, A. (2007) Energy Models for Graph Clustering. Journal of Graph Algorithms & Applications, 11, 453-480.
https://doi.org/10.7155/jgaa.00154