基于情感分析的移动图书馆用户生成内容评价效果研究

Research on the Evaluation of Mobile Library User-generated Content Based on Sentiment Analysis

[目的／意义]通过对移动图书馆用户生成内容的情感分析，预测用户情感倾向对移动图书馆资源的评价效果，从而更好地实现移动图书馆资源推广和精准推荐服务。[方法／过程]基于情感分析提出移动图书馆用户生成内容评价效果分析过程，以获取的“掌阅图书馆”中15部年度畅销书籍的用户生成内容为研究样本，对数据进行预处理，在此基础上从领域词典构建、情感分类、评价效果3个过程入手进行分析。[结果／结论]数据分析结果表明，移动图书馆UGC用户情感倾向具有多元性和一致性，中性评价具有重要性，能够较为准确地预测移动图书馆资源的得分情况。将情感分析相关理论和方法引入移动图书馆UGC研究，能够为移动图书馆完善其服务措施、提高其服务质量提供参考意见。

[ Purpose/significance ] This study aims to make an analysis of the evaluation effect of the mobile library user-generated content （UGC）, which is beneficial to improve the recommendation and promotion of mobile library re- sources. [ Method/process ] An evaluation model is constructed based on the sentimental analysis. After the preprocess- ing of the UGC of the 16 best-selling books from Zhangyue library, data analysis is divided into three stages, including dictionary construction, sentiment classification and evaluation. [ Result/conclusion] The results show that the sentimen- tal tendencies of UGC in mobile libraries are pluralistic and consistent, and the neutral evaluation is important, as well as the final evaluation results is reasonable. In this paper, the theory of sentiment analysis is introduced into the research of mobile library UGC, which can provide reference for mobile library to improve its social service measures and improve its social service quality.