中文摘要：

运用一套紧跟前沿领域的微博情感倾向判定算法对微博主题中的用户情感倾向进行判定与分析.首先,对该算法进行介绍;其次,运用真实的微博数据测试样本对算法进行测试,评估该算法对微博分析的准确度,表明该算法在微博用户情感倾向判定方面具有较高的预测精度;然后,分别对以中国两会和日本地震为主题的新浪微博用户的情感倾向进行判定与分析,每个主题的统计结果符合现实社会及历史因素的常规逻辑分析,验证了该算法的可靠性和应用性;最后,通过只统计每个主题一半用户的情感倾向,得到与统计全部用户情感倾向基本一致.这说明对于具有超大用户数量参与的微博主题大数据,只需要统计一定比例用户的情感倾向,可以得到全部用户情感倾向的统计结果,因此可以大幅减少统计时间,且保证了结果有效性,对微博用户情感倾向的判定和分析具有重要理论意义和实际应用价值.

英文摘要：

We use a new algorithm to determine and analyze the sentiment orientation of Chinese micro-biog. Firsly, the algorithm is introduced in detail. Secondly, a test sample is used to evaluate the accuracy of the algorithm, which shows that our algorithm has good prediction accuracy in the determination of micro-blog sentiment orientation. Thirdly, we determine and analyze the sentiment orientation of sina micro-blog users with the two topics of the China two sessions and Japanese earthquake. The statistical results of each topic are in line with the logical analysis of the social and historical factors, which verify the reliability and applicability of our algorithm. Finally, we have further found that the statistical results of half micro-blog users are almost the same with the ones of total micro-blog users for each topic. This implies that the statistical results of total micro-blog users can be gotten only through analyzing a certain proportion of the big data. This is significant for theoretical research and practical application in the determination and analysis of micro-blog sentiment orientation.