Abstract:

In the era of big data, the opinions, tendencies and arguments of users on the forum have formed a large amount of data. Sentiment mining for forum will help relevant personnel understand and control the information, and will also have a direct impact on decision-making. So it is necessary to pay attention to the sentiment mining of forums. First, starting from the concept and meaning of forum data mining related technologies, it focuses on the research status of forum sentiment mining based on sentiment lexicon and machine learning. Then comparing the shortcomings and applications of each method, it gives the improvement plan. Finally, the problems and challenges to be solved in the field of sentiment mining in the forum are given, and the future development direction of the technology is predicted.

Key words: forum, sentiment mining, sentiment lexicon, machine learning

摘要：

大数据时代，论坛上用户的看法、倾向、观点和争论形成了大量数据。对这些能表达作者情绪的数据进行挖掘，有助于相关人员对信息的理解、把控，亦会对决策形成直接影响。为此，关注论坛情感挖掘十分重要。从论坛数据挖掘相关技术的概念和意义出发，重点讨论了论坛情感挖掘中基于情感词典和基于机器学习两种方法的研究现状，对每种方法的适用任务、不足之处、改进方案、发展趋势等进行对比和阐述。给出论坛情感挖掘领域尚待解决的难题与挑战，并对该技术未来的发展方向做出预测。

关键词: 论坛, 情感挖掘, 情感词典, 机器学习