[目的/意义]提出基于在线评论的网络社区信息可信度评价方法，为信息治理提供有效依据。[方法/过程]构建基于在线评论的网络社区信息可信度评价指标体系，利用改进AHP理论确定指标权重；利用LSTM模型对评论情感分类，采用改进的D-S证据理论模型融合情感分类数据作为指标量化计算方法。以知乎网络社区为例，从3个角度计算网络信息内容的可信度：经过筛选的具有可信观点评价的在线评论、所有在线评论、调查问卷。[结果/结论]实验结果表明，基于可信观点评论的可信度排序与基于调查问卷的可信度排序基本一致，说明利用在线评论对网络信息可信度进行评价具有一定的可行性。
[Purpose/significance] A social Q&A community information credibility evaluation method based on online reviews was proposed, aiming to provid an effective methods for information governing.[Method/process] The paper constructed the evaluation system of online community information reliability based on online reviews, and the indicator weight was determined by improved AHP theory. The LSTM model was used to classify the reviews emotion, and the improved D-S evidence theory model was used to fuse the emotion classification data. Taking ZhiHu Community as an example, the credibility of network information content was calculated from three perspectives:screened online reviews with credible opinion evaluation, all online reviews and questionnaires.[Result/conclusion] The experimental results showed that the ranking of the credibility value of this method was basically the same as that of the ranking obtained by the questionnaire. It showed that it is feasible to evaluate the credibility of the network information by online review.
[1] KIOUSIS S. Public trust or mistrust? perceptions of media credibility in the information age[J]. Mass communication & society, 2001, 4(4):381-403.
[2] LUCASSEN T, SCHRAAGEN J M. Factual accuracy and trust in information:the role of expertise[J].Journal of the American Society for Information Science and Technology, 2011,62(7):1232-1242
[3] WESTERMAN D, SPENCE P R, HEIDE B V D. Social media as information source:recency of updates and credibility of information[J].Journal of computer-mediated communication, 2014, 19(2):171-183.
[4] 高明霞, 陈福荣. 基于信息融合的中文微博可信度评估方法[J]. 计算机应用, 2016, 36(8):2071-2075.
[5] ALRUBAIAN M, AL-QURISHI M, AL-RAKHAMI M, et al. Reputation-based credibility analysis of twitter social network users[J]. Concurrency & computation practice & experience, 2017, 29(7):1-12.
[6] RIEH S Y, DANIELSON D R. Credibility:a multidisciplinary framework[J].Annual review of information science & technology, 2010, 41(1):307-364.
[7] WATHEN C N, BURKELL J. Believe it or not:factors influencing credibility on the Web[J]. Journal of the Association for Information Science & Technology, 2002, 53(2):134-144.
[8] LI R, SU A. Factors influencing information credibility on social media platforms:evidence from facebook pages[J]. Procedia computer science, 2015, 72:314-328.
[9] SARNA G, BHATIA M P S. Content based approach to find the credibility of user in social networks:an application of cyberbullying[J]. International journal of machine learning & cybernetics, 2015, 8(2):1-13.
[10] 李保珍, 王亚, 周可. 基于贝叶斯理论的社会化媒体网络信息内容可信度测度[J]. 数据分析与知识发现, 2017, 1(6):83-92.
[11] FOGG B J, SOOHOO C, DANIELSON D R, et al. How do users evaluate the credibility of Websites?:a study with over 2,500 participants[M]. New York:Association for Computing Machinery,2003.
[12] 沈旺,康霄普,王佳馨,等.用户视角下社会化问答社区信息可信度评价研究[J].图书情报工作,2018,62(17):104-111.
[13] PARK M, SAMPATHKUMAR H, LUO B, et al. Content-based assessment of the credibility of online healthcare information[C]//IEEE international conference on big data. USA:IEEE,2013:51-58.
[14] KAKOL M, NIELEK R, WIERZBICKI A. Understanding and predicting Web content credibility using the content cedibility corpus[J]. Information processing & management, 2017, 53(5):1043-1061.
[15] NAMIHIRA Y, SEGAWA N, IKEGAMI Y, et al. High precision credibility analysis of information on Twitter[C]//International conference on signal-image technology & internet-based systems. Kyoto:IEEE, 2013:909-915.
[16] 孟美任, 丁晟春. 在线中文商品评论可信度研究[J]. 现代图书情报技术, 2013, 29(9):60-66.
[17] 唐晓波,兰玉婷.基于特征本体的微博产品评论情感分析[J].图书情报工作,2016,60(16):121-127,136.
[18] 曹宇,王名扬,贺惠新.情感词典扩充的微博文本多元情感分类研究[J].情报杂志,2016,35(10):185-189.
[19] 何跃,赵书朋,何黎.基于情感知识和机器学习算法的组合微文情感倾向分类研究[J].情报杂志,2018,37(5):189-194.
[20] 王勤勤, 张玉红, 李培培,等. 基于word2vec的跨领域情感分类方法[J]. 计算机应用研究, 2018, 35(10):50-53.
[21] 张冬雯,杨鹏飞,许云峰.基于word2vec和SVMperf的中文评论情感分类研究[J].计算机科学,2016,43(S1):418-421,447.
[22] 何晶晶,蔡德胜,介飞,等.利用D-S证据理论进行特征融合的同义实体识别[J].计算机应用研究,2018,35(5):1429-1433.
[23] 杨风暴, 王肖霞. D-S证据理论的冲突证据合成方法[M]. 北京:国防工业出版社, 2010.
[24] 梁樑, 盛昭翰, 徐南荣. 一种改进的层次分析法[J]. 系统工程, 1989(3):5-7.
