Title: Research and development of community's opinion discovery and influence analysis system in social network

Authors: Chunlin Yin; Jie Li; Pengfeng Qiu; Zheng Yang; Li Yang; Meng Su; Na Zhao

Addresses: Electric Power Research Institute of Yunnan Power Grid Co., Ltd., Kunming Economic and Technological Development Zone, No. 105, Yunda West Road, Yunnan Province, China ' Electric Power Research Institute of Yunnan Power Grid Co., Ltd., Kunming Economic and Technological Development Zone, No. 105, Yunda West Road, Yunnan Province, China ' Electric Power Research Institute of Yunnan Power Grid Co., Ltd., Kunming Economic and Technological Development Zone, No. 105, Yunda West Road, Yunnan Province, China ' Electric Power Research Institute of Yunnan Power Grid Co., Ltd., Kunming Economic and Technological Development Zone, No. 105, Yunda West Road, Yunnan Province, China ' Electric Power Research Institute of Yunnan Power Grid Co., Ltd., Kunming Economic and Technological Development Zone, No. 105, Yunda West Road, Yunnan Province, China ' Electric Power Research Institute of Yunnan Power Grid Co., Ltd., Kunming Economic and Technological Development Zone, No. 105, Yunda West Road, Yunnan Province, China ' Yunnan University, University East Outer Ring South Road, Chenggong District, Kunming City, China

Abstract: In the current internet era, the detection of hidden topics in social media and the discovery of key leaders of public opinion can monitor and guide the dissemination of public opinion on hot events, which is conducive to maintaining social stability. However, the topics learned by the existing topic discovery methods are sometimes ambiguous and need complicated follow-up processing. Therefore, this paper provides some solutions to these problems by introducing community network structure and analysing the connectivity between users. Based on the community discovery algorithm and topic model, a system for discovering and analysing the influence of opinion communities in social networks is developed. The software system carries out topic detection on the basis of community structure division. Combined with the text emotion analysis model, the system also analyses the social emotion trend of each community to hot events. The experimental results on microblog data verify the superiority of the algorithm and the system in topic detection and the effectiveness of mining social emotions.

Keywords: social network; community structure; topic model; LDA algorithm; sentiment analysis.

DOI: 10.1504/IJICT.2024.140502

International Journal of Information and Communication Technology, 2024 Vol.25 No.3, pp.281 - 302

Received: 23 Mar 2022
Accepted: 24 Jun 2022
Published online: 20 Aug 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article