Asia Service Sciences and Software Engineering Conference最新文献

筛选

英文中文

Simulation Model Of Stampede Accident Based On Multi-Agent And Game Theory 基于多智能体和博弈论的踩踏事故仿真模型

Asia Service Sciences and Software Engineering Conference Pub Date : 1900-01-01 DOI: 10.1145/3456126.3456136

Sui Yang, Baoyun Wang, Linna Li

引用次数: 1

Call Attention to Stances: Detect Rumor with a Stance Attention Network 注意姿态:用姿态注意网络检测谣言

Asia Service Sciences and Software Engineering Conference Pub Date : 1900-01-01 DOI: 10.1145/3399871.3399890

Lingyu Zeng, Bin Wu, Bai Wang

{"title":"Call Attention to Stances: Detect Rumor with a Stance Attention Network","authors":"Lingyu Zeng, Bin Wu, Bai Wang","doi":"10.1145/3399871.3399890","DOIUrl":"https://doi.org/10.1145/3399871.3399890","url":null,"abstract":"With the advent of web2.0 era, social media such as microblogging platforms have become important information distribution channels. Considering the amount of users on microblogging platforms, the rumor spreading on microblogging platform could have a negative effect on individuals, groups and the whole society. Hence, automatic rumor detection method is required. Stance features are a crucial for rumor detection, because users discussing rumors tend to express more querying and denying stances. However, different user stances have different importance. Motivated by this inspiration, in this paper, we propose a Rumor Detection Model with a Stance Attention Network (RDM-SAN). The RDM-SAN consists of three modules: stance module, rumor module and integrate module. The first module is a stance module, in which recurrent neural network and convolutional neural network are used to capture grammatical and event-related information released by individual users respectively. Then attention mechanism is leveraged to make sure that the most valuable information is contained in the event stance latent representation. The second module is a rumor module in which the content features and temporal features of a microblogging event are captured. The third module is an integrate module in which stance representations and rumor representations are concatenated together to detect rumors. Experiments on a real-world dataset from Weibo platform demonstrate that our proposed model RDM-SAN improves the performance of rumor detection in terms of both efficiency and accuracy compared to other methods, and the accuracy of our model achieves 94.6%.","PeriodicalId":174808,"journal":{"name":"Asia Service Sciences and Software Engineering Conference","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123043458","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 1