Purpose: This paper intends to explore a quantitative method for investigating the characteristics of information diffusion through social media like weblogs and microblogs.By using the social network analysis methods...Purpose: This paper intends to explore a quantitative method for investigating the characteristics of information diffusion through social media like weblogs and microblogs.By using the social network analysis methods,we attempt to analyze the different characteristics of information diffusion in weblogs and microblogs as well as the possible reasons of these differences.Design/methodology/approach: Using the social network analysis methods,this paper carries out an empirical study by taking the Chinese weblogs and microblogs in the field of Library and Information Science(LIS) as the research sample and employing measures such as network density,core/peripheral structure and centrality.Findings: Firstly,both bloggers and microbloggers maintain weak ties,and both of their social networks display a small-world effect. Secondly,compared with weblog users,microblog users are more interconnected,more equal and more capable of developing relationships with people outside their own social networks. Thirdly,the microblogging social network is more conducive to information diffusion than the blogging network,because of their differences in functions and the information flow mechanism. Finally,the communication mode emerged with microblogging,with the characteristics of micro-content,multi-channel information dissemination,dense and decentralized social network and content aggregation,will be one of the trends in the development of the information exchange platform in the future.Research limitations: The sample size needs to be increased so that samples are more representative. Errors may exist during the data collection. Moreover,the individual-level characteristics of the samples as well as the types of information exchanged need to be further studied.Practical implications: This preliminary study explores the characteristics of information diffusion in the network environment and verifies the feasibility of conducting a quantitative analysis of information diffusion through social media. In addition,it provides insight into the characteristics of information diffusion in weblogs and microblogs and the possible reasons of these differences.Originality/value: We have analyzed the characteristics of information diffusion in weblogs and microblogs by using the social network analysis methods. This research will be useful for a quantitative analysis of the underlying mechanisms of information flow through social media in the network environment.展开更多
This paper explores the uses’ influences on microblog. At first, according to the social network theory, we present an analysis of information transmitting network structure based on the relationship of following and...This paper explores the uses’ influences on microblog. At first, according to the social network theory, we present an analysis of information transmitting network structure based on the relationship of following and followed phenomenon of microblog users. Informed by the microblog user behavior analysis, the paper also addresses a model for calculating weights of users’ influence. It proposes a U-R model, using which we can evaluate users’ influence based on PageRank algorithms and analyzes user behaviors. In the U-R model, the effect of user behaviors is explored and PageRank is applied to evaluate the importance and the influence of every user in a microblog network by repeatedly iterating their own U-R value. The users’ influences in a microblog network can be ranked by the U-R value. Finally, the validity of U-R model is proved with a real-life numerical example.展开更多
Considering that there exists a strong similarity between behaviors of users and intelligence of swarm of agents,in this paper we propose a novel user recommendation strategy based on particle swarm optimization(PSO)f...Considering that there exists a strong similarity between behaviors of users and intelligence of swarm of agents,in this paper we propose a novel user recommendation strategy based on particle swarm optimization(PSO)for Microblog network. Specifically,a PSO-based algorithm is developed to learn the user influence,where not only the number of followers is incorporated,but also the interactions among users(e.g.,forwarding and commenting on other users' tweets). Three social factors,the influence and the activity of the target user,together with the coherence between users,are fused to improve the performance of proposed recommendation strategy. Experimental results show that,compared to the well-known Page Rank-based algorithm,the proposed strategy performs much better in terms of precision and recall and it can effectively avoid a biased result caused by celebrity effect and zombie fans effect.展开更多
Microblog is a new Internet featured product, which has seen a rapid development in recent years. Researchers from different countries are making various technical analyses on microblogging applications. In this study...Microblog is a new Internet featured product, which has seen a rapid development in recent years. Researchers from different countries are making various technical analyses on microblogging applications. In this study, through using the natural language processing(NLP) and data mining, we analyzed the information content transmitted via a microblog, users' social networks and their interactions, and carried out an empirical analysis on the dissemination process of one particular piece of information via Sina Weibo.Based on the result of these analyses, we attempt to develop a better understanding about the rule and mechanism of the informal information flow in microblogging.展开更多
情感分析能从用户言论中快速准确地挖掘用户的情感倾向,有着极大的应用市场。针对微博语言语法结构复杂多样的特性,提出了一种基于语法依存结构的图卷积神经网络(SGCN)模型对中文微博进行细粒度的情感分类。所提模型兼具结构表达和语义...情感分析能从用户言论中快速准确地挖掘用户的情感倾向,有着极大的应用市场。针对微博语言语法结构复杂多样的特性,提出了一种基于语法依存结构的图卷积神经网络(SGCN)模型对中文微博进行细粒度的情感分类。所提模型兼具结构表达和语义表达丰富的特点:基于词语间的依赖关系构建文本图,并通过点互信息(PMI)量化词语间的相关程度,作为相应边的权重以充分表现句子的结构信息;将融合位置信息的语义特征作为节点的初始特征,增加文本图中点的语义特征。为了验证所提模型的性能,在SMP2020(Social Media Processing 2020)微博情感分类数据集上,对两组包含开心、悲伤、愤怒、恐惧、惊讶和无情绪的6类微博情感数据进行了分析。实验结果表明,所提模型的平均F1分数可达到72.64%,相较于BERT(Bidirectional Encoder Representations from Transformers)词向量特征图卷积网络(BGCN)模型和文本级图神经网络(Text-Level-GNN)模型分别提高了2.75和3.87个百分点,验证了所提模型能更有效地利用句子的结构信息,提升模型的分类性能。展开更多
基金supported by Sun Yat-sen University Cultivation Fund for Young Teachers(Grant No.:20000-3161102)the National Social Science Fundation of China(Grant No.:08CTQ015)
文摘Purpose: This paper intends to explore a quantitative method for investigating the characteristics of information diffusion through social media like weblogs and microblogs.By using the social network analysis methods,we attempt to analyze the different characteristics of information diffusion in weblogs and microblogs as well as the possible reasons of these differences.Design/methodology/approach: Using the social network analysis methods,this paper carries out an empirical study by taking the Chinese weblogs and microblogs in the field of Library and Information Science(LIS) as the research sample and employing measures such as network density,core/peripheral structure and centrality.Findings: Firstly,both bloggers and microbloggers maintain weak ties,and both of their social networks display a small-world effect. Secondly,compared with weblog users,microblog users are more interconnected,more equal and more capable of developing relationships with people outside their own social networks. Thirdly,the microblogging social network is more conducive to information diffusion than the blogging network,because of their differences in functions and the information flow mechanism. Finally,the communication mode emerged with microblogging,with the characteristics of micro-content,multi-channel information dissemination,dense and decentralized social network and content aggregation,will be one of the trends in the development of the information exchange platform in the future.Research limitations: The sample size needs to be increased so that samples are more representative. Errors may exist during the data collection. Moreover,the individual-level characteristics of the samples as well as the types of information exchanged need to be further studied.Practical implications: This preliminary study explores the characteristics of information diffusion in the network environment and verifies the feasibility of conducting a quantitative analysis of information diffusion through social media. In addition,it provides insight into the characteristics of information diffusion in weblogs and microblogs and the possible reasons of these differences.Originality/value: We have analyzed the characteristics of information diffusion in weblogs and microblogs by using the social network analysis methods. This research will be useful for a quantitative analysis of the underlying mechanisms of information flow through social media in the network environment.
文摘This paper explores the uses’ influences on microblog. At first, according to the social network theory, we present an analysis of information transmitting network structure based on the relationship of following and followed phenomenon of microblog users. Informed by the microblog user behavior analysis, the paper also addresses a model for calculating weights of users’ influence. It proposes a U-R model, using which we can evaluate users’ influence based on PageRank algorithms and analyzes user behaviors. In the U-R model, the effect of user behaviors is explored and PageRank is applied to evaluate the importance and the influence of every user in a microblog network by repeatedly iterating their own U-R value. The users’ influences in a microblog network can be ranked by the U-R value. Finally, the validity of U-R model is proved with a real-life numerical example.
基金supported by National Natural Science Foundation of China(No.61171109)Applied Basic Research Programs of Sichuan Science and Technology Department(No.2014JY0215)Basic Research Plan in SWUST(No.13zx9101)
文摘Considering that there exists a strong similarity between behaviors of users and intelligence of swarm of agents,in this paper we propose a novel user recommendation strategy based on particle swarm optimization(PSO)for Microblog network. Specifically,a PSO-based algorithm is developed to learn the user influence,where not only the number of followers is incorporated,but also the interactions among users(e.g.,forwarding and commenting on other users' tweets). Three social factors,the influence and the activity of the target user,together with the coherence between users,are fused to improve the performance of proposed recommendation strategy. Experimental results show that,compared to the well-known Page Rank-based algorithm,the proposed strategy performs much better in terms of precision and recall and it can effectively avoid a biased result caused by celebrity effect and zombie fans effect.
文摘Microblog is a new Internet featured product, which has seen a rapid development in recent years. Researchers from different countries are making various technical analyses on microblogging applications. In this study, through using the natural language processing(NLP) and data mining, we analyzed the information content transmitted via a microblog, users' social networks and their interactions, and carried out an empirical analysis on the dissemination process of one particular piece of information via Sina Weibo.Based on the result of these analyses, we attempt to develop a better understanding about the rule and mechanism of the informal information flow in microblogging.
文摘情感分析能从用户言论中快速准确地挖掘用户的情感倾向,有着极大的应用市场。针对微博语言语法结构复杂多样的特性,提出了一种基于语法依存结构的图卷积神经网络(SGCN)模型对中文微博进行细粒度的情感分类。所提模型兼具结构表达和语义表达丰富的特点:基于词语间的依赖关系构建文本图,并通过点互信息(PMI)量化词语间的相关程度,作为相应边的权重以充分表现句子的结构信息;将融合位置信息的语义特征作为节点的初始特征,增加文本图中点的语义特征。为了验证所提模型的性能,在SMP2020(Social Media Processing 2020)微博情感分类数据集上,对两组包含开心、悲伤、愤怒、恐惧、惊讶和无情绪的6类微博情感数据进行了分析。实验结果表明,所提模型的平均F1分数可达到72.64%,相较于BERT(Bidirectional Encoder Representations from Transformers)词向量特征图卷积网络(BGCN)模型和文本级图神经网络(Text-Level-GNN)模型分别提高了2.75和3.87个百分点,验证了所提模型能更有效地利用句子的结构信息,提升模型的分类性能。
