共查询到20条相似文献，搜索用时 46 毫秒
1.
Human collaborative relationship inference is a meaningful task for online social networks and is called link prediction in network science. Real-world networks contain multiple types of interacting components and can be modeled naturally as heterogeneous information networks (HINs). The current link prediction algorithms in HINs fail to effectively extract training samples from snapshots of HINs; moreover, they underutilise the differences between nodes and between meta-paths. Therefore, we propose a meta-circuit machine (MCM) that can learn and fuse node and meta-path features efficiently, and we use these features to inference the collaborative relationships in question-and-answer and bibliographic networks. We first utilise meta-circuit random walks to obtain training samples in which the basic idea is to perform biased meta-path random walks on the input and target network successively and then connect them. Then, a meta-circuit recurrent neural network (mcRNN) is designed for link prediction, which represents each node and meta-path by a dense vector and leverages an RNN to fuse the features of node sequences. Experiments on two real-world networks demonstrate the effectiveness of our framework. This study promotes the investigation of potential evolutionary mechanisms for collaborative relationships and offers practical guidance for designing more effective recommendation systems for online social networks. 相似文献
2.
在知识经济时代,学科知识的传播与扩散促进了学科的协同、交叉、融合、发展与创新。文章利用复杂网络算法对学科引证知识扩散时序演化网络进行动态链路预测分析,以期探索学科知识流动结构变化及演进态势,为学科知识管理及决策制定提供可资借鉴的理论和实践参考。文章以学科引证知识扩散时序演化网络结构信息为基础,采用10项基于局部信息的相似性指标分别对无权和加权知识扩散网络进行动态链路预测分析,并将各指标的预测性能进行了对比。最后,利用无权RA指标和加权AA指标对学科引证知识扩散态势进行了预测。研究表明:不同指标的预测精度在不同的时间段内会动态变化;在学科引证知识扩散网络中,存在一定程度的弱连接效应;不同链路预测指标在无权和加权学科引证知识扩散网络中的适用性存在一定差异。 相似文献
3.
《Information processing & management》2020,57(2):102151
Learning latent representations for users and points of interests (POIs) is an important task in location-based social networks (LBSN), which could largely benefit multiple location-based services, such as POI recommendation and social link prediction. Many contextual factors, like geographical influence, user social relationship and temporal information, are available in LBSN and would be useful for this task. However, incorporating all these contextual factors for user and POI representation learning in LBSN remains challenging, due to their heterogeneous nature. Although the encouraging performance of POI recommendation and social link prediction are delivered, most of the existing representation learning methods for LBSN incorporate only one or two of these contextual factors. In this paper, we propose a novel joint representation learning framework for users and POIs in LBSN, named UP2VEC. In UP2VEC, we present a heterogeneous LBSN graph to incorporate all these aforementioned factors. Specifically, the transition probabilities between nodes inside the heterogeneous graph are derived by jointly considering these contextual factors. The latent representations of users and POIs are then learnt by matching the topological structure of the heterogeneous graph. For evaluating the effectiveness of UP2VEC, a series of experiments are conducted with two real-world datasets (Foursquare and Gowalla) in terms of POI recommendation and social link prediction. Experimental results demonstrate that the proposed UP2VEC significantly outperforms the existing state-of-the-art alternatives. Further experiment shows the superiority of UP2VEC in handling cold-start problem for POI recommendation. 相似文献
4.
能有效预测在线社交网络中人的行为有较高的应用价值,对当前在线社交网络存在多种交互方式的现象进行分析,以此构建一种多维社交关系的在线社交网络链路预测的方法。基于社会影响和选择两种机制构建在线社交网络多维网络结构,综合各个维度节点和连边的拓扑结构,实现多维社交网络预测矩阵的集结,在降维后的网络上进行链路预测。利用百度贴吧数据集实验验证,从AUC和精确度两方面评价模型,结果表明本文提出的多维预测方法是有效的。 相似文献
5.
Fattane Zarrinkalam Mohsen Kahani Ebrahim Bagheri 《Information processing & management》2018,54(2):339-357
Inferring users’ interests from their activities on social networks has been an emerging research topic in the recent years. Most existing approaches heavily rely on the explicit contributions (posts) of a user and overlook users’ implicit interests, i.e., those potential user interests that the user did not explicitly mention but might have interest in. Given a set of active topics present in a social network in a specified time interval, our goal is to build an interest profile for a user over these topics by considering both explicit and implicit interests of the user. The reason for this is that the interests of free-riders and cold start users who constitute a large majority of social network users, cannot be directly identified from their explicit contributions to the social network. Specifically, to infer users’ implicit interests, we propose a graph-based link prediction schema that operates over a representation model consisting of three types of information: user explicit contributions to topics, relationships between users, and the relatedness between topics. Through extensive experiments on different variants of our representation model and considering both homogeneous and heterogeneous link prediction, we investigate how topic relatedness and users’ homophily relation impact the quality of inferring users’ implicit interests. Comparison with state-of-the-art baselines on a real-world Twitter dataset demonstrates the effectiveness of our model in inferring users’ interests in terms of perplexity and in the context of retweet prediction application. Moreover, we further show that the impact of our work is especially meaningful when considered in case of free-riders and cold start users. 相似文献
6.
《Information processing & management》2023,60(4):103348
Graph neural networks (GNN) have emerged as a new state-of-the-art for learning knowledge graph representations. Although they have shown impressive performance in recent studies, how to efficiently and effectively aggregate neighboring features is not well designed. To tackle this challenge, we propose the simplifying heterogeneous graph neural network (SHGNet), a generic framework that discards the two standard operations in GNN, including the transformation matrix and nonlinear activation. SHGNet, in particular, adopts only the essential component of neighborhood aggregation in GNN and incorporates relation features into feature propagation. Furthermore, to capture complex structures, SHGNet utilizes a hierarchical aggregation architecture, including node aggregation and relation weighting. Thus, the proposed model can treat each relation differently and selectively aggregate informative features. SHGNet has been evaluated for link prediction tasks on three real-world benchmark datasets. The experimental results show that SHGNet significantly promotes efficiency while maintaining superior performance, outperforming all the existing models in 3 out of 4 metrics on NELL-995 and in 4 out of 4 metrics on FB15k-237 dataset. 相似文献
7.
[目的/意义]总结了基于在线社交媒体数据的广度学习工作研究进展,从情报学的视角分析了广度学习的应用展望及未来发展趋势。[方法/过程]利用文献统计分析方法,重点分析了广度学习技术在网络嵌入、链路预测、社区检测等在线社交网络分析领域的应用现状。[结果/结论]广度学习可以将多个不同种类的大型异构数据源融合在一起,设计并使用一套统一的分析方法来跨越这些融合的数据源执行协同数据挖掘任务。广度学习在异构社交网络分析中的这些成功应用为其在情报学领域中的研究奠定了理论基础和技术支持,将会有更广泛更深远的研究成果出现。 相似文献
8.
A systematic review of empirical research on knowledge and growth in small firms is reported. The findings cover how human and social capital, organizational systems, and knowledge networks combine to facilitate or restrict growth. Findings highlight the situated, complex and idiosyncratic nature of small firm growth and the tensions between this experience and a prevailing view of knowledge in the existing research as a codifiable and transferable asset. A need for supplementary small firm heuristics (to age, size and sector) and epistemologies (to knowledge as an asset) and research approaches are identified to better investigate this diversity. 相似文献
9.
Enterprises in both the public and private sector undertake knowledge management (KM) initiatives through which they hope to engender a new, more adaptive and flexible culture of learning and innovation in their organisations. Creative activities involving social learning and innovation are, however, more common in less formal entities such as communities of practice at work and community service organisations in civil society. This paper presents the results and implications of collaborative research into the understanding, development and evaluation of socio-technical systems (STS) designed to mobilise collective knowledge in diverse community settings. The research concerns information and communication technologies (ICT)-mediated activities of communities in the broader civil society and also those in formal organisations. The paper describes and critically evaluates a set of three STS that have the potential to support the collective knowledge of innovative groups, teams and networks, which can all be considered forms of community. The findings could be of strategic value to business, government and community service organisations initiating KM programmes aimed at using collective learning to support innovation. 相似文献
10.
This paper investigates cluster synchronization in community networks with nonidentical nodes. Several effective strategies to enhance the coupling weights are designed. For the first time, adaptive enhancing factor method combined with edge-based pinning control is adopted to achieve synchronization. Furthermore, distributed adaptive pinning control scheme is adopted based on the local information of node dynamics. Noticeably, only the coupling weights of spanning trees in each community are tuned, which are low-cost and more practicable. Based on Lyapunov stability theory, some sufficient conditions for cluster synchronization are derived. Numerical simulations are provided to verify the effectiveness of the theoretical results. 相似文献
11.
借助社会网络分析方法,通过与知识管理活动尤其是知识共享的结合,构建了一个基于知识管理背景的知识创新与共享模型,探讨社会网络的联结特性和结构形态如何对知识创新与共享的实现过程产生促进作用,最后利用社会网络分析工具对知识管理虚拟社区中的知识共享行为进行了量化分析,从而找出人与人之间隐含的知识分布特点与共享规律,并对如何提高社区知识共享效率提出相应的策略,以促进网络结构的优化和虚拟社区的集体智慧发展。 相似文献
12.
以索尼公司1995～2011年的技术联盟网络为对象,考察了索尼与其他公司建立的联盟网络类型与知识管理动机之间的匹配关系。首先,基于社会网络的结构特征与类型,提出弱联系、强联系和标准联盟网络与知识获取、探索型知识共享和利用型知识共享、知识商业化之间的匹配假设;其次,通过对研究变量的操作化定义,分析了许可证、股权合资、合作联盟、标准联盟网络与知识管理动机之间的匹配性。最后从重视边界管理者的作用、提高联盟网络类型与知识管理动机的匹配性以及提升联盟网络管理能力等方面提出了管理启示。 相似文献
13.
Minna JanhonenAuthor Vitae Jan-Erik JohansonAuthor Vitae 《International Journal of Information Management》2011,31(3):217-225
This study aims to find out how different processes of knowledge management and patterns of social networking affect team performance. Our data on teams originate from a sample of different organizations from a variety of both public and private industries in Finland (76 teams; 499 employees). One of the main deficiencies in the current literature on knowledge and networks is that they tend to concentrate on specific types of teams in a single organization context. Our aim was to put the team phenomenon into an everyday context by analysing the interplay of knowledge creation and social networks in teams which function on a permanent basis in a variety of industry contexts. Both knowledge creation and social networking contributed to performance, but the results showed that whereas team members see the knowledge conversion processes as central to performance, top management emphasize the importance of social networks in value creation. In our examination, lively interaction between team members, combined with team leaders’ intra-organizational networks, contributed to team performance. 相似文献
14.
Kalpana Shankar 《The Information Society》2013,29(2):116-120
In emergencies, information sharing among and between officials, volunteers, and citizens is essential for effective recovery and management. Recently, volunteers and others have been using community technology centers, community wireless networks, and end-user social technologies such as blogs and Wikis to prepare for emergencies and communicate and coordinate response when they happen. This article argues that there is a need for a research agenda that combines our knowledge of community informatics with the principles of disaster management to understand how social networks form and mobilize in disasters and how information and communication technologies should be designed and deployed to engage, inform, and mobilize those volunteer and citizen networks. 相似文献
15.
《The Information Society》2008,24(2):116-120
In emergencies, information sharing among and between officials, volunteers, and citizens is essential for effective recovery and management. Recently, volunteers and others have been using community technology centers, community wireless networks, and end-user social technologies such as blogs and Wikis to prepare for emergencies and communicate and coordinate response when they happen. This article argues that there is a need for a research agenda that combines our knowledge of community informatics with the principles of disaster management to understand how social networks form and mobilize in disasters and how information and communication technologies should be designed and deployed to engage, inform, and mobilize those volunteer and citizen networks. 相似文献
16.
Social networks provide individuals with diverse or redundant information depending on the network structure. Both types of information offer advantages for generating new ideas. At the same time, network structure and network content are independent. As a result, two individuals with the same network position can access diverse or redundant content from their social peers. In this study, we investigate the function of social networks in innovative endeavors given individuals’ different kinds of information accessing behavior. In accordance with previous research, we argue that individuals with a broker status access more diverse information through non-redundant network structures and develop, on average, more novel ideas. We further propose that redundancy in content complements brokers’ structural non-redundancy by providing familiar knowledge elements and therefore interpretability, while non-redundancy in both content and structure leads to information overload. Thus, we hypothesize that brokers accessing more information depth, and independently, less information breadth generate newer ideas. To test our hypotheses, we collected data from a popular online maker community containing 18,146 ideas, 19,919 profiles, and 52,663 comments. We used topic modeling (Latent Dirichlet Allocation) to extract hidden knowledge elements and social network analysis to identify brokers. In line with our hypotheses, we find that information depth (breadth) strengthens (weakens) a favorable broker position. These findings have implications for the literature on idea generation in social networks and household sector innovation. 相似文献
17.
互联网众筹可视为一种大众参与和共同创造的新兴创业孵化范式。本文基于社会资本理论,构建了异质性社会资本影响互联网众筹绩效的理论模型。特别的,注意到发起人两种不同的社会资本,即“社交资本”和“社群资本”,分别具有“对人不对事”和“对事不对人”的特质,因而在众筹事前风险甄别和事后监督惩罚中具有不同作用机理。最后,采集“众筹网”微观数据对理论假设进行检验。研究结果表明：1）发起人的社群资本和社交资本均对众筹绩效有正向影响;2）相对于社交资本,发起人的社群资本对众筹绩效的影响更加显著。 相似文献
18.
In this paper, social network analysis techniques and regression models are used to explain the impact of the level of knowledge development on ego-network redundancy in a community of hospital physicians. Our findings document that the level of knowledge development and the extent to which knowledge is homogeneously distributed amongst collaborating physicians are related to the redundancy of their advice networks, albeit with opposite effects. Our results highlight also that the impact of these relationships on network redundancy is moderated based on whether partnering individuals belong to different professional groups. Our results provide valuable input for the management of knowledge networks within professional organizations. 相似文献
19.
This study examined the development of individual social capital in a distributed learning community. Feld's theory of focused choice predicts that the formation of network ties is constrained by contextual factors that function as foci of activities. In our research, we examined how group assignment and location could function as such foci to influence the development of individual social capital in a distributed learning community. Given that networks with different content flows may possess different properties, we examined two different types of networks—task-related instrumental networks and non-task-related expressive networks. A longitudinal research design was used to evaluate the evolution of networks over time. Hypotheses were tested using a sample of 32 students enrolled in a distributed learning class. The results show strong support for Feld's theory. While serving as foci of activities to organize social interactions, both group assignment and geographic separation can also function to fragment a learning community. 相似文献