Consensus dynamics in online collaboration systems

Wang

Physica A: Statistical Mechanics and its Applications

et al. 2019

Most previous works on opinion modeling lack the simultaneous study of individual mental activity and group behavior. This paper is motivated to propose an agent-based online opinion formation model based on attitude change theory, group behavior theory and evolutionary game theory in the perspective of sociology and psychology. In this model, there are three factors influencing the persuasion process, including credibility of the leaders, characteristic of the recipient, and group environment. The proposed model is applied to Twitter to analyze the influence of topic type, parameter changing, and opinion leaders on opinion formation.Experimental results show that the opinion evolution of controversial topic shows greater uncertainty and sustainability. The ratio of benefit to cost has a significant impact on opinion formation and a moderate ratio will result in the longest relaxation time or most unified global opinions. Furthermore, celebrities with a large number of followers are more capable of influencing public opinion than experts. This paper enriches the researches on opinion formation modeling, and the results provide managerial insights for business on public relations and market prediction.

Section: Opinion Modelsmentioning

confidence: 99%

Agent-based opinion formation modeling in social network: A perspective of social psychology

Wang

Physica A: Statistical Mechanics and its Applications

et al. 2019

“…The DML encapsulates all database-related CRUD operations (i. e., create, retrieve, update, delete) in one module and thus, enables easy access to the underlying data backend. As shown in Figure 2, we utilize the high-performance search engine Apache Solr 5 . This data backend solution not only guarantees scalability and (near) real-time recommendations but also the support of multiple entities like the users, datasets and services we encounter here.…”

Section: System Architecturementioning

confidence: 99%

“…These interactions are created by users voting for a service. Finally, we establish a collaboration network between datasets and services (see e.g., [5]). Thus, we create a link between a dataset and a service when a user has interacted with both, the dataset and the service, which leads to 95,249 interactions.…”

Section: Datamentioning

confidence: 99%

Using the Open Meta Kaggle Dataset to Evaluate Tripartite Recommendations in Data Markets

Kowald,

Traub,

Theiler

et al. 2019

Preprint

Self Cite

This work addresses the problem of providing and evaluating recommendations in data markets. Since most of the research in recommender systems is focused on the bipartite relationship between users and items (e. g., movies), we extend this view to the tripartite relationship between users, datasets and services, which is present in data markets. Between these entities, we identify four use cases for recommendations: (i) recommendation of datasets for users, (ii) recommendation of services for users, (iii) recommendation of services for datasets, and (iv) recommendation of datasets for services. Using the open Meta Kaggle dataset, we evaluate the recommendation accuracy of a popularity-based as well as a collaborative filtering-based algorithm for these four use cases and find that the recommendation accuracy strongly depends on the given use case. The presented work contributes to the tripartite recommendation problem in general and to the under-researched portfolio of evaluating recommender systems for data markets in particular.

“…In particular, community structures are often found in empirical real-world complex networks. These structures have a major impact on information propagation across graphs [1] as they provide barriers for propagation [9]. The process of information propagation forms the basis for many studied applications on graphs.…”

Section: Introductionmentioning

confidence: 99%

On the Impact of Communities on Semi-supervised Classification Using Graph Neural Networks

Hussain,

Duricic,

Lex

et al. 2020

Preprint

Self Cite

Graph Neural Networks (GNNs) are effective in many applications. Still, there is a limited understanding of the effect of common graph structures on the learning process of GNNs. In this work, we systematically study the impact of community structure on the performance of GNNs in semi-supervised node classification on graphs. Following an ablation study on six datasets, we measure the performance of GNNs on the original graphs, and the change in performance in the presence and the absence of community structure. Our results suggest that communities typically have a major impact on the learning process and classification performance. For example, in cases where the majority of nodes from one community share a single classification label, breaking up community structure results in a significant performance drop. On the other hand, for cases where labels show low correlation with communities, we find that the graph structure is rather irrelevant to the learning process, and a feature-only baseline becomes hard to beat. With our work, we provide deeper insights in the abilities and limitations of GNNs, including a set of general guidelines for model selection based on the graph structure.