Recommender systems are powerful tools developed to mitigate information overload in e-commerce platforms. Social recommender systems leverage social relations among users to predict their preferences. Recently, graph neural networks have been utilized for social recommendations, modeling user-user social relations and user–item interactions as graph-structured data. Despite their improvement over traditional systems, most existing social recommender systems exploit only first-order social relations and overlook the importance of social influence diffusion from higher-order neighbors in social networks. Additionally, these techniques often treat all neighboring nodes equally, without highlighting the most influential ones. To address these challenges, we introduce GATE-SR, a novel model that leverages a multi-head graph attention autoencoder to capture indirect social influence from higher-order neighbors while emphasizing the most relevant users. Moreover, we incorporate implicit social connections derived from coherent communities within the network. While GATE-SR performs comparably to baseline models in rich data environments, its strength lies in excelling at cold-start scenarios—where other models often fall short. This focus on cold-start performance aligns with our goal of building a robust recommender system for real-world challenges. Through extensive experiments on three real-world datasets, we demonstrate that GATE-SR outperforms several state-of-the-art baselines in cold-start scenarios. These results highlight the crucial role of accentuating the most influential neighbors, both explicit and implicit, when modeling higher-order social connections for more accurate recommendations. • Varied attention enhances recommendations by assigning importance to neighbors. • Autoencoder’s stacked layers model high-order social relations effectively. • Community detection cuts over-individualized recommendations, optimizing complexity. • Mitigate data sparsity using implicit social connections. • Adept at mitigating cold-start probelm, emphasizing higher-order social influence.

Previous research has identified various organisational factors that influence the use of Enterprise Social Media (ESM) applications in companies, as well as the outcomes of ESM use. However, these valuable contributions manifest themselves as isolated findings and sometimes provide conflicting results, for example, regarding the impact on organisational hierarchies. It appears that the context of ESM use, which may explain the diversity of results, is not yet sufficiently understood. Against this backdrop, we develop a theoretical framework that adopts a morphogenetic perspective and draws on organisational theory to conceptualise the interplay between four organisational context dimensions and ESM use. By applying this framework to an empirical analysis of a global organisation, we develop a systematic understanding of the contextual nature of ESM use. Our findings help to position and connect previous research on hierarchies, the role of management, and the transformative outcomes of ESM use.

Self-tracking devices and applications have become popular in recent years and changed user behaviour. Previous research has primarily focused on the adoption of self-tracking devices and their effects on self-assessment. As adoption increases, user engagement becomes prominent for the continuous use of the devices and the applications. In this study, we focus on user engagement with activity tracking applications, e.g., Fitbit Flex and Jawbone Up that offer data on user performance. We collected data from semi-structured interviews with 54 participants. We propose a process model comprising four stages which involve distinct user interactions with data: review, react, reflect, and respond. We advance research in this domain by the proposed process model that explicates user engagement in two cases: when the user encounters satisfactory or unsatisfactory results. In the latter case, we depict four response tactics when users are confronted with unsatisfactory results.

Over the decades, the evolving information systems (IS) research community and its academic output has greatly expanded. This paper offers an integrated analysis of multiple dimensions of network interconnectedness of the growing IS discipline. In line with the social and intellectual dimensions of the underlying theory of science, we synthesise multiple network views on authors, institutions, journal outlets, citations, and themes into a multi-dimensional knowledge network infrastructure analysis of collaborative networks of IS researchers and their academic output. We further introduce two fragmentation types to better address the dynamics of the IS discourse discussed in previous research. Based on a corpus of all 3587 AIS basket of 8 journal articles over 20 years, we use the analytical framework to study whether the fast growth of the IS discipline resulted in a reduced coherence of the overall academic collaboration and the research themes. The analysis reveals that the sampled IS researcher community developed a large core component with influential bridging people who mitigate fragmentation and centralisation. This IS community structure constitutes a valuable asset to cope with fragmentation tendencies in the intellectual dimension (research topics) resulting from many short-term topic bursts and from centralisation of conceptual terms.

Despite growing interest in delineating the social identity of Information Systems (IS) research and the network structures of its scholarly community, little is known about how the IS community network is shaped by individual conceptions and what motivates IS researchers to engage in research collaboration. Using an exploratory theoretical framework that is based on three dimensions of social capital theory, we examined 32 years of scientific co-authorship in an international IS researcher community. We formulated propositions to empirically examine the multi-level relationships between personal drivers and the resulting complex network organization of the IS community. Our propositions are refined with qualitative interviews and tested using a survey. This process revealed a collaborative research culture with several individual dispositions, including a strategic structural focus, a cognitive focus and a relational focus. These exist among actors displaying a range of differing behaviours such as active engagement and passive serendipity. Our study indicates individual differences at the conception stage of engaging in academic collaboration impact on the resulting network-level configuration. We identified that regional preference, maturity life cycles and lack of small-world properties highlight the important role of senior members as structural backbones and brokers within the IS community.

Communication in online social networks has been analyzed for some time regarding the expression of sentiments. So far, very little is known about the relationship between sentiments and network emergence, dissemination patterns and possible differences between positive and negative sentiments. The dissemination patterns analyzed in this study consist of network motifs based on triples of actors and the ties among them. These motifs are associated with common social network effects to derive meaningful insights about dissemination activities. The data basis includes several thousand social networks with textual messages classified according to embedded positive and negative sentiments. Based on this data, sub-networks are extracted and analyzed with a dynamic network motif analysis to determine dissemination patterns and associated network effects. Results indicate that the emergence of digital social networks exhibits a strong tendency towards reciprocity, followed by the dominance of hierarchy as an intermediate step leading to social clustering with hubs and transitivity effects for both positive and negative sentiments to the same extend. Sentiments embedded in exchanged textual messages do only play a secondary role in network emergence and do not express differences regarding the emergence of network patterns.