Multiplex graphs have been recently proposed as a model to represent high-level complexity in real-world networks such as heterogeneous social networks where actors could be characterized by heterogeneous properties and could be linked with different types of social interactions. This has brought new challenges in community detection, which aims to identify pertinent groups of nodes in a complex graph. In this context, great efforts have been made to tackle the problem of community detection in multiplex graphs. However, most of the proposed methods until recently deal with static multiplex graph and ignore the temporal dimension, which is a key characteristic of real networks. Even more, the few methods that consider temporal graphs, they just propose to follow communities over time and none of them use the temporal aspect directly to detect stable communities, which are often more meaningful in reality. Thus, this paper proposes a new two-step method to detect stable communities in temporal multiplex graphs. The first step aims to find the best static graph partition at each instant by applying a new hybrid community detection algorithm, which considers both relations heterogeneities and nodes similarities. Then, the second step considers the temporal dimension in order to find final stable communities. Finally, experiments on synthetic graphs and a real social network show that this method is competitive and it is able to extract high-quality communities.
In the era of data science and big data analytics, people analytics help organizations and their human resources (HR) managers to reduce attrition by changing the way of attracting and retaining talent. In this context, employee attrition presents a critical problem and a big risk for organizations as it affects not only their productivity but also their planning continuity. In this context, the salient contributions of this research are as follows. Firstly, we propose a people analytics approach to predict employee attrition that shifts from a big data to a deep data context by focusing on data quality instead of its quantity. In fact, this deep data-driven approach is based on a mixed method to construct a relevant employee attrition model in order to identify key employee features influencing his/her attrition. In this method, we started thinking 'big' by collecting most of the common features from the literature (an exploratory research) then we tried thinking 'deep' by filtering and selecting the most important features using survey and feature selection algorithms (a quantitative method). Secondly, this attrition prediction approach is based on machine, deep and ensemble learning models and is experimented on a large-sized and a medium-sized simulated human resources datasets and then a real small-sized dataset from a total of 450 responses. Our approach achieves higher accuracy (0.96, 0.98 and 0.99 respectively) for the three datasets when compared previous solutions. Finally, while rewards and payments are generally considered as the most important keys to retention, our findings indicate that 'business travel', which is less common in the literature, is the leading motivator for employees and must be considered within HR policies to retention.
Abstract. With the growing number of users and the huge amount of information in dynamic social networks, contextualizing community detection has been a challenging task. Thus, modeling these social networks is a key issue for the process of contextualized community detection. In this work, we propose a temporal multiplex information graphbased model to represent dynamic social networks: we consider simultaneously the social network dynamicity, its structure (different social connections) and various members' profiles so as to calculate similarities between "nodes" in each specific context. Finally a comparative study on a real social network shows the efficiency of our approach and illustrates practical uses.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.