Enhancing Team Composition in Professional Networks: Problem Definitions and Fast Solutions

Abstract

In this paper, we study ways to enhance the composition of teams based on new requirements in a collaborative environment. We focus on recommending team members who can maintain the team’s performance by minimizing changes to the team’s skills and social structure. Our recommendations are based on computing team-level similarity, which includes skill similarity, structural similarity as well as the synergy between the two. Current heuristic approaches are one-dimensional and not comprehensive, as they consider the two aspects independently. To formalize team-level similarity, we adopt the notion of graph kernel of attributed graphs to encompass the two aspects and their interaction. To tackle the computational challenges, we propose a family of fast algorithms by (a) designing effective pruning strategies, and (b) exploring the smoothness between the existing and the new team structures. Extensive empirical evaluations on real world datasets validate the effectiveness and efficiency of our algorithms.

Extra

“…With graph kernel, we aim to find a candidate that makes the new team most similar to the old team. For details regarding other scenarios (i.e., team expansion and shrinkage), please refer to [2,4]. B -Explaining Team Recommendation: we seek to understand the team recommendation results via the influence of various graph elements to the corresponding graph kernels.…”

“…For instance, in team member replacement, by applying random walk graph kernel to the team networks before and after replacement, it encodes both the skill match, structure match as well as the interaction between the two during the replacement process [3]. Team member replacement further enables other team recommendation scenarios [4]. However, these existing methods lack intuitive ways to explain why the underlying algorithm gives the specific recommendation for a given team optimization scenario.…”

Skill ranking of researchers via hypergraph

“…Many efforts have been made to analyze and optimize teams in terms of their topology structure (Milojević, 2014;Kong et al, 2018;Wang et al, 2017), which have made great contributions. However, team effectiveness not only depends on the appropriate team topology structure, but also depends on their function component, such as the abilities and skill distributions of team members (Li et al, 2017). Some works built, evaluated and refined teams in consideration of both skills of team members and the team topology structure (Li et al, 2017;Wang, Zhao & Ng, 2016).…”

“…However, team effectiveness not only depends on the appropriate team topology structure, but also depends on their function component, such as the abilities and skill distributions of team members (Li et al, 2017). Some works built, evaluated and refined teams in consideration of both skills of team members and the team topology structure (Li et al, 2017;Wang, Zhao & Ng, 2016).…”

“…Researchers' skill sets have been used in solving many problems, such as team formation, team member replacement and team refinement (Kong et al, 2016). Many researchers use terms extracted from the paper's keywords and title or conference and journal name as authors' skills (Li et al, 2017). They only consider a binary relationship between researchers and skills, that is, whether a researcher has a particular skill.…”