Modeling Teams Performance Using Deep Representational Learning on Graphs

TL;DR

A graph neural network model is proposed to predict a team’s performance while identifying the drivers determining such outcome, based on three architectural channels: topological, centrality, and contextual, which capture different factors potentially shaping teams’ success.

Abstract

The large majority of human activities require collaborations within and across formal or informal teams. Our understanding of how the collaborative efforts spent by teams relate to their performance is still a matter of debate. Teamwork results in a highly interconnected ecosystem of potentially overlapping components where tasks are performed in interaction with team members and across other teams. To tackle this problem, we propose a graph neural network model designed to predict a team's performance while identifying the drivers that determine such an outcome. In particular, the model is based on three architectural channels: topological, centrality, and contextual which capture different factors potentially shaping teams' success. We endow the model with two attention mechanisms to boost model performance and allow interpretability. A first mechanism allows pinpointing key members inside the team. A second mechanism allows us to quantify the contributions of the three driver effects in determining the outcome performance. We test model performance on a wide range of domains outperforming most of the classical and neural baselines considered. Moreover, we include synthetic datasets specifically designed to validate how the model disentangles the intended properties on which our model vastly outperforms baselines.

Figures (14)

• Figure 1: MENTOR. The architecture of our model is based on the usage of three channels: topology (T), centrality (C) and contextual (L). Each channel returns a corresponding embedding vector for each subgraph $S_i$. The outputs of the three channels are then merged by means of an attention mechanism that estimates the importance of a specific effect.
• Figure 2: Isolation procedure. Graphical illustration of the isolation procedure of the subgraphs $S_i$ and $S_j$ from $G$, performed by the topology channel. During this phase, the shared member $v$ is duplicated in order to be present in both $\tilde{S}_i$ and $\tilde{S}_j$ subgraphs.
• Figure 3: Hypernodes creation. Graphical illustration of the preprocessing phase of the centrality channel: subgraphs $S_i$ and $S_j$ of $G$ are collapsed to the hypernodes $v_i'$ and $v_j'$. Edges in the new hypergraph are weighted according to the connectivity structure of the original graph $G$.
• Figure 4: Anchor-sets. Graphical illustration of the anchor-sets $A_i$ generated by the P-GNN algorithm in order to potentially cover the entire volume of the graph $H$.
• Figure 5: Confusion matrices on real-world datasets. The confusion matrices on real-world datasets: a) IMDb; b) Dribbble c) Kaggle. The results refer to the configuration corresponding to the seed which returns the highest accuracy.