Towards Optimal Teams in Composite Networks

Abstract

Composite networks, including social networks, information networks, and cognitive networks, have become an important platform where people collaborate with each other to collectively perform a task in the form of teams. The objective of this project is to develop new algorithms and tools to (1) query and detect a desired team with a specific set of requirements, (2) configure a good team for a given task that is more likely to succeed, and (3) refine existing teams to improve the performance over time. By bringing the key network metrics that drive the peak team performance into the team search, formation and refinement process, this project aims to help (1) an organization better configure the limited network and human resource, (2) the decision-makers make a better decision to perform certain tasks that need collaborative effort within a team; and (3) provide feedback to individuals to foster productive behavior change. This project takes a multi-disciplinary approach, consisting of network query, visualization and optimization. First, they formulate team search as a complex network browsing and query problem, where the researchers seek to build interactive, user-friendly, and scalable systems to search the desired teams. Second, they formulate the team formation problem as a multi-team, multi-objective and multi-level optimization problem whose objective function integrates different key network metrics that drive the peak team performance with a careful balance between different metrics. Third, in order to refine existing teams, they will design a similarity measure between two individuals in the context of the team itself, that quantifies both the skill matching and the structural matching as well as the interaction of both. Overall, this project integrates interactive visualization mechanisms and advanced network analysis algorithms for optimizing teams, leading to new algorithms and tools for searching, forming and refining teams in composite networks.

Document Details

Document Type
DoD Grant Award
Publication Date
Oct 11, 2018
Source ID
W911NF1610168

Entities

People

  • Hanghang Tong

Organizations

  • Arizona State University
  • Army Contracting Command
  • United States Army

Tags

Fields of Study

  • Computer science

Readers

  • Neural Network Machine Learning.
  • Systems Analysis and Design
  • Team-Based Human-Centered Cognitive Task Decision Making and Information Performance.