TY - JOUR

T1 - Parameterized complexity of team formation in social networks

AU - Bredereck, Robert

AU - Chen, Jiehua

AU - Hüffner, Falk

AU - Kratsch, Stefan

N1 - Funding Information:
This work was started at the research retreat of the TU Berlin Algorithms and Computational Complexity group held in April 2014. Robert Bredereck was partially supported by the DFG , project PAWS (NI 369/10 ), and from September 2016 to September 2017 on postdoctoral leave at the University of Oxford (GB), supported by the DFG fellowship BR 5207/2 . We thank the anonymous reviewers of AAIM 2016 and of Theoretical Computer Science for improving our work.
Funding Information:
This work was started at the research retreat of the TU Berlin Algorithms and Computational Complexity group held in April 2014. Robert Bredereck was partially supported by the DFG, project PAWS (NI 369/10), and from September 2016 to September 2017 on postdoctoral leave at the University of Oxford (GB), supported by the DFG fellowship BR 5207/2. We thank the anonymous reviewers of AAIM 2016 and of Theoretical Computer Science for improving our work.
Publisher Copyright:
© 2017 Elsevier B.V.

PY - 2018/3/22

Y1 - 2018/3/22

N2 - Given a task that requires some skills and a social network of individuals with different skills, the TEAM FORMATION problem asks to find a team of individuals that together can perform the task, while minimizing communication costs. Since the problem is NP-hard, we identify the source of intractability by analyzing its parameterized complexity with respect to parameters such as the total number of skills k, the team size l, the communication cost budget b, and the maximum vertex degree Δ. We show that the computational complexity strongly depends on the communication cost measure: when using the weight of a minimum spanning tree of the subgraph formed by the selected team, we obtain fixed-parameter tractability for example with respect to the parameter k. In contrast, when using the diameter as measure, the problem is intractable with respect to any single parameter; however, combining Δ with either b or l yields fixed-parameter tractability.

AB - Given a task that requires some skills and a social network of individuals with different skills, the TEAM FORMATION problem asks to find a team of individuals that together can perform the task, while minimizing communication costs. Since the problem is NP-hard, we identify the source of intractability by analyzing its parameterized complexity with respect to parameters such as the total number of skills k, the team size l, the communication cost budget b, and the maximum vertex degree Δ. We show that the computational complexity strongly depends on the communication cost measure: when using the weight of a minimum spanning tree of the subgraph formed by the selected team, we obtain fixed-parameter tractability for example with respect to the parameter k. In contrast, when using the diameter as measure, the problem is intractable with respect to any single parameter; however, combining Δ with either b or l yields fixed-parameter tractability.

KW - Computational complexity analysis

KW - Parameterized complexity analysis

KW - Team formation on social networks

UR - http://www.scopus.com/inward/record.url?scp=85020874250&partnerID=8YFLogxK

U2 - 10.1016/j.tcs.2017.05.024

DO - 10.1016/j.tcs.2017.05.024

M3 - Article

AN - SCOPUS:85020874250

SN - 0304-3975

VL - 717

SP - 26

EP - 36

JO - Theoretical Computer Science

JF - Theoretical Computer Science

ER -