gptkbp:instanceOf
|
computational complexity theory concept
|
gptkbp:category
|
theoretical computer science
|
gptkbp:complexity
|
NP
|
gptkbp:defines
|
A problem is NP-complete if every problem in NP can be reduced to it in polynomial time.
A decision problem is NP-complete if it is in NP and as hard as any problem in NP.
|
gptkbp:describedBy
|
gptkb:Cook's_theorem
gptkb:Cook-Levin_theorem
gptkb:Levin's_theorem
|
gptkbp:example
|
gptkb:Hamiltonian_cycle_problem
gptkb:3-SAT
gptkb:Clique_problem
gptkb:Graph_coloring_problem
gptkb:Subset_sum_problem
gptkb:Traveling_salesman_problem_(decision_version)
gptkb:Vertex_cover_problem
Boolean satisfiability problem
|
gptkbp:field
|
gptkb:mathematics
computer science
|
gptkbp:firstProblem
|
Boolean satisfiability problem
|
https://www.w3.org/2000/01/rdf-schema#label
|
NP-completeness
|
gptkbp:importantFor
|
central to computational complexity theory
|
gptkbp:introduced
|
gptkb:Stephen_Cook
gptkb:Leonid_Levin
|
gptkbp:introducedIn
|
1971
|
gptkbp:namedFor
|
gptkb:Richard_Karp
|
gptkbp:notablePublication
|
gptkb:Reducibility_Among_Combinatorial_Problems_(Karp,_1972)
gptkb:The_Complexity_of_Theorem-Proving_Procedures_(Cook,_1971)
|
gptkbp:openProblem
|
gptkb:P_vs_NP
|
gptkbp:property
|
If any NP-complete problem can be solved in polynomial time, all problems in NP can be.
If any NP-complete problem is not in P, then P ≠ NP.
|
gptkbp:reductionType
|
polynomial-time reduction
|
gptkbp:relatedTo
|
gptkb:P_vs_NP_problem
gptkb:NP-hardness
NP
|
gptkbp:seeAlso
|
gptkb:Karp's_21_NP-complete_problems
gptkb:Cook-Levin_theorem
gptkb:NP-hardness
P-completeness
Polynomial-time reduction
co-NP-completeness
|
gptkbp:symbol
|
gptkb:NPC
|
gptkbp:bfsParent
|
gptkb:Richard_M._Karp
gptkb:complexity_theory
gptkb:Algorithms_and_Complexity
|
gptkbp:bfsLayer
|
4
|