A Simple Linear Time (1+έ)-Approximation Algorithm for k-Means Clustering in Any Dimensions.
Amit KumarYogish SabharwalSandeep SenPublished in: FOCS (2004)
Keyphrases
- worst case
- learning algorithm
- computational complexity
- optimization algorithm
- dynamic programming
- recognition algorithm
- k means
- cost function
- computationally efficient
- high accuracy
- probabilistic model
- convergence rate
- segmentation algorithm
- optimal solution
- data sets
- experimental evaluation
- np hard
- search space
- detection algorithm
- clustering method
- times faster
- approximation ratio
- computational cost
- theoretical analysis
- preprocessing
- objective function
- improved algorithm