A New Method of K-Means Clustering Algorithm with Events Based on Variable Time Granularity.
Mengxing HuangHongjing LinPublished in: IEEE WISA (2016)
Keyphrases
- high accuracy
- support vector machine svm
- optimization method
- theoretical analysis
- clustering method
- segmentation method
- computational cost
- computer vision
- computationally efficient
- support vector machine
- classification accuracy
- significant improvement
- prior knowledge
- detection method
- error rate
- fully automatic
- evaluation method
- segmentation algorithm
- feature set
- experimental evaluation
- dynamic programming
- cost function
- k means
- pairwise
- preprocessing