The Univariate Flagging Algorithm (UFA): a Fully-Automated Approach for Identifying Optimal Thresholds in Data.
Mallory Bounds ShethRoy E. WelschNatasha MarkuzonPublished in: CoRR (2016)
Keyphrases
- fully automated
- dynamic programming
- input data
- noisy data
- data sets
- detection algorithm
- optimal solution
- learning algorithm
- data analysis
- data reduction
- computational complexity
- k means
- data sources
- worst case
- image data
- simulated annealing
- segmentation algorithm
- adaptive threshold
- data mining
- missing data
- data mining techniques
- preprocessing
- energy function
- training data
- synthetic datasets
- cluster centers
- semi automated
- three dimensional
- seeded region growing