The centroidal algorithm in molecular similarity and diversity calculations on confidential datasets.
Sergey V. TrepalinNikolay OsadchiyPublished in: J. Comput. Aided Mol. Des. (2005)
Keyphrases
- np hard
- dynamic programming
- experimental evaluation
- computational cost
- clustering method
- preprocessing
- detection algorithm
- similarity measure
- synthetic and real datasets
- synthetic datasets
- times faster
- matching algorithm
- simulated annealing
- worst case
- learning algorithm
- optimal solution
- computational complexity
- search space
- similarity metric
- objective function
- distance metric
- search algorithm
- tree structure
- probabilistic model
- optimization algorithm
- segmentation algorithm
- expectation maximization
- particle swarm optimization
- input data
- k means
- high accuracy