Using Machine Learning Ensemble Methods to Predict Execution Time of e-Science Workflows in Heterogeneous Distributed Systems.
Farrukh NadeemDaniyal M. AlghazzawiAbdulfattah S. MashatKhalid FaqeehAbdullah Al-Malaise Al-GhamdiPublished in: IEEE Access (2019)
Keyphrases
- distributed systems
- ensemble methods
- machine learning methods
- machine learning
- decision trees
- loosely coupled
- prediction accuracy
- base classifiers
- fault tolerant
- workflow management systems
- benchmark datasets
- random forests
- ensemble learning
- distributed environment
- generalization ability
- machine learning algorithms
- drifting concepts
- bias variance analysis
- load balancing
- data mining
- fault tolerance
- geographically distributed
- random forest
- web services
- mobile agents
- business processes
- bootstrap sampling
- distributed database systems
- classifier ensemble
- distributed computing
- learning algorithm
- ensemble feature selection
- data replication
- feature selection
- support vector machine
- business process
- learning tasks
- neural network
- decision tree ensembles
- deadlock detection
- classification accuracy
- meta learning
- supervised learning
- feature subset