Assessment of machine learning algorithms in cloud computing frameworks
Authors:
- 1 April 2013
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 98-103
Abstract
In the past decade, digitization of information has led to a data explosion in both volume and complexity. While traditional computing frameworks have failed to provide adequate computing power for the now common data-intensive computing tasks, cloud computing provides an effective alternative to enhance computing power. Machine learning algorithms are powerful analytical methods that allow machines to recognize patterns and facilitate human learning. However, the performance of individual machine learning algorithms within each cloud computing framework remains largely unknown. Furthermore, the lack of a robust selection methodology matching input data with effective machine learning algorithms limits the ability of practitioners to make effective use of cloud computing. This research compares various machine learning algorithms on the widely adopted Apache Mahout framework and the recently introduced GraphLab framework. Whereas previous work has examined the computational architectures of various cloud computing frameworks, this work focuses on a problem-based approach to architecture selection. The experimental results demonstrate that GraphLab generally outperforms Mahout with respect to runtime, scalability, and usability. However, Mahout outperforms GraphLab when the experiment focus shifts to error measurement.Keywords
This publication has 6 references indexed in Scilit:
- Evaluation of trauma care using TRISS method: the role of adjusted misclassification rate and adjusted w-statisticWorld Journal of Emergency Surgery, 2009
- Large-Scale Parallel Collaborative Filtering for the Netflix Prize2008
- MapReduceCommunications of the ACM, 2008
- Logistic regression analysis for experimental determination of forming limit diagramsInternational Journal of Machine Tools and Manufacture, 2006
- Face recognition using LDA-based algorithmsIEEE Transactions on Neural Networks, 2003
- 10.1162/jmlr.2003.3.4-5.993Applied Physics Letters, 2000