KFUPM ePrints

Bacterial Growth Classification with Support Vector Machines: A Comparative Study

Faisal, Kanaan A (2007) Bacterial Growth Classification with Support Vector Machines: A Comparative Study. In: The 2007 International Conference on Machine Learning; Models, Technologies and Applications (MLMTA'07). The 2007 World Congress in Computer Science, Computer Science, Computer Engineering, and Applied Computing (WORLDCOMP’07), June 25-28, 2007, Las Vegas, Nevada, USA.90-98, 2007.



In this paper, we propose to use support vector machines for classification of bacterial growth and non growth database and modeling the probability= of growth. Unlike artificial neural networks paradigms, support vector machines use the kernel functions and support vectors with maximum margin, which allows a better performance. As a practical application of the new approach, support vector machines were investigated for their quality and accuracy in classifi-cation of growth/no-growth state of a pathogenic Escherichia coli R31 in response to temperature and water activity. A comparison with the most common used statistics, machine learning, and data mining schemes was carried out. The results shows that support vector machines classifier based on the Gaussian RBF Kernel was found to do better than most of logistic regression, K-nearest neighbor, probabilistic networks, and multilayer perceptron classifiers.

Item Type:Conference or Workshop Item (Paper)
Uncontrolled Keywords:Bacterial growth; Logistic regression; Support Vector Machines; K-nearest neighbors,neural networks.
Date:June 2007
Date Type:Publication
Divisions:College Of Computer Sciences and Engineering > Information and Computer Science Dept
Creators:Faisal, Kanaan A
ID Code:14623
Deposited On:25 Jun 2008 13:59
Last Modified:12 Apr 2011 13:16

Repository Staff Only: item control page