J
Johnson P. Thomas
Researcher at Oklahoma State University–Stillwater
Publications - 81
Citations - 2631
Johnson P. Thomas is an academic researcher from Oklahoma State University–Stillwater. The author has contributed to research in topics: Quality of service & Big data. The author has an hindex of 18, co-authored 78 publications receiving 2476 citations. Previous affiliations of Johnson P. Thomas include University of Reading & Pace University.
Papers
More filters
Journal ArticleDOI
Feature deduction and ensemble design of intrusion detection systems
TL;DR: This study investigated the performance of two feature selection algorithms involving Bayesian networks and Classification and Regression Trees and an ensemble of BN and CART and proposed an hybrid architecture for combining different feature selection algorithm for real world intrusion detection.
Journal ArticleDOI
Modeling intrusion detection system using hybrid intelligent systems
TL;DR: Two hybrid approaches for modeling IDS are presented as a hierarchical hybrid intelligent system model (DT-SVM) and an ensemble approach combining the base classifiers to maximize detection accuracy and minimize computational complexity.
Journal ArticleDOI
Factors Affecting Longevity of Homograft Valves Used in Right Ventricular Outflow Tract Reconstruction for Congenital Heart Disease
James S. Tweddell,Andrew N. Pelech,Peter C. Frommelt,Kathleen A. Mussatto,J. D. Wyman,Raymond T. Fedderly,Stuart Berger,Michele A. Frommelt,Derek Lewis,David Z. Friedberg,Johnson P. Thomas,R. Sachdeva,S. B. Litwin +12 more
TL;DR: Homograft valves used for RVOT reconstruction provide effective intermediate palliation with excellent late survival and factors that adversely affect graft longevity include younger age, longer donor warm ischemic time, smaller homografted size, use of aortic homograft in the older patient, and extracardiac operative technique.
Proceedings ArticleDOI
QoS impact on user perception and understanding of multimedia video clips
TL;DR: Results show that the quality of video clips can be severely degraded without the user having to perceive any significant 10ss of informational content.
Journal ArticleDOI
D-SCIDS: distributed soft computing intrusion detection system
TL;DR: In this article, the authors evaluated three fuzzy rule-based classifiers to detect intrusions in a network and compared them with other machine learning techniques like decision trees, support vector machines and linear genetic programming.