February 2021 Issue Vol.11 No.2
USE of VARIOUS INTEGRATED TECHNOLOGIES in PREVENTING FLOOD DISASTER
https://ia601407.us.archive.org/9/items/vol11no201/vol11no201.pdf
J. Michael Antony Sylvia1, Dr.M.Pushpa Rani2
Department of Computer Science, Mother Teresa Women’s University, Kodaikanal, India.
Abstract:
Abstract - Monitoring and analyzing the environmental parameters accurately predict the occurrence of flood disaster. Appropriate disaster warning in turn would support the concerned authorities to take precautionary measures. Traditional technologies step behind in efficient disaster management process.
Hence need arouse in further technological development to accurately predict and warn people about the disaster occurrence. Recently developed disaster management systems with integrated technologies such as IoT, Machine Learning, Deep Learning, Crowd Sourcing and Artificial Intelligence
have enhanced the entire process of disaster management with few limitations. This paper surveys the various existing integrated technologies used in disaster management process.
MIXED MODEL OF EXTREME LEARN MACHINE TREE AND RANDOM FOREST CLASSIFIER FOR PREDICTION OF ORAL CANCER
https://ia601407.us.archive.org/9/items/vol11no201/vol11no202.pdf
Dr. M Natarajan
Assistant Professor, Department of Computer Science,
Thanthai Hans Roever College, Perambalur, TamilNadu, India.
Dr. A. Muruganandam
Principal, Department of Computer Science,
Sri Aravindar Arts and Science College, Vanur, Villupuram, TamilNadu, India.
Abstract:
Oral Cancer is one of the deadliest diseases and most of the human are infected by this crucial disease in several parts of the world. It may occur in any part of the oral cavity. The early detection and prevention of oral cancel is very critical issue but it can improve the survival chances considerably, allow for simple treatment and provided the better quality of life for survivors. In existing system, the genetic algorithm is used for feature selection and the Support Vector Machine classifier algorithm is used for classification to predict the oral cancer. The feature selection and the classification is performed separately so the time complexity of the accuracy and prediction time quite complex So to solve this issue in proposed system the firefly algorithm is used for the feature selection and for the classification, mixed model of Extreme Learn Machine (ELM) Random forest classifier technique is used to improve the classification accuracy. The proposed system is tested with normal clinical data set which is improved the classification accuracy and the prediction time compared to existing system.
Read complete February 2021
Read complete February 2021
https://ia601407.us.archive.org/9/items/vol11no201/vol11no202.pdf
Dr. M Natarajan
Assistant Professor, Department of Computer Science,
Thanthai Hans Roever College, Perambalur, TamilNadu, India.
Dr. A. Muruganandam
Principal, Department of Computer Science,
Sri Aravindar Arts and Science College, Vanur, Villupuram, TamilNadu, India.
Abstract: Oral Cancer is one of the deadliest diseases and most of the human are infected by this crucial disease in several parts of the world. It may occur in any part of the oral cavity. The early detection and prevention of oral cancel is very critical issue but it can improve the survival chances considerably, allow for simple treatment and provided the better quality of life for survivors. In existing system, the genetic algorithm is used for feature selection and the Support Vector Machine classifier algorithm is used for classification to predict the oral cancer. The feature selection and the classification is performed separately so the time complexity of the accuracy and prediction time quite complex So to solve this issue in proposed system the firefly algorithm is used for the feature selection and for the classification, mixed model of Extreme Learn Machine (ELM) Random forest classifier technique is used to improve the classification accuracy. The proposed system is tested with normal clinical data set which is improved the classification accuracy and the prediction time compared to existing system.