PREDICTION OF LUNG CANCER USING ARTIFICIAL NEURAL NETWORK
N
Numan
Faculty of Computer and Information,Mansoura University, Egypt
author
S
Abuelenin
Faculty of Computer and Information,Mansoura University, Egypt
author
M
Rashad
Computer Science Department,Faculty of Computers and Information,
Mansoura University, Egypt
author
text
article
2016
eng
International Journal of Intelligent Computing and Information Sciences
Ain Shams University, Faculty of Computer and Information Science
1687-109X
16
v.
2
no.
2016
1
19
https://ijicis.journals.ekb.eg/article_10013_63b00a23ece071c520656fe6dfe039cf.pdf
dx.doi.org/10.21608/ijicis.2018.10013
CIRCLES ROUTING PROTOCOL FOR WIRELESS SENSOR NETWORK
S
Koriem
Computer Systems Eng. Department, Faculty of Engineering,Al-Azhar University, Cairo, Egypt
author
M
Bayoumi
Systems and Computers Engineering department, Faculty of Engineering, AL-Azhar University,Nasr city, Cairo, Egypt.
author
S
Nouh
Systems and Computers Engineering department, Faculty of Engineering, AL-Azhar University,Nasr city, Cairo, Egypt.
author
text
article
2016
eng
Abstract: Wireless Sensor Network (WSN) is an emerging technology for monitoring physical world.WSN consists of large number of sensor nodes operated by battery mostly in harsh environment. These wireless nodes are very limited in battery power and communication processes. Gathering sensed information in an efficient manner is critical to operate the sensor network for a long period time. In this paper, a new Routing Protocol CRP (Circles Routing Protocol) is proposed. The CRP is developed for achieving QoS (Quality of Service) in terms of network life time, power consumption, packet delivery, and network throughput by distributing the energy load among all sensor nodes. The dynamic behavior of the proposed CRP depends on executing the following steps.Firstly, CRP cutting down the playground into many clusters by using the advantage of the grid construction to physically partition the playground into many small individual clusters. Secondly, CRP electing one node in each cluster, as a Cluster Head (CH), form a circular chain within each cluster to collect and fuse data from the other nodes. Thirdly, CRP collecting every four adjacent clusters in one group called inter four clusters. Then, CRP constructing a circular chain within these four clusters containing the four CH nodes. Fourthly, CRP electing one of these CH nodes as an inter CH node to collect and fuse data from the other CH nodes. Finally, CRP constructing a circular chain containing the four inter CH nodes.
Then CRP electing one of inter CH nodes to be the outer CH node which collects and fuses the datafrom the other nodes and subsequently transmit this data to the Base Station (BS). In the performanceanalysis, we use the NS-2 Simulator as a simulation technique to study and analysis the performance ofCRP protocol. To verify from the correctness of the obtained performance results, we compare the CRPresults with those obtained from LEACH (Low-Energy Adaptive Clustering Hierarchy) and PEGASIS(Power-Efficient Gathering in Sensor Information System) protocols.
International Journal of Intelligent Computing and Information Sciences
Ain Shams University, Faculty of Computer and Information Science
1687-109X
16
v.
2
no.
2016
21
36
https://ijicis.journals.ekb.eg/article_10903_6c66771a16939742237fe21b8eab3072.pdf
dx.doi.org/10.21608/ijicis.2018.10903
IMPROVING FEATURE MAPS IN EARLY LAYERS OF CONVOLUTIONAL NEURAL NETWORKS USING OTSU METHOD
A
Al-furas
Faculty of Computer and Information,Mansoura University, Egypt.
author
M
AL-dosuky
Faculty of Computer and Information,Mansoura University, Egypt.
author
Taher
Hamza
Computer Science Department Faculty of Computer and Information Sciences, Mansoura University - Egypt
author
text
article
2016
eng
Abstract: A novel deep architecture Thresholding Convolution Neural Network (ThCNN) progresses in this paper; Which is a simple and effective method to regularizing features map in the early layers of Convolution Neural Network(CNN). One of the issues identified with deep learning is the features in early layers that robustness and discriminativeness. In this paper, we compute the optimal global threshold to determine the features that are passed to the next layers. We then evaluate ThCNN on an MNIST dataset comparing it CNN by applying multiple trained models. It yield decent accuracy compared to traditional CNN. It gives a 99.5%
International Journal of Intelligent Computing and Information Sciences
Ain Shams University, Faculty of Computer and Information Science
1687-109X
16
v.
2
no.
2016
37
45
https://ijicis.journals.ekb.eg/article_10905_ae527d2d07b489a78d5cbcd28d164f12.pdf
dx.doi.org/10.21608/ijicis.2018.10905