DocumentCode
560912
Title
Real-time neural network-based network analyzer for hotspot area
Author
Handayanto, Rahmadya Trias ; Haryono ; Prianggono, Jarot
Author_Institution
Lab. of Software, Islam 45 Univ., Indonesia
fYear
2011
fDate
17-18 Dec. 2011
Firstpage
323
Lastpage
330
Abstract
We present a real-time neural network-based network analyzer system for hotspot area. There are many applications that available in the market today for provide us the network graph of our hotspot areas. These graphs will be analyzed by a network administrator. Because a hotspot area often runs 24 hours, an administrator has a difficulty to monitor the traffic all the time. Therefore we proposed the automatic system to help a network administrator in monitoring the network. This system will replace human skill in interpreting the graph with an Artificial Neural Network System. To minimize the number of input vector we use mean value of axis, so the micro-computer e.g. notebook, laptop, PDA, and other gadgets can handle this system. Testing result showed this system could classify between normal, high and un-normal traffic of network graph periodically.
Keywords
computer network security; inference mechanisms; microcomputers; network analysers; network theory (graphs); neural nets; real-time systems; telecommunication traffic; artificial neural network system; graph interpreting; hotspot area; input vector minimization; microcomputer; network administrator; network graph; network monitoring; real-time neural network-based network analyzer; traffic monitoring; Artificial neural networks; Classification algorithms; Computer crime; Forensics; Monitoring; Software; Training;
fLanguage
English
Publisher
ieee
Conference_Titel
Advanced Computer Science and Information System (ICACSIS), 2011 International Conference on
Conference_Location
Jakarta
Print_ISBN
978-1-4577-1688-1
Type
conf
Filename
6140744
Link To Document