Please use this identifier to cite or link to this item: http://dspace.dtu.ac.in:8080/jspui/handle/123456789/516
Full metadata record
DC FieldValueLanguage
dc.contributor.authorKANWAT, JITENDER-
dc.date.accessioned2010-12-03T06:41:55Z-
dc.date.available2010-12-03T06:41:55Z-
dc.date.issued2010-12-03-
dc.identifier.urihttp://dspace.dtu.ac.in:8080/jspui/handle/123456789/516-
dc.descriptionME THESISen_US
dc.description.abstractThe present research relates to the computer system user authentication, identification and security in a small to large multi-user environment. This presented program supports the automatic system administration to limit the access of the unauthorized person/program by tracking the uses of application software pattern in a network. In this program if suspicious program/activity/commands are happen then this limit the unauthorized person/program to disrupt system operations by using a neural network and set of rules to track the application software uses pattern and mark/identify suspicious activity on the system. Through this research we not only remove the dependency of user name and password for the user authentication but also track the suspicious activity within the network and prevent the user/network/company/organization through hacking .once we start the program in a network the neural network automatically train itself for the rapidly used application software patterns and We train the neural network beyond the tasks. Once we limit the tasks user never cross the boundaries it all depends on neural network’s learning process and duration of learning period where we teach the neural network in future we fix the work deadlines The main objective of this research work is to provide support to system administrators in limiting the ability of unauthorized user/ abnormal activities by monitoring the application software usage pattern and decision taking by the neural network. The present research work is a method that will prevent destructive programs, malicious software/ hacking tools from being executed. A combination of security rules and application software usages pattern are used to identify suspicious activities on a network/computer.en_US
dc.language.isoenen_US
dc.relation.ispartofseriesTD 715;-
dc.subjectNETWORKen_US
dc.subjectNEURALen_US
dc.subjectSECURITYen_US
dc.titleNETWORK SECURITY USING PATTERN BASED NEURAL NETWORKen_US
Appears in Collections:M.E./M.Tech. Computer Technology & Applications

Files in This Item:
File Description SizeFormat 
1.doc194 kBMicrosoft WordView/Open
2.doc1.25 MBMicrosoft WordView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.