Using Ensemble Methods for Improving Classification of the KDD CUP ’99 Data Set
Journal Title: IOSR Journals (IOSR Journal of Computer Engineering) - Year 2014, Vol 16, Issue 5
Abstract
Abstract: The KDD CUP ’99 data set has been widely used for intrusion detection and pattern mining in the last decade or so. Umpteen number of experiments pertaining to classification have been conducted on it.Many researchers have dedicated their resources for analysing this data set. But it has yet to be analysed by using Ensemble methods of classification. This paper contains experimental results obtained after classifying 10 % of the KDD CUP ’99 data set using ensemble methods like Bagging,Boosting and compares their performance with the standard J-48 classification algorithm.Weka experimenter has been used to classify the 494020 records using the aforementioned classifiers and the advantages of ensembling have been discussed in accordance with the obtained results..
Authors and Affiliations
Rohan D. Kulkarni
Keywords
Related Articles
- EP ID EP142279
- DOI 10.9790/0661-16535761
- Views 125
- Downloads 0
Opportunistic Networks: A Review
Abstract: Opportunistic network is a wireless ad hoc network. It is a type of delay tolerant network and an extension of mobile ad hoc networks. This network has a special characteristic which is intermittent connectivit...
Design and Development of Secure Electronic Voting System Using Radio Frequency Identification and Enhanced Least Significant Bit Audio Steganographic Technique
Abstract: Electronic decision making process has been adjudged as an alternative measure to address the flaws of ballot voting system for the delivery of free, fair, confident, credible and transparent elections. Electro...
Study of P2P Botnet
Abstract: Today, centralized botnets are still widely used. In a centralized botnet, bots are connected to several servers (called C&C servers) to obtain commands. This architecture is easy to construct and eff...
Analysis of Data Mining Tasks, Techniques, Tools, Applications And Trend
Data mining is a process which finds useful patterns from huge amount of data. It is a powerful new technology with great potential to help companies focus on the most important information in their data warehouses. It u...
Manual Unpacking Of Upx Packed Executable Using Ollydbg and Importrec
A 'Packer' is a compression routine that compress an executable file. Packers are used on executable for two main reasons: to shrink programs or to thwart detection or analysis. When malware has been pack...