In today's era, network security has become very important and a severe issue in information and data security. The data present over the network is profoundly confidential. In order to perpetuate that data from malicious users a stable security framework is required. Intrusion detection system (IDS) is intended to detect illegitimate access to a computer or network systems. With advancement in technology by WWW, IDS can be the solution to stand guard the systems over the network. Over the time data mining techniques are used to develop efficient IDS. Here,a new approach is introduced by assembling data mining techniques such as data preprocessing, feature selection and classification for helping IDS to attain a higher detection rate. The proposed techniques have three building blocks: data preprocessing techniques are used to produce final subsets. Then, based on collected training subsets various feature selection methods are applied to remove irrelevant & redundant features. The efficiency of above ensemble is checked by applying it to the different classifiers such as naive bayes, J48. By experimental results, for credit-gdataset, using discretize or normalize filter with CAE accuracy of both classifiers i. e. naive bayes & J48 is increased. For vote dataset, using discretize or normalize filter with CFS accuracy of the naive bayes classifier increased.

1. J. Gomez & D. Dasgupta, (2002), S. K. , and Peterson, L. L. 1993. Reasoning about naming systems.
2. Mr. Suraj S. Morkhade1, Prof. Mahip Bartere2,"Survey on Data Mining based Intrusion Detection Systems", International Journal of Application or Innovation in Engineering & Management (IJAIEM)
3. J. Gomez & D. Dasgupta, (2002) "Evolving Fuzzy Classifiers for Intrusion Detection", IEEEProceedings of the IEEE Workshop on Information Assurance, West Point, NY.
4. R. H. Gong, M. Zulkernine & P. Abolmaesumi, (2005) "A Software Implementation of a GeneticAlgorithm Based Approach to Network Intrusion Detection", Sixth InternationalConference on Software Engineering, Artificial Intelligence, Networking and Parallel/DistributedComputing and First ACIS International Workshop on Self-Assembling Wireless Networks.
5. Jiawei Han,Micheline Kamber,Jian Pei,"Data Mining : Concept and Techniques ", 3rd edition, Morgan Kaufmann,2011. ( 1st edition. ,2000-2001 )(2nd edition 2006).
6. H Liu andL Yu "Feature Selection for High-Dimensional Data – A Fast Correlation-Based Filter Solution", In Machine Learning-International Workshop Then Conference, 2003, Vol. 20(2), p. 856
7. P Langley, Selection of Relevant Features in Machine Learning, DefenseTechnical Information Center, 1994, pp. 140-144.
8. J Hua, WD Tembe, ER Dougherty, Performance of feature-selection methods in the classification of high-dimension data, Pattern Recognition, 2009, Vol. 42(3), pp. 409-424.
9. H Liu, H Motoda, L Yu,Feature selection with selective sampling,Machine Learning-International Workshop Then Conference, 2002, pp. 395-402.
10. Mark A. Hall, Lloyd A. Smith , "Practical Feature Subset Selection forMachine Learning", Computer ScienceDepartment, University of Waikato, Hamilton, New Zealand.
11. Margaret H. Danham,S. Sridhar, " Data mining,Introductory and Advanced Topics", Personeducation , 1st ed. , 2006
12. George Dimitoglou, James A. Adams, and CarolM. Jim," Comparison of the C4. 5 and a Naive BayesClassifier for the Prediction of Lung CancerSurvivability"
13. Aman Kumar Sharma, Suruchi Sahni, "AComparative Study of Classification Algorithms forSpam Email Data Analysis", IJCSE, Vol. 3, No. 5, 2011, pp. 1890-1895

Ids J48 Classifier Naive Bayes Classifier.