I've read the paper titled "Molded Intrusion Detection enforcing RBF" which you sent the other day. If I were asked to review this paper with a figure on a scale of one to ten, it would be one. It would be worthless if we took no notice of authors' lack of basic knowlede on how a scientific paper should be described and also the lack of careful proof reading before the submission. First, we should know the fact that "attackers always learn the state of the art technique of intrusion detection." Therefore, the system against the data published in 1999 in a public domain would be nothing more than useless nowadays. Second, survey was real poor. The author claimed the system is made up of (1) feature reduction, (2) rule sets elaborated by GA, and (3) training by RBF. Nothing new. Especially the authors never descrived "intrusion detection with RBF in an enormous amount of those successful report of approaches by RBF. It might give innocent readers a false impression that the approach is authors' innovation. Third, the authors never mentioned a singularity of the KDD data set frequently reported so far. Again the lack of survey. And finally, detection rate the authors claimed as the results of their experiment is not so impressive. the feature reduction by mutual Information Correlation, framing the rule set based on genetic algorithm and the training done by Neural Network - Radial Basis Function The author should know the basic way of describing a scientific paper as well as a caful proof reading. (IDS, order of rererences, no year in reference, irregular indent) attacker always leanr new detection system so... "Experimental results found that Genetic Algorithm is highly successful in detecting known attacks, neural networks are found to be more effective to detect unknown attacks. The proposed methods outperform previous work in detecting both known and new attacks." "In this research work KDD 99 data sets have been used without modification." too old to use KDD 99 nevertheless few survey google rbf "intrusion detection" ‚ÌŒŸõŒ‹‰Ê –ñ 9,610 (1) -------------------------------------------------------------------------------------------------- Comparison of BPL and RBF Network in Intrusion Detection System ‘ЃVƒŠ[ƒY Lecture Notes in Computer Science o”ÅŽÐ Springer Berlin / Heidelberg ISSN 0302-9743 (Print) 1611-3349 (Online) Šª Volume 2639/2003 ‘Ð Rough Sets, Fuzzy Sets, Data Mining, and Granular Computing DOI 10.1007/3-540-39205-X ’˜ìŒ  2003 ISBN 978-3-540-14040-5 DOI 10.1007/3-540-39205-X_79 Page 581 Abstract In this paper, we present the performance comparison results of the backpropagation learning (BPL) algorithm in a multilayer perceptron (MLP) neural network and the radial basis functions (RBF) network for intrusion detection. The results show that RBF network improves the performance of intrusion detection systems (IDSs) in anomaly detection with a high detection rate and a low false positive rate. RBF network requires less training time and can be optimized to balance the detection and the false positive rates. (2) -------------------------------------------------------------------------------------------------- 2009 International Symposium on Information Engineering and Electronic Commerce Intrusion Detection Based on RBF Neural Network Radial Basis Function (RBF) has been one of the most common neural networks used in the intrusion detection system(IDS). To improve the approximation performance and calculation speed of RBF, we describe a method to deal with the benchmark datasets adopted in the research. It includes converting the string to numeric elements firstly, then omitting the unnecessary data and ensuring that the data has the reasonable range limit. The simulation results built upon Matlab software show that the RBF neural network has better performance than BP neural network. Ternopil, Ukraine May 16-May 17 ISBN: 978-0-7695-3686-6 ASCII Text x Jing Bi, Kun Zhang, Xiaojing Cheng, "Intrusion Detection Based on RBF Neural Network," Information Engineering and Electronic Commerce, International Symposium on, pp. 357-360, 2009 International Symposium on Information Engineering and Electronic Commerce, 2009. BibTex x @article{ 10.1109/IEEC.2009.80, author = {Jing Bi and Kun Zhang and Xiaojing Cheng}, title = {Intrusion Detection Based on RBF Neural Network}, journal ={Information Engineering and Electronic Commerce, International Symposium on}, volume = {0}, year = {2009}, isbn = {978-0-7695-3686-6}, pages = {357-360}, doi = {http://doi.ieeecomputersociety.org/10.1109/IEEC.2009.80}, publisher = {IEEE Computer Society}, address = {Los Alamitos, CA, USA}, } RefWorks Procite/RefMan/Endnote x TY - CONF JO - Information Engineering and Electronic Commerce, International Symposium on TI - Intrusion Detection Based on RBF Neural Network SN - 978-0-7695-3686-6 SP357 EP360 A1 - Jing Bi, A1 - Kun Zhang, A1 - Xiaojing Cheng, PY - 2009 KW - RBF network KW - Intrusion Detection KW - Network Security VL - 0 JA - Information Engineering and Electronic Commerce, International Symposium on ER - (3) -------------------------------------------------------------------------------------------------- An intrusion detection system based on RBF neural network Zhimin Yang Xiumei Wei Luyan Bi Dongping Shi Hui Li Dept. of Comput. Sci., Shandong Univ., Weihai, China; This paper appears in: Computer Supported Cooperative Work in Design, 2005. Proceedings of the Ninth International Conference on Publication Date: 24-26 May 2005 Volume: 2, On page(s): 873- 875 Vol. 2 ISSN: ISBN: 1-84600-002-5 INSPEC Accession Number: 8588002 Current Version Published: 2005-09-06 Abstract Based on the information system of Shanhua Company, this paper discusses the structure and function of intrusion detection system based on RBF, the steps and method of intrusion detection. In the experiment of network simulation, through continuous training of input normal samples and abnormal sample, keeping an eye on if the RBF neural networks can distinguish the known intrusion behavior character among the training samples with high exactness and distinguish new intrusion behavior character and mutation of known intrusion behavior character with some probability. The result of experiment proves that RBF network is better than BP network in its property of optimal approximation, classify ability and the rapidity of study, RBF can improve the detection (4) -------------------------------------------------------------------------------------------------- RBF-based real-time hierarchical intrusion detection systems Ju Jiang Chunlin Zhang Kamel, M. Dept. of Syst. Design, Waterloo Univ., Ont., Canada; This paper appears in: Neural Networks, 2003. Proceedings of the International Joint Conference on Publication Date: 20-24 July 2003 Volume: 2, On page(s): 1512- 1516 vol.2 ISSN: 1098-7576 ISBN: 0-7803-7898-9 INSPEC Accession Number: 7883930 Digital Object Identifier: 10.1109/IJCNN.2003.1223922 Current Version Published: 2003-08-26 Abstract An intrusion detection system (IDS) is an art to detect network intrusions by monitoring the network traffic patterns. Generally, an IDS uses only a single-layer detection structure; therefore it cannot adjust its structure adaptively and automatically. In this paper, two hierarchical IDSs, the serial hierarchical and parallel hierarchical IDSs, are proposed. Both of them are based on radial basis function (RBF) neural networks. Because of the short training time and high accuracy of the RBF neural networks, two hierarchical IDSs can monitor network traffic in real-time, train new classifiers for novel intrusions automatically, and modify their structures adaptively after new classifiers are trained. -------------------------------------------------------------------------------------------------- Computer Physics Communications Volume 180, Issue 10, October 2009, Pages 1795-1801 Font Size: Find more full-text articles: Your search for "rbf "intrusion detection" " would return 59 results on ScienceDirect. View Results Abstract Abstract - selected Article Figures/Tables Figures/Tables - selected References References - selected PDF (632 K) Article Toolbox E-mail Article Cited By Save as Citation Alert Citation Feed Export Citation Add to my Quick Links Cited By in Scopus (0) Related Articles in ScienceDirect Design and performance of an intelligent predictive con... Information Sciences Design and performance of an intelligent predictive controller for a six-degree-of-freedom robot using the Elman network Information Sciences, Volume 176, Issue 12, 22 June 2006, Pages 1781-1799 Ra?it Koker Abstract The aim of this paper was to propose a recurrent neural network-based predictive controller for robotic manipulators. A neural network controller for a six-joint Stanford robotic manipulator was designed using the generalized predictive control (GPC) and the Elman network. The GPC algorithm, which is a class of digital control method, requires long computational time. This is a disadvantage in real-time robot control; therefore, the Elman network controller was designed to reduce processing time by avoiding the highly mathematical and computational complexity of the GPC. The main reason for choosing the Elman network, amongst several neural network algorithms, was that the presence of feedback loops have a profound impact on the learning capability of the network. The designed neural network controller was able to recover quickly because of its significant generalization capability, which allowed it to adapt very rapidly to changes in inputs. The performance of the controller was also shown graphically using simulation software, including the dynamics and kinematics of the robot model. PDF (620 K) An integrated hybrid neural system for noise filtering,... Biochemical Engineering Journal An integrated hybrid neural system for noise filtering, simulation and control of a fed-batch recombinant fermentation Biochemical Engineering Journal, Volume 15, Issue 3, September 2003, Pages 165-175 P. R. Patnaik Abstract Fermentations employing genetically modified (recombinant) bacteria have complex metabolisms that are sensitive to the operating conditions. Therefore it is difficult to propose simple and accurate mathematical models suitable for industrial conditions. Under such a simulated condition, a fed-batch fermentation has been analyzed here for recombinant ƒÀ-galactosidase production by Escherichia coli containing the plasmid pOU140. A previous study had shown that representing the bioreactor by an Elman neural network and coupling this to neural networks to filter the disturbances selectively and to control the fermentation generated more ƒÀ-galactosidase than a well-mixed, noise-free fermentation. To overcome the limitations of large recurrent networks, the bioreactor has now been represented by a hybrid model which combines a partial mathematical model with two small Elman neural networks for the intra-cellular variables, an autoassociative network to filter the noise and a feedforward network as the controller. This integrated system permits more robust operation in the presence of disturbances and produces smoother profiles of the concentrations, improved gene expression and more ƒÀ-galactosidase than previously achieved with an Elman network alone. PDF (205 K) A modified particle swarm optimization-based dynamic re... Computers & Structures A modified particle swarm optimization-based dynamic recurrent neural network for identifying and controlling nonlinear systems Computers & Structures, Volume 85, Issues 21-22, November 2007, Pages 1611-1622 Hong-Wei Ge, Yan-Chun Liang, Maurizio Marchese Abstract In this paper, we first present a learning algorithm for dynamic recurrent Elman neural networks based on a modified particle swarm optimization. The proposed algorithm computes concurrently both the evolution of network structure, weights, initial inputs of the context units and self-feedback coefficient of the modified Elman network. Thereafter, we introduce and discuss a novel control method based on the proposed algorithm. More specifically, a dynamic identifier is constructed to perform speed identification and a controller is designed to perform speed control for Ultrasonic Motors (USM). Numerical experiments show that the novel identifier and controller based on the proposed algorithm can both achieve higher convergence precision and speed than other state-of-the-art algorithms. In particular, our experiments show that the identifier can approximate the USMfs nonlinear input?output mapping accurately. The effectiveness of the controller is verified using different kinds of speeds of constant, step and sinusoidal types. Besides, a preliminary examination on a randomly perturbation also shows the robust characteristics of the two proposed models. PDF (425 K) A feasible intrusion detector for recognizing IIS attac... Computers & Security A feasible intrusion detector for recognizing IIS attacks based on neural networks Computers & Security, Volume 27, Issues 3-4, May-June 2008, Pages 84-100 Shi-Jinn Horng, Pingzhi Fan, Yao-Ping Chou, Yen-Cheng Chang, Yi Pan Abstract Most activities on the Internet can be recorded as log files of websites and website administrators can inspect log files to locate problems after any network intrusion occurs. However, since log files usually contain a huge quantity of data, without effective methods, it is generally not feasible for administrators to determine the concealed meanings within log files. One method for dealing with this issue is to use neural networks; this is an effective means to distinguish and classify abnormal data in log files, thus alleviating the administrator's burden. This paper presents the results of a study on intrusion detection on IIS (Internet information services) utilizing a hybrid intrusion detection system (IDS). The feasibility of the hybrid IDS is validated based on the Internet scanner system (ISS). In the intrusion detection system proposed, we used four different training data sets: 200, 800, 1400, and 2000. The system is trained either by Taguchi's experimental design or full factorial experimental design under different training data sets; the former can save much more time than the latter. Under Taguchi's experimental design, the best results are obtained when the training data set is of size 1400; overall accuracy in this case is 97.5%. On the contrary, for the full factorial experimental design, the best results are reached when the training data set is of size 2000; overall accuracy is 97.6%. Our study indicates that when to retrain the detector and how much time to allow for this training fully depend on the downgrade percentage of the detection rate, which determines the size of the retraining data set. To reduce the void time for updating the detector, the downgrade percentage should be restricted. PDF (1253 K) Two-days ahead prediction of daily maximum concentratio... Atmospheric Environment Two-days ahead prediction of daily maximum concentrations of SO2, O3, PM10, NO2, CO in the urban area of Palermo, Italy Atmospheric Environment, Volume 41, Issue 14, May 2007, Pages 2967-2995 U. Brunelli, V. Piazza, L. Pignato, F. Sorbello, S. Vitabile Abstract Artificial neural networks are functional alternative techniques in modelling the intricate vehicular exhaust emission dispersion phenomenon. Pollutant predictions are notoriously complex when using either deterministic or stochastic models, which explains why this model was developed using a neural network. Neural networks have the ability to learn about non-linear relationships between the used variables. In this paper a recurrent neural network (Elman model) based forecaster for the prediction of daily maximum concentrations of SO2, O3, PM10, NO2, CO in the city of Palermo is proposed. The effectiveness of the presented forecaster was tested using a time series recorded between 1 January 2003 to 31 December 2004 in eight monitoring stations in urban area of Palermo (Italy). Experimental trials show that the developed and tuned model is appropriate, giving small values of root mean square error (RMSE) , mean absolute error (MAE) and mean square error (MSE). In addition, the related correlation coefficient ranges from 0.72 to 0.97 for each forecasted pollutant, underlying a small difference between the forecasted and the measured values. The above results make the proposed forecaster a powerful tool for pollution management systems. PDF (2311 K) View More Related Articles View Record in Scopus doi:10.1016/j.cpc.2009.05.004 Copyright c 2009 Elsevier B.V. All rights reserved. A research using hybrid RBF/Elman neural networks for intrusion detection system secure model This article is not included in your organization's subscription. However, you may be able to access this article under your organization's agreement with Elsevier. Xiaojun Tonga, , , Zhu Wangb and Haining Yua aSchool of Computer Science and Technology, Harbin Institute of Technology, Weihai, 264209, China bCollege of Information, Harbin Institute of Technology, Weihai, 264209, China Received 17 August 2008; revised 2 May 2009; accepted 6 May 2009. Available online 14 May 2009. Abstract A hybrid RBF/Elman neural network model that can be employed for both anomaly detection and misuse detection is presented in this paper. The IDSs using the hybrid neural network can detect temporally dispersed and collaborative attacks effectively because of its memory of past events. The RBF network is employed as a real-time pattern classification and the Elman network is employed to restore the memory of past events. The IDSs using the hybrid neural network are evaluated against the intrusion detection evaluation data sponsored by U.S. Defense Advanced Research Projects Agency (DARPA). Experimental results are presented in ROC curves. Experiments show that the IDSs using this hybrid neural network improve the detection rate and decrease the false positive rate effectively. Keywords: Intrusion detection; Hybrid RBF/Elman neural network; Memory of events; Anomaly detection; Misuse detection -------------------------------------------------------------------------------------------------- Intrusion Detection Based on Adaptive RBF Neural Network Full text Publisher Site Source ISDA archive Proceedings of the Sixth International Conference on Intelligent Systems Design and Applications - Volume 02 table of contents Pages: 1081 - 1084 Year of Publication: 2006 ISBN:0-7695-2528-8 Authors Jiang Zhong University of Chongqing, China Zhiguo Li University of Chongqing, China Yong Feng University of Chongqing, China Cunxiao Ye University of Chongqing, China Publisher IEEE Computer Society Washington, DC, USA Bibliometrics Downloads (6 Weeks): n/a, Downloads (12 Months): n/a, Citation Count: 0 Additional Information: abstract index terms collaborative colleagues Tools and Actions: Review this Article Save this Article to a Binder Display Formats: BibTeX EndNote ACM Ref ABSTRACT Recently the machine learning-based intrusion detection approaches have been subjected to extensive researches because they can detect both misuse and anomaly. In this paper, we propose a new method to design classifier based on multiple granularities immune network. Firstly a multiple granularities immune network (MGIN) algorithm is employed to reduce the data and get the candidate hidden neurons and construct an original RBF network including all candidate neurons. Secondly, the removing redundant neurons procedure is used to get a smaller network. Experimental results on the real network data set show that the new classifier has higher detection and lower false positive rate than traditional RBF classifier.