PDF Publication Title:
Text from PDF Page: 030
[76] KDD99 - It applies Restricted Boltzmann Machines (RBM) in anomaly detection using the KDD99 dataset, obtaining an accuracy of 94% on the training dataset and 83% on the test dataset. [77] KDD99 - A Deep Belief Networks (DBN) is used to train a classifier to detect intrusions, comparing the performance to SVM and ANN. DBN obtains the best accuracy of 93.49% which is achieved with a deep DBN of 122-150-90-50-5 (nodes per layer) plus a final backpropagation 2-layer neural network ended with a softmax (classification) layer. [78] NSL-KDD - For the 2-labels scenario, and using Naive Bayes with several feature engineering methods, they report an accuracy of 96.5% but the test set used is unclear. [79] NSL-KDD - They report an accuracy of 99.1% using several methods (SVM, NB...) with a previously performed dimensionality reduction on the features, but again, it is not clear the test set used, and the metrics are given on subsets of the anomaly types. [80] NSL-KDD - they report an accuracy of 99.9% with AdaBoost and a selection of features using a wrapper model; they use a subset of the NSL-KDD dataset for training and an unclear test set [81] NSL-KDD - For the 2-labels scenario and using AdaBoost with weak learners being simple decision stumps, they report a detection rate of 90% on test data. [82] KDD99 - The authors propose a least squares support vector machines (LS-SVM) based intrusion detection model, using kernel space approximation through greedy searching. They obtain a best detection rate of 0.98 [83] KDD99 -This paper proposes using optimized Regularized Least Square (RLS) classification combined with k- means clustering plus kernel approximation techniques. It obtains a best accuracy of 0.986 for DoS label detection in a one vs. rest detection mode. Table 1. Intrusion detection - related works Doctoral Thesis: Novel applications of Machine Learning to NTAP autoencoder provides better final accuracy than PCA, factor analysis and Kernel/PCA. - 28PDF Image | Novel applications of Machine Learning to Network Traffic Analysis
PDF Search Title:
Novel applications of Machine Learning to Network Traffic AnalysisOriginal File Name Searched:
456453_1175348.pdfDIY PDF Search: Google It | Yahoo | Bing
Cruise Ship Reviews | Luxury Resort | Jet | Yacht | and Travel Tech More Info
Cruising Review Topics and Articles More Info
Software based on Filemaker for the travel industry More Info
The Burgenstock Resort: Reviews on CruisingReview website... More Info
Resort Reviews: World Class resorts... More Info
The Riffelalp Resort: Reviews on CruisingReview website... More Info
CONTACT TEL: 608-238-6001 Email: greg@cruisingreview.com (Standard Web Page)