Novel applications of Machine Learning to Network Traffic Analysis

PDF Publication Title:

Novel applications of Machine Learning to Network Traffic Analysis ( novel-applications-machine-learning-network-traffic-analysis )

Previous Page View | Next Page View | Return to Search List

Text from PDF Page: 141

misclassifications. The U2R label has the worst results, which is expected given the very low frequency of this label (1.77%). Table 1. Classification confusion matrix. Prediction DoS Normal Probe R2L U2R Total Percentage (%) 6295 119 368 4 6786 11860 2490 1081 327 22544 100.00% 30.10% 52.61% 11.05% 4.80% 1.45% 100.00% In Table 2 we present the empirical results to determine the position for inserting the labels in the decoder. The insertion in the second layer provides the best classification results, we can also observe that the position of insertion is an important element to consider when defining the architecture of the model. DoS 916 61 8917 610 162 24 7458 36 29 9711 18 21 2421 33.08% 43.08% 10.74% 11.33% Ground Truth Normal Probe 252 1762 1430 32 R2L 858 230 2554 0 345 25 7 23 400 1.77% U2R Total Percentage (%) Table 2. Impact of layer used in the decoder to insert the labels. 0.7791 0.7625 0.7888 0.8010 0.7908 0.8159 0.7547 0.7389 0.7584 0.7791 0.8010 0.7547 Labels inserted in first layer of decoder Model Accuracy F1 Precision Recall Labels inserted in second layer of decoder (ID-CVAE) Labels inserted in third layer of decoder All results presented in this Section (label frequency, confusion matrix, and performance metrics) are calculated using the full NSL-KDD test dataset 4.2. Feature Reconstruction ID-CVAE can perform feature reconstruction. Figure 7 shows the process that consists of three phases (in order): training, prediction and reconstruction phase. The objective here will be to reconstruct missing features in an intrusion detection test dataset, i.e., one with unknown labels. To do that, we start by training an ID-CVAE with the dataset with missing features (shown in Figure 7 as ⁡𝑿−𝑓), but using the complete dataset as reference (Figure 7, training phase). Doctoral Thesis: Novel applications of Machine Learning to NTAP - 139

PDF Image | Novel applications of Machine Learning to Network Traffic Analysis

PDF Search Title:

Novel applications of Machine Learning to Network Traffic Analysis

Original File Name Searched:

456453_1175348.pdf

DIY 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)