PDF Publication Title:
Text from PDF Page: 175
Figure 7. Extended histogram for original NSL-KDD training dataset and a synthesized one generated from the same labels as the original. Regarding the second approach to check the similarity between original and synthesized features, in Table 1 we present the accuracy obtained in reproducing the three discrete features when using the NSL-KDD Training and Test dataset as original data. In each case, we compare the accuracy between the original dataset and a synthesized dataset composed of the same labels as the original. We see that the accuracy depends on the number of different values of the feature, and, in general, it is quite high. Option C offers the best results, providing a reconstruction accuracy for discrete features that is greater than 90% for most features. We can see that even when the much smaller NSL-KDD test dataset is used to train VGM, we still get very good accuracy results (the three columns on the right in Table 1). The values in Table 1 are color-coded; where the greenest is better and the redder is worse (comparison of values is applied column-wise). We base our definition of accuracy in the usually accepted one [20]. Table 1. Accuracy when reproducing discrete features We can see in Table 1 that the results differ for the column: flags. This is due to the fact that NSL-KDD has differences between the training and test datasets, as presented in Section 3.1. Doctoral Thesis: Novel applications of Machine Learning to NTAP - 173PDF 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)