PDF Publication Title:
Text from PDF Page: 012
2. THESIS FRAMEWORK The modality of this thesis is the compendium of publications in JCR-indexed journals in the telecommunications and data networking field. In Fig. 1 the objectives and scope of the thesis are presented. In Fig. 2 more details on how the different papers fit into the different areas of study are shown. Each paper fits in a different area. The first paper: “Review of methods to predict connectivity of IoT wireless devices” has been published in Ad Hoc & Sensor Wireless Networks. The second paper: “Network Traffic Classifier with Convolutional and Recurrent Neural Networks for Internet of Things” has been published in IEEEAccess in the Special Section on Big Data Analytics in Internet of Things and Cyber-physical Systems. The third paper: “Conditional Variational Autoencoder for Prediction and Feature Recovery Applied to Intrusion Detection in IoT” has been published in Sensors. The fourth paper: “Deep learning model for multimedia Quality of Experience prediction based on network flow packets” has been published in IEEE Communications Magazine with the feature topic of: “Enabling Technologies for Smart Internet of Things”. Finally, the fifth paper: “Variational data generative model for intrusion detection” has been published in Knowledge and Information Systems. Paper 1 [1] provides a thorough study of the activity of real IoT devices, which parameters affect it and how different algorithms can be used to predict it. The paper proposes a dataset preparation which fits classic time-series algorithms (ARIMA, HMM...) as well as other machine learning algorithms (non-time-series based: Random Forest, Logistic Regression...) which are not usually applied to time-series problems. In particular, as far as we know, it is the first time that a Random Forest algorithm is presented in the literature to deal with a time- series problem as the one treated in the paper. Paper 2 [2] studies the application of deep learning algorithms to detect the type-of-service of a network flow. It provides a comparison of results with other machine learning methods. It is, as far as we know, the first application of a CNN + RNN model to an NTC problem. Paper 3 [3] proposes a new solution to intrusion detection in data networks. The proposal is based on a generative model using a variant of a Variational Autoencoder (VAE) whose name is Conditional Variational Autoencoder (CVAE). This variant of a VAE provides many advantages and it is not only able to give better prediction results than other machine learning algorithms, but it is also able to generate new synthetic samples with feature values that have a Doctoral Thesis: Novel applications of Machine Learning to NTAP - 10PDF 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 | RSS | AMP |