I want to apply deep neural network for NIDS and when I use MATLAB, uploading the test and train data is easy. The question is how to use Deep neural Network?
Thanks
Different from previous review studies, we present more descriptive statistics to evaluate published studies in focused period. Therefore, the following statistics has been extracted from the reviewed 149 studies: 1. Classification output classes PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.1954v1 | CC-BY 4.0 Open Access | rec: 14 Apr 2016, publ: 14 Apr 2016 10 Atilla Ozg¨ur, Hamit Erdem ¨ 2. Training and Testing Dataset Usage 3. Cross Validation 4. Dataset sizes used in training and testing machine learning algorithms 5. Applied algorithms in proposed method 6. Classifiers used for comparison 7. Software Toolbox Usage 8. Other Datasets used in Reviewed Studies 9. Performance Metrics used in Experiments 10. IDS vs Not IDS 11. Main IDS Type according to study These descriptive statistics are presented using figures and tables and has been discussed in detail.
Although KDD99 dataset is more than 15 years old, it is still widely used in academic
research. To investigate wide usage of this dataset in Machine Learning Research (MLR)
and Intrusion Detection Systems (IDS); this study reviews 149 research articles from 65
journals indexed in Science Citation In- dex Expanded and Emerging Sources Citation
Index during the last six years (2010–2015). If we include papers presented in other
indexes and conferences, number of studies would be tripled. The number of published
studies shows that KDD99 is the most used dataset in IDS and machine learning areas,
and it is the de facto dataset for these research areas. To show recent usage of KDD99 and
the related sub-dataset (NSL-KDD) in IDS and MLR, the following de- scriptive statistics
about the reviewed studies are given: main contribution of articles, the applied algorithms,
compared classification algorithms, software toolbox usage, the size and type of the used
dataset for training and test- ing, and classification output classes (binary, multi-class). In
addition to these statistics, a checklist for future researchers that work in this area is
provided.
- Badges
- Science topic