Embedded Feature Selection Method for a Network-Level Behavioural Analysis Detection Model

Yusof, Mohammad Hafiz Mohd; Mokhtar, Mohd Rosmadi; Zain, Abdullah Mohd.; Maple, Carsten

doi:10.14569/ijacsa.2018.091271

Cited by 8 publications

(9 citation statements)

References 17 publications

(35 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, studies by [17] stated that the use of the Bayesian Network method may improve the result of prediction. Thus, this research is to extend the previous works in [8] which has established the Feature Selection Model. The model is utilized to obtain optimized features which are subsequently used in this proposed model.…”

Section: Literature Reviewmentioning

confidence: 99%

“…( 8) is utilized. This work has been published in previous work in [8]. Then, it will be used to extract optimized features from the ground truth dataset of the largest healthcare provider in Malaysia.…”

Section: Stage 2: Feature Selection and Density Distribution Function...mentioning

confidence: 99%

“…. Detailed discussion on Feature Selection Model is explained in paper [8]. If a network administrator observes these graphs over the available state-of-the-art, signaturebased CNMS (Centralized Network Monitoring System), they could conclude some normalcy over its distribution.…”

Section: Falsealarmrate Farmentioning

confidence: 99%

“…Studies have shown that existing network-level behavioral analysis methods suffered a few issues, for instance; lack of distribution modelling refinement processes, limited ground truth testing dataset, non-inferential analysis which leads to the inability to predict zero-day attacks, and produces high-level assumptions [4]. The reduction of millions of instances, disregarded parameters, removed similarities of most of the traffic flows to reduce information noise, insufficient number of optimized features, and ignored instances which do not involve an entity are amongst other problems that have been identified as the main issues contributing to the inability to predict zero-day attacks [8].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Behavioral Intrusion Prediction Model on Bayesian Network over Healthcare Infrastructure

Yusof¹,

Zin²,

Satar³

2022

Computers, Materials &Amp; Continua

View full text Add to dashboard Cite

Due to polymorphic nature of malware attack, a signature-based analysis is no longer sufficient to solve polymorphic and stealth nature of malware attacks. On the other hand, state-of-the-art methods like deep learning require labelled dataset as a target to train a supervised model. This is unlikely to be the case in production network as the dataset is unstructured and has no label. Hence an unsupervised learning is recommended. Behavioral study is one of the techniques to elicit traffic pattern. However, studies have shown that existing behavioral intrusion detection model had a few issues which had been parameterized into its common characteristics, namely lack of prior information (p (θ)), and reduced parameters (θ). Therefore, this study aims to utilize the previously built Feature Selection Model subsequently to design a Predictive Analytics Model based on Bayesian Network used to improve the analysis prediction. Feature Selection Model is used to learn significant label as a target and Bayesian Network is a sophisticated probabilistic approach to predict intrusion. Finally, the results are extended to evaluate detection, accuracy and false alarm rate of the model against the subject matter expert model, Support Vector Machine (SVM), k nearest neighbor (k-NN) using simulated and ground-truth dataset. The ground-truth dataset from the production traffic of one of the largest healthcare provider in Malaysia is used to promote realism on the real use case scenario. Results have shown that the proposed model consistently outperformed other models.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Section: Stage 2: Feature Selection and Density Distribution Function...mentioning

confidence: 99%

Section: Falsealarmrate Farmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Behavioral Intrusion Prediction Model on Bayesian Network over Healthcare Infrastructure

Yusof¹,

Zin²,

Satar³

2022

Computers, Materials &Amp; Continua

View full text Add to dashboard Cite

show abstract

“…Therefore, to further improve the predictive ability and generalization performance of the selected model, feature selection is carried out . The embedded method is employed to select the optimal features for GBR models. As for SVR, the mRMR approach is used for feature screening.…”

Section: Regression On E G Ssa and Cs Of Abo3-type Perovskitesmentioning

confidence: 99%

Multiobjective Stepwise Design Strategy-Assisted Design of High-Performance Perovskite Oxide Photocatalysts

Tao

Chang

et al. 2021

J. Phys. Chem. C

View full text Add to dashboard Cite

The rapid discovery of high-performance photocatalysts is one of the challenges in the field of photocatalysis research. Here, a multiobjective stepwise design strategy is developed to accelerate the design of potential ABO 3type perovskites with enhanced photocatalytic activity. The strategy includes model building, stepwise screening, and prediction of the hydrogen production rate (R H 2 ) of candidate perovskites. Based on the strategy, 35 candidate perovskites with a high specific surface area (SSA) (>60 m 2 g −1 ), a suitable bandgap (E g ) (1.4−2.6 eV), and a small crystallite size (CS) (<36 nm) are proposed. The predicted results of R H 2 of the candidates show they have high R H 2 (>6000 μmol h −1 g −1 ). The statistical analysis results reveal that a suitable E g is more likely to acquire when the main elements at the A site are Bi, La, and Pr, and those at the B site are Fe, Ti, and Mn. The perovskites prepared using the sol−gel, polymer complex, combustion method, and lower calcination temperature preferably have a higher SSA and a smaller CS. The relationship between target properties and R H 2 demonstrates that candidate perovskites with E g in the range of 1.9−2.4 eV, CS in the range of 20−32 nm, and a high SSA (>50 m 2 g −1 ) have a higher R H 2 . Furthermore, these models have been developed for online prediction applications.

show abstract

Discovery of Novel Photocatalysts Using Machine Learning Approach

Priyanga

Pransu

Krishna

et al. 2023

Machine Learning for Advanced Functional Materials

View full text Add to dashboard Cite

Embedded Feature Selection Method for a Network-Level Behavioural Analysis Detection Model

Cited by 8 publications

References 17 publications

Behavioral Intrusion Prediction Model on Bayesian Network over Healthcare Infrastructure

Behavioral Intrusion Prediction Model on Bayesian Network over Healthcare Infrastructure

Multiobjective Stepwise Design Strategy-Assisted Design of High-Performance Perovskite Oxide Photocatalysts

Discovery of Novel Photocatalysts Using Machine Learning Approach

Contact Info

Product

Resources

About