Explainable Artificial Intelligence (XAI) to Enhance Trust Management in Intrusion Detection Systems Using Decision Tree Model

Mahbooba, Basim; Timilsina, Mohan; Sahal, Radhya; Serrano, Martin M.

doi:10.1155/2021/6634811

Cited by 150 publications

(71 citation statements)

References 41 publications

(35 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since this data set is relatively small as compared to the traditional data science settings we did not evaluate a Bayesian machine learning model [1,8,9]. In addition, we did not want to start with any independence or dependence relationships but wanted to mainly focus on associations since we did not have prior assumptions about any of the eleven features…”

Section: Methodsmentioning

confidence: 99%

Computational Design Map for Heterogeneous Experimental Studies

Kulkarni

Maisha

et al. 2021

Preprint

View full text Add to dashboard Cite

This paper focuses on the discovery of a computational design map of disparate heterogeneous outcomes from bioinformatics experiments in pig (porcine) studies to help identify key variables impacting the experiment outcomes. Specifically we aim to connect discoveries from disparate laboratory experimentation in the area of trauma, blood loss and blood clotting using data science methods in a collaborative ensemble setting. Trauma related grave injuries cause exsanguination and death, constituting up to 50% of deaths especially in the armed forces. Restricting blood loss in such scenarios usually requires the presence of first responders, which is not feasible in certain cases. Moreover, a traumatic event may lead to a cytokine storm, reflected in the cytokine variables. Hemostatic nanoparticles have been developed to tackle these kinds of situations of trauma and blood loss. This paper highlights a collaborative effort of using data science methods in evaluating the outcomes from a lab study to further understand the efficacy of the nanoparticles. An intravenous administration of hemostatic nanoparticles was executed in pigs that had to undergo hemorrhagic shock and blood loss and other immune response variables, cytokine response variables are measured. Thus, through various hemostatic nanoparticles used in the intervention, multiple data outcomes are produced and it becomes critical to understand which nanoparticles are critical and what variables are key to study further variations in the lab. We propose a collaborative data mining framework which combines the results from multiple data mining methods to discover impactful features. We used frequent patterns observed in the data from these experiments. We further validate the connections between these frequent rules by comparing the results with decision trees and feature ranking. Both the frequent patterns and the decision trees help us identify the critical variables that stand out in the lab studies and need further validation and follow up in future studies. The outcomes from the data mining methods help produce a computational design map of the experimental results. Our preliminary results from such a computational design map provided insights in determining which features can help in designing the most effective hemostatic nanoparticles.

show abstract

Section: Methodsmentioning

confidence: 99%

Computational Design Map for Heterogeneous Experimental Studies

Kulkarni

Maisha

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Mahbooba et al. [ 66 ] used XAI to enhance trust management in IDSs using a DT on the KDD dataset and a DT algorithm for decision making.…”

Section: Related Workmentioning

confidence: 99%

Classification and Explanation for Intrusion Detection System Based on Ensemble Trees and SHAP Method

Kim

Kang

et al. 2022

Sensors

View full text Add to dashboard Cite

In recent years, many methods for intrusion detection systems (IDS) have been designed and developed in the research community, which have achieved a perfect detection rate using IDS datasets. Deep neural networks (DNNs) are representative examples applied widely in IDS. However, DNN models are becoming increasingly complex in model architectures with high resource computing in hardware requirements. In addition, it is difficult for humans to obtain explanations behind the decisions made by these DNN models using large IoT-based IDS datasets. Many proposed IDS methods have not been applied in practical deployments, because of the lack of explanation given to cybersecurity experts, to support them in terms of optimizing their decisions according to the judgments of the IDS models. This paper aims to enhance the attack detection performance of IDS with big IoT-based IDS datasets as well as provide explanations of machine learning (ML) model predictions. The proposed ML-based IDS method is based on the ensemble trees approach, including decision tree (DT) and random forest (RF) classifiers which do not require high computing resources for training models. In addition, two big datasets are used for the experimental evaluation of the proposed method, NF-BoT-IoT-v2, and NF-ToN-IoT-v2 (new versions of the original BoT-IoT and ToN-IoT datasets), through the feature set of the net flow meter. In addition, the IoTDS20 dataset is used for experiments. Furthermore, the SHapley additive exPlanations (SHAP) is applied to the eXplainable AI (XAI) methodology to explain and interpret the classification decisions of DT and RF models; this is not only effective in interpreting the final decision of the ensemble tree approach but also supports cybersecurity experts in quickly optimizing and evaluating the correctness of their judgments based on the explanations of the results.

show abstract

“…In [58], the authors use decision tree algorithms to boost trust management using eXplainable Artificial Intelligence (XAI) for intrusion detection. Simple decision tree algorithms are applied to split the sub-choices for the intrusion detection system (IDS), which resemble a human approach to decision-making.Results show that the accuracy of the proposed approach is comparable with state-of-the-art algorithms.…”

Section: Securitymentioning

confidence: 99%

Machine Learning in Beyond 5G/6G Networks—State-of-the-Art and Future Trends

et al. 2021

View full text Add to dashboard Cite

Artificial Intelligence (AI) and especially Machine Learning (ML) can play a very important role in realizing and optimizing 6G network applications. In this paper, we present a brief summary of ML methods, as well as an up-to-date review of ML approaches in 6G wireless communication systems. These methods include supervised, unsupervised and reinforcement techniques. Additionally, we discuss open issues in the field of ML for 6G networks and wireless communications in general, as well as some potential future trends to motivate further research into this area.

show abstract

Explainable Artificial Intelligence (XAI) to Enhance Trust Management in Intrusion Detection Systems Using Decision Tree Model

Cited by 150 publications

References 41 publications

Computational Design Map for Heterogeneous Experimental Studies

Computational Design Map for Heterogeneous Experimental Studies

Classification and Explanation for Intrusion Detection System Based on Ensemble Trees and SHAP Method

Machine Learning in Beyond 5G/6G Networks—State-of-the-Art and Future Trends

Contact Info

Product

Resources

About