GEP-based Framework for Immune- Inspired Intrusion Detection

Tang, Wan

doi:10.3837/tiis.2010.12.017

Cited by 2 publications

(2 citation statements)

References 26 publications

(53 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The novel concept behind the linear chromosomes and the Expression Trees helps GEP to considerably overhaul prevailing adaptive algorithms. Therefore, GEP algorithms having new prospects for solving more complicated and logical problems [16]. The most important factors of GEP are the function set, terminal set, fitness function, control parameters, and stop condition which requires to be preset while using the GEP model to resolve a particular problem [17,18].…”

Section: B Gep Modellingmentioning

confidence: 99%

A Comparative Study of ANN and GEP Model to Predict the Pressure Drop in the Water Transportation System

Chakraborty¹,

Mandal²,

Barman³

2020

IJIEEB

View full text Add to dashboard Cite

In the present study, the parameter responsible to find out pressure drops in a pipeline network system has been modeled by Gene Expression Programming Based on the experimental data. The different factors like Pipe diameter, Particle diameter, liquid density, Solid density liquid Viscosity, Volume fraction, Velocity, Solid concentration are taken into consideration as the input parameter. GEP model was developed to predict the pressure drop within the pipeline system. GEP model predicts the pressure drop with an accuracy of mean R-Square 0.999153373.As the input parameter is responsible for the selection of soft computing method and both ANN and GEP model is considered in order to validate the output parameters. The result of GEP has been compared with an ANN model, to observe the level of accuracy of the predicted pressure drop with a correlation to predict pressure drop shown by equation 6. The obtained results of both GEP and ANN models are being compared and GEP predicted results are found to be better in predicting the output parameter. The mean absolute error is found to be 15.566 % by the ANN model wherein the GEP model predicts with an accuracy of 8.993 %.The results indicate that the GEP is better tool to predict pressure drop with more accuracy.

show abstract

Section: B Gep Modellingmentioning

confidence: 99%

A Comparative Study of ANN and GEP Model to Predict the Pressure Drop in the Water Transportation System

Chakraborty¹,

Mandal²,

Barman³

2020

IJIEEB

View full text Add to dashboard Cite

show abstract

“…This paper does not intend to suggest that there are no BDMs in existence that would detect the covert‐botnet. For instance, Tang et al present an immune‐inspired IDS that may be extended to prevent the covert channel described in this paper. This paper intends to suggest that BDMs should undergo an evasion evaluation as suggested in Gu et al .…”

Section: Related Work and Backgroundmentioning

confidence: 99%

Bot detection evasion: a case study on local‐host alert correlation bot detection methods

Shirley

Babu

Mano

2012

Security Comm Networks

View full text Add to dashboard Cite

Botnets have continuously evolved since their inception as a malicious entity. Attackers come up with new botnet designs that exploit the weaknesses in existing defense mechanisms and continue to evade detection. It is necessary to analyze the weaknesses of existing defense mechanisms to find out the lacunae in them. This research exposes a weakness found in an existing bot detection method (BDM) by implementing a specialized P2P botnet model and carrying out experiments on it. Weaknesses that are found and validated can be used to predict the development path of botnets, and as a result, detection and mitigation measures can be implemented in a proactive fashion. The main contribution of this work is to demonstrate the exploitation pattern of an inherent weakness in local-host alert correlation (LHAC) based methods and to assert that current LHAC implementations could allow pockets of cooperative bots to hide in an enterprise size network. This work suggests that additional monitoring capabilities must be added to current LHAC-based methods in order for them to remain a viable bot detection mechanism. This paper takes a forward-looking approach and presents a specific botnet development path based on a weakness evident in a successful BDM. This BDM will be defined as a local-host alert correlation (LHAC) based method. Further, this paper elaborates on the design and implementation of a different model for P2P botnet communication and describes the experiments conducted. The experiments demonstrate that the proposed botnet model can effectively hide from an LHAC-based BDM without having to rely on additional communication hiding tactics.It is often necessary to create a "proof-of-concept" that validates that a given vulnerability is exploitable. Therefore, this botnet model is tested against BotHunter [7]. BotHunter is a successful LHAC-based BDM. BotHunter is able to map a host's behavior to a malware infection life-cycle in order to flag compromised hosts as malware. The covert-botnet model tested in this work is able to circumvent BotHunter, and is implemented with this specific intent. The concept behind this model is extensible. Slight modifications to the covert-botnet framework would enable it to hide from any LHAC-based BDM that does not have the ability to account for all of a local host's network activity. Any botnet that is able to, somehow, circumvent the monitoring points employed by a BDM would be able to avoid detection in a similar manner.

show abstract

GEP-based Framework for Immune- Inspired Intrusion Detection

Abstract: This version proposes a GDP algorithm and includes a concrete analysis and supporting implementation results on GEP-IIDS.

Cited by 2 publications

References 26 publications

A Comparative Study of ANN and GEP Model to Predict the Pressure Drop in the Water Transportation System

A Comparative Study of ANN and GEP Model to Predict the Pressure Drop in the Water Transportation System

Bot detection evasion: a case study on local‐host alert correlation bot detection methods

Contact Info

Product

Resources

About