Egyptian Informatics Journal · 2013 · 192 citations · 17 references
ComputingComputer ScienceCybersecurityComputer SecurityData MiningNetwork SecurityAnomaly DetectionComputer IntrusionDenial-of-service AttacksIntrusion DetectionAccuracyComputer Intrusion DetectionDistributed Denial Of ServiceFalse Alarm Rate AnalysisFalse Alarm ReductionHybrid Intrusion DetectionNetwork Intrusion DetectionSemi-supervised Intrusion Detection
The rapid expansion of computer use and networked applications has heightened security concerns, exposing systems to vulnerabilities that threaten integrity, availability, and confidentiality, and underscoring the critical role of Intrusion Detection Systems in detecting anomalies and attacks. This study integrates data mining techniques with an IDS to efficiently uncover relevant hidden data for users while reducing execution time. To address classification, human interaction, unlabeled data, and DDoS effectiveness, the authors employ EDADT, a Hybrid IDS model, a semi‑supervised approach, and a Varying HOPERAA algorithm. Evaluated on the KDD Cup dataset, the proposed algorithms achieve higher accuracy and lower false‑alarm rates than existing methods.
With the tremendous growth of the usage of computers over network and development in application running on various platform captures the attention toward network security. This paradigm exploits security vulnerabilities on all computer systems that are technically difficult and expensive to solve. Hence intrusion is used as a key to compromise the integrity, availability and confidentiality of a computer resource. The Intrusion Detection System (IDS) plays a vital role in detecting anomalies and attacks in the network. In this work, data mining concept is integrated with an IDS to identify the relevant, hidden data of interest for the user effectively and with less execution time. Four issues such as Classification of Data, High Level of Human Interaction, Lack of Labeled Data, and Effectiveness of Distributed Denial of Service Attack are being solved using the proposed algorithms like EDADT algorithm, Hybrid IDS model, Semi-Supervised Approach and Varying HOPERAA Algorithm respectively. Our proposed algorithm has been tested using KDD Cup dataset. All the proposed algorithm shows better accuracy and reduced false alarm rate when compared with existing algorithms.
17
Intrusion detection using data mining techniques
Mohammadreza Ektefa, Sara Memar, Fatimah Sidi et al. · 2010 · 173 citations