I am working on "Intrusion Detection System with Data mining Techniques". Can anyone help me with which algorithms/techniques are useful, or which keywords are good to find better research papers. Thanks
Your question is very broad.
I will suggest you to do some googlescholar searches for keywords like KDD99 dataset (https://tinyurl.com/3acekz7). The dataset is old, but most literature still use it as a benchmark. Related publications discuss about DM techniques. Perhaps that will give you some roadmap.
There was also a new dataset from University of New South Whales- UNSW-NB15 (https://tinyurl.com/znuxf4b), published at the ECU SRI Security Congress last week (http://conferences.secau.org/) at Perth WA.
Hope that helps
I suggest you also take a serious look at how you pre-process your data, as this will influence the performance different classifiers.
Feature selection , Feature extraction , Neural Networks , Statistical approach , Anomaly approach
Hi,
You can use principal components analysis (PCA) or partial least squares regression (PLS) these techniques are very used in data mining approach
Hi,
I have published two research papers on Intrusion Detection System for Cloud environment. You can read those two papers. It will give you an overall idea and expand your vision of research.
Thank you
Look at into the infrastructure Historian - Techniques such as feature extraction, data normalisation, machine learning and neural networks are an ideal approach.
I am very thankful to all of you for help. can anyone tell me on which tool i can do experiment and get reading ?
First get a solid idea about the data mining concepts related with building IDS.I recommend "Data Mining and Machine Learning in Cybersecurity" by Sumeet Dua, Xian Du for reading.Then come to the decision of the approach you are going to use based on below criteria,
1. IDS - signature,anomaly or hybrid based
2. Which Techniques - supervised,unsupervised or semisupervised
3. Dataset - e.g KDDCup 99
4.Algorithms based on 2
5.languages based on 4
https://www.crcpress.com/Data-Mining-and-Machine-Learning-in-Cybersecurity/Dua-Du/9781439839423
Read many of recent Cisco Talos , Cognitive blogs , they are from top class security researchers and scientists in the field working on real world data
http://blog.talosintel.com/
http://blogs.cisco.com/security/malware-stealing-gigabytes-of-your-data-as-seen-by-cognitive-threat-analytics
http://blogs.cisco.com/security/how-opendns-predicts-attacks-when-hacker-infrastructure-is-cheap-and-plenty
http://blogs.cisco.com/security/how-not-to-sample-network-traffic
Most of the industry blogs have been offlate very good , check from others like Cylance, SANS, etc too. they are very much latest and up to date .
KDD should not be used for any real research , it is quite old and heavily biased.
Check the below links for latest and realworld data.
http://www.mlsecproject.org/
http://www.secrepo.com/
- Badges
- Science topic
- Similar topics
- Computer Science
- Data Mining