Data mining

This page is a place-holder for data mining and machine learning, as applied to computer forensics. Right now it is a list of algorithms and resources that will be useful for people doing forensic data mining.

Algorithms

• https://en.wikipedia.org/wiki/DBSCAN - DBScan clustering algorithm.
• https://en.wikipedia.org/wiki/Okapi_BM25 - Okapi BM25 - A ranking function used by search engines.
• https://en.wikipedia.org/wiki/Soundex - Soundex, for doing "sounds like" searching.

Tools

• Weka data mining toolkit - java, has programmatic and GUI interface.
• Ping He has created an Open Source C4.5 implementation in C
• Apache Mahout: goal is to "build scalable, Apache licensed machine learning libraries" (java). also includes a focus on using hadoop.
• The Journal of Machine Learning maintains an archive of non-trivial machine learning algorithms, toolboxes, and languages.