Yooreeka is a library for data mining, machine learning, soft computing, and mathematical analysis. The project started with the code of the book "Algorithms of the Intelligent Web".^[1] Although the term "Web" prevailed in the title, in essence, the algorithms are valuable in any software application.

It covers all major algorithms and provides many examples.

Yooreeka 2.x is licensed under the Apache License rather than the somewhat more restrictive LGPL (which was the license of v1.x).

The library is written 100% in the Java language.

Algorithms

The following algorithms are covered:

• Clustering
  • Hierarchical—Agglomerative (e.g. MST single link; ROCK) and Divisive
  • Partitional (e.g. k-means)
• Classification
  • Bayesian
  • Decision trees
  • Neural Networks
  • Rule based (via Drools)
• Recommendations
  • Collaborative filtering
  • Content based
• Search
  • PageRank
  • DocRank
  • Personalization