Hi Jason,
First, congratulations on your book, Webwork in Action -- also from Manning.
Let me answer your question on the kinds of algorithms available. Within the area of collective intelligence, I classify intelligence into three main parts : explicit intelligence (e.g. recommendations, ratings), implicit intelligence (e.g., derived from unstructured text), and derived intelligence, which deals with applying machine learning algorithms to data.
The second part of the book deals with machine learning algorithms. Here, I cover the data mining process, build a text analyzing toolkit, clustering data and predictive analysis. Here, I cover WEKA, which is an open-source framework for data mining in Java and the Java Datamining (JDM) API's developed under JSR 73 and JSR 247. WEKA has an implementation for most of the data mining algorithms. The book provides code for leveraging WEKA and JDM in your application.
Also, there is a chapter on finding
patterns from data using clustering. Here, I work through an example of clustering blog entries using k-means and hierarchical clustering algorithms.
Chapter 12 is devoted to the process of building recommendation systems using both content-based and collaborative-based techniques. I also cover three real-world examples: Amazon personalization, Google News personalization, and Netflix.
thanks
Satnam