• KDnuggets: site for all things data mining related.
• Kaggle: organization that runs various data science competitions.
• Nine rules not to follow: common pitfalls when first attempting to tackle data mining problems.
• Video lectures on machine learning and data mining.

A fairly extensive list of software for machine learning and data mining can be found at KDnuggets.

Commonly used software for statistics, machine learning and/or data mining include:

• Commercial databases: Oracle, DB2, and SQL Server all have OLAP components that implement data mining algorithms.
• Dedicated commerical pacakges: dominated by SPSS and SAS (Enterprise Miner)
• MATLAB and its statistics toolbox.
• Python: pacakges like NumPy, SciPy, and scikit-learn.
• Weka: Java implementation of various machine learning algorithms
• KNIME: "Graphical workbench" for data analysis; has components for R, Python, and Weka integration

A fairly extensive list of data sets can be found at KDnuggets. This is especially useful if you don't already have a large data set for your term project or a smaller data set for a demo during your in-depth presentation (should you choose to run a demo).

Some of the more popular sources for data:

• KDD Cup Center: Data and tasks from past KDD Cup competitions
• UCI Machine Learning Repository: Source for "classic" data sets (iris, Boston housing, etc.)
• Google, Amazon, and InfoChimps: Sources for large, public data sets (but might take some effort to search)