I've been thinking about writing a blog entry on Hadoop, highlighting the benefits that it brings to the current data analysis framework adopted by most insurers. There are a number of different aspects that could be covered, including the basics of pig/hive programming, benefits of HDFS and mapreduce for data processing, and possible integration with R or Apache Mahout(an open source machine learning project).
I'd like to maintain focus on the insurance applications for these tools. If anyone has prior experience with Hadoop and would like to collaborate on this piece, I'll happily share what I have thus far.
Thanks