I am a SAS/STAT user in my company and our company has no budget for any expensive data mining packages like SAS EM or SPSS CLEMENTINE. Can someone suggest to me good open source data mining packges that are good for commercial usage? What I am trying to get is a package that is capable of handling SAS table with size of about 14 million records and about 10 variables.
Also, I know there are tons of data mining algorithms out there and I am totally new to this field. As I said, we use classical statistics methods in our work. Can someone recommend some resources that will help me learn these algorithms? For example, a book or an web site that explains to me when and what algorithm is good for what situaton. Thanks a lot.