The Hadoop ecosystem has several other specialized tools and projects for data analysis that might be applicable for your use case. I'll mention a few popular open source tools to be considered.
R is a language created for statisticians and has over 2 million users. RHadoop is a set of R packages that run on Hadoop, which are as follows:
rmrpackage provides interface to MapReduce using R code
rhdfspackage provides interface to HDFS with R
rhbasepackage is required if HBase is used
plyrmrpackage has common data manipulation functions
ravropackage allows reading and writing to files in an
With RHadoop, data scientists can explore the Data Lake, perform regression analysis to predict the future, identify ...