The following is the implementation of the MINI algorithm, see here for the correct reference.

It is fairly inefficient (e.g. transposing the data can be optimized a lot), and although we have done our best to ensure correctness it is provided without any guarantees. It may be used research purposes only -- pls contact me if you would like to use it for other purposes.

The C++ source code.
This .sh file illustrates how to compile the code, and how to run it on the chess dataset. Note that it makes use of CHARM (but in principle other itemset miners can be used as well).