Towards easy prototyping of pattern mining problems

In the last decade, plenty of algorithms, benchmarks, and experimental studies have been carried out for pattern mining problems. In this paper, we focus on the special class of pattern mining problems known to be ”representable as sets”. In this setting, the main contribution of this paper is to take advantage of the common theoretical background of these problems from an implementation point of view by proposing a library of efficient data structures and algorithms for pattern mining. Thus, every problem fulfilling the theoritical requirements could be implemented with only minimal effort. According to our first results, the programs obtained using our library offer a very good tradeoff between performances and simplicity of their development.