Using Answer Set Programming for pattern mining
This work addresses pattern mining for data analysis, but it is incremental as it applies an existing declarative method to a known task with limited performance gains.
The paper tackled the problem of serial pattern mining by using Answer Set Programming (ASP) to implement both non-incremental and incremental resolutions, finding that the incremental version was more efficient but both were less efficient than dedicated algorithms.
Serial pattern mining consists in extracting the frequent sequential patterns from a unique sequence of itemsets. This paper explores the ability of a declarative language, such as Answer Set Programming (ASP), to solve this issue efficiently. We propose several ASP implementations of the frequent sequential pattern mining task: a non-incremental and an incremental resolution. The results show that the incremental resolution is more efficient than the non-incremental one, but both ASP programs are less efficient than dedicated algorithms. Nonetheless, this approach can be seen as a first step toward a generic framework for sequential pattern mining with constraints.