Interactive visual sequence mining based on pattern-growth
Katerina Vrotsou, Aida Nordman · 2014
Sequential pattern mining aims to discover valuable patterns from datasets and has a vast number of applications in various fields. Due to the combinatorial nature of the problem, the existing algorithms tend to output long lists of patterns that often suffer from a lack of focus from the user perspective. Our aim is to tackle this problem by combining interactive visualization techniques with sequential pattern mining to create a "transparent box" execution model for existing algorithms. This paper describes our first step in this direction and gives an overview of a system that allows the user to guide the execution of a pattern-growth algorithm at suitable points, through a powerful visual interface.