Discovering sequential patterns is an important problem in many data mining
Conventional algorithms proposed for this problem show
sequential patterns with items only. However, there are many cases
in which the quantity information can provide more useful insights to the
In this thesis, we introduce the problem of mining sequential patterns with
quantities. We propose the naive extension of traditional algorithms. Those
algorithms blindly enumerate the search space and result in bad performance.
To alleviate this problem, we propose the filtering and sampling techniques
that reduce the number of candidates for enumeration. To see the effectiveness
of our schemes, we implemented our algorithms and did a performance study.
The experimental result confirms that our schemes are much faster than
the traditional algorithms with naive extensions.