Work

An effective algorithm for mining sequential generators

Year: 2011

Type: conference-paper

Abstract: Mining frequent sequences patterns invokes the interests of many searchers. However, the result set of frequent sequences is tremendous. While frequent sequential generator patterns can compact the result set of frequent sequences greatly and are superior to the frequent sequential closed patterns in classification and model selection. The existing sequential generator patterns mining algorithm fails to make full use of the relationship between a sequence and its subsequence. And when mining the frequent sequential generators, the existing algorithms are ineffective and inefficient. In order to resolve this problem, an effective and efficient algorithm is presented. A lot of experiments about the performance on datasets are conducted. The results show that the algorithm proposed is better than the existing algorithms for mining the sequential generators.

Source: Procedia Engineering

DOI: 10.1016/j.proeng.2011.08.684

Authors: Shengwei Yi, Tianheng Zhao, Yuanyuan Zhang, Shilong Ma, Zhanbin Che

Institutions: State Key Laboratory of Software Development Environment, Beihang University

Language: English

Topic: Data Mining Algorithms and Applications

Subfield: Information Systems

Field: Computer Science

Domain: Physical Sciences

Open Access status: diamond

Funder: National Key Research and Development Program of China

Award: G5179843675