.png){kind=icon}
返回
Mining frequent closed inter-sequence patterns efficiently using dynamic bit vectors
.png){kind=icon}
PRE
AI
DOI：10.1007/s10489-014-0630-1.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Mining frequent sequences is a critical stage before rule generation for sequence databases. Currently, there are two main ways for mining frequent sequences, namely intra-sequence mining and inter-sequence mining. Inter-sequence mining is more attractive than intra-sequence mining because it considers the relationship between sequences in transactions. However, mining all possible frequent inter-sequences takes a long time and requires a lot of memory. Mining frequent closed inter-sequences is efficient because such sequences are compact, and only the necessary information is maintained. CISP-Miner was proposed for mining frequent closed inter-sequence patterns, but it consumes a lot of memory since many closed patterns are mined. This paper proposes an algorithm called ClosedISP for mining frequent closed inter-sequence patterns. The proposed algorithm uses a checking scheme for early eliminating and checking closed patterns without candidate maintenance. ClosedISP uses a dynamic bit vector that combines transaction information to compress data. In addition, ClosedISP adopts a prefix tree and a depth-first search order to reduce the search space and generate non-redundant sequential rules efficiently. Experiments were conducted to compare the proposed algorithm with CISP-Miner to demonstrate the effectiveness of the proposed algorithm in terms of runtime and memory usage.
Keyword:
Dynamic bit vector
Closed inter-sequence pattern
Vertical data format
AI总结
对已上传原文的论文进行重点信息的提取，主要内容包括：简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。
期刊
IF:
3.5
论文数:
7.6K
被引数:
1.7W
机构
引用论文
.png){kind=icon}
Index-BitTableFI: An improved algorithm for mining frequent itemsetsIndex-bittablefi: 一种改进的频繁项集挖掘算法
PREAI
分享收藏
Inherited arrhythmias and gene therapy: Are there any ethical considerations to take into account?
OAAI
OAAI分享收藏
查看更多内容