| 本文已被：浏览 4710次 下载 6174次 |
码上扫一扫！ |
|
|
| 挖掘多数据流的异步偶合模式的抗噪声算法 |
|
陈安龙1,2, 唐常杰1, 元昌安1,3, 彭京1, 胡建军1
|
|
1.四川大学,计算机学院,四川,成都,610065;2.电子科技大学,计算机科学与工程学院,四川,成都,610054;3.广西师范学院,信息技术系,广西,南宁,530001
|
|
| 摘要: |
| 挖掘多数据流的异步偶合模式是具有挑战性的工作.主要的研究工作包括:(1) 研究Haar小波滤波技术在挖掘流数据的异步偶合模式中的应用;(2) 引入小波系数序列来度量数据流的异步局域偶合度;证明了一系列定理,保证了度量方法的正确性;(3) 设计了环形滑动窗口和挖掘异步偶合模式的抗噪声增量算法,其时间复杂性小于O(n2);(4) 使用真实数据进行模拟实验,验证了算法的有效性. |
| 关键词: 多数据流 异步偶合模式 Haar小波 环形滑动窗口 |
| DOI： |
| 分类号: |
| 基金项目:Supported by the National Natural Science Foundation of China under Grant No.60473071 (国家自然科学基金); the National Research Foundation for the Doctoral Program of Higher Education of China under Grant No.20020610007 (国家教育部博士点基金) |
|
| An Anti-Noise Algorithm for Mining Asynchronous Coincidence Pattern in Multi-Streams |
|
CHEN An-Long,TANG Chang-Jie,YUAN Chang-An,PENG Jing,HU Jian-Jun
|
| Abstract: |
| Mining asynchronous coincidence pattern is a difficult task in multi-data streams. The main contributions of this work included: (1) The filter technique of Haar Wavelet is investigated and applied to mining asynchronous coincidence pattern in multi-streams; (2) The Wavelet coefficient series are applied to the measurement of asynchronous coincidence between data streams. A series of theorems are proved to ensure the validity of measuring asynchronous coincidence; (3) The anti-noise increment algorithms are designed on loop sliding windows to mine asynchronous coincidence pattern and implemented with complexity O(n2); (4) The extensive experiments on real data are given to validate algorithms. |
| Key words: multi-data stream asynchronous coincidence pattern Haar wavelet loop sliding window |
