Abstract: Detecting anomalous behavior from terabytes of collected record data has emerged as a crucial component for many systems for data mining system.Very often，processing record data collected from various locations or providers cannot be directly aggregated for anomaly analysis due to the proprietary nature of the data.This paper proposes a novel general framework for anomaly detection from distributed data sources that cannot be directly merged.In the proposed method，anomaly detection algorithm is firstly applied to data from individual provider and then their results are combined.It investigates ten semi-supervised anomaly detection algorithms，as well as four methods for combining anomaly detection results.The experiments performed on simulated data have shown that particular anomaly detection algorithms and combining methods are more suitable for the task of distributed anomaly detection than others.

Key words: data mining, ensemble learning, distributed, anomaly detection

摘要： 研究了基于模型共享的集成学习分布式异常检测模型，采用多数投票、边界扩展、平均叠加和距离加权4种不同的集成学习方法得到全部的局部模型;采用交换本地数据挖掘模型的方式来实现数据共享，从而构造出一个总体的集成学习模型。从全局的观点检测异常，减少了集中式检测所需数据的传输量，有效保护了数据提供者的隐私性。仿真实验结果表明，该方法的检测性能与集中式检测的性能相当，甚至更好。

关键词: 数据挖掘, 集成学习, 分布式, 异常检测