基于移動代理的層次優(yōu)化挖掘模型

對于大規(guī)模分布式數(shù)據(jù)挖掘問題，提出一種基于移動代理的層次結(jié)構(gòu)挖掘模型，該模型對OIKI DDM模型進行擴展，利用層次設(shè)計思想，基于移動代理和增量優(yōu)化技術(shù)進行挖掘和增量集成。實驗結(jié)果表明該模型對于數(shù)據(jù)站點大小具有更好的伸縮性，實現(xiàn)更加靈活，可根據(jù)網(wǎng)絡(luò)特點有效降低通訊代價，特別適合于大規(guī)模分布式環(huán)境。

For large-scale distributed data sets, a new hierarchical mining model based on mobile agent is proposed to perform distributed mining tasks. Based on hierarchical idea, the proposed approach integrates multiple local results using mobile agent and incremental optimization. The experimental results demonstrate that the proposed approach is scalable and particularly suited to large-scale distributed environments. In addition, the proposed model can reduce dramatically communication cost based on network characteristics.