自适应的混沌粒子群算法优化XML文档聚类策略

XML Document Clustering Strategy Based on Adaptive Particle Swam Optimization Algorithm with Chaos

为了提高海量XML文档集的聚类质量,提出了一种基于粒子群的XML自适应混沌聚类算法(简称ACPSO);为了简化XML文档相似性判定,该算法以XML键为基础,结合混沌原理与粒子群算法划分XML文档;为了加速算法的收敛性,通过对算法相关参数的自适应学习与权重调整,增强XML文档的全局寻优能力,改善XML文档聚类的质量。对比其它聚类算法,仿真表明本算法不仅能有效避免聚类停滞现象的发生,而且是一种高效的XML文档聚类方法。

To improve the clustering quality of massive XML document sets, an adaptive clustering algorithm with chaos of XML document sets was proposed based on PSO （ACPSO）. To predigest similarity judgment of XML document, the algorithm is based on xml keys, chaos principles and PSO to carve up XML documents by similarity distance. While enhancing the algorithm＇s astringency, with some parameters of organize-self clustering learning and power gene optimization, the algorithm swells the capability to seek the optimization in the full XML document, and improves the clustering quality of XML documents. Contrasted with other clustering algorithms, a series of emulation experiments show that the algorithm not only avoids bringing the stagnancy in clustering but also is an effective clustering method.