基于半结构特征分割的Web数据挖掘算法

Web Data Mining Algorithm Based on Semi Structure Feature Segmentation

提出一种基于半结构特征分割的Web数据挖掘算法.进行Web热点数据的信息流信号模型构建,对Web热点信息流进行包络特征分解,为了提高数据挖掘的纯度和抗干扰性能,采用前馈调制滤波器进行数据干扰滤波,采用半结构特征分割进行Web热点数据的特征提取,实现数据挖掘算法改进.仿真结果表明,采用该算法能提高对Web数据特征的检测性性能,数据挖掘中受到的旁瓣干扰较小,挖掘精度较高,性能优于传统算法.

A Web data mining algorithm based on semi structure feature segmentation is proposed.The information stream signal model of Web hot date is constructed and the characteristic erwelope decomposition of Web hot information stream is finished,in order to improve the purity of data mining and the anti-interference performance by feedforward filter modulation data interference filter,using semi structural feature segmentation for web hot number according to feature extraction.The data mining algorithm is realized.Simulation results show that the new algorithm can improve the detection capability of characteristics of Web data,data mining has little sidelobe interference,mining precision is high,performance is better than traditional algorithm.