大型Web站点逻辑域挖掘算法

Large Scale Website Logical Domain Mining Algorithm

通过进一步发展Wen-SyanLi等人提出的Web站点逻辑域理论,该文提出Web站点逻辑域核模型及建立在其上的逻辑域挖掘算法。该算法通过对Web站点超链接的图结构进行运算,得到Web站点逻辑域。与Wen-SyanLi算法对比测试,结果表明在获得相同逻辑域个数的情况下,克服了其采用启发式方法所带来的效率问题。在对4个大型Web站点的单独测试中,平均能够达到85%的逻辑域挖掘精度。

By developing Wen-Syan Li＇s website logical domain theory, the paper proposes a website logical domain core model and logical domain mining algorithm based upon it. The algorithm computes website＇s hyperlink graph structure to obtain its logical domain. In comparative test with Wen-Syan Li＇s algorithm, it overcomes the efficiency defect of Wen-Syan Li＇s huristic method while obtaining the same quantity of logical domain. In separate test of 4 large scale websites, the logical domain core mining precision can averagely reach 85%.