FICW: Frequent itemset based text clustering with window constraint
Chong Zhou, Yansheng Lu, Lei Zou, Hu Rong · Wuhan University Journal of Natural Sciences · 2006
Most of the existing text clustering algorithms overlook the fact that one document is a word sequence with semantic information. There is some important semantic information existed in the positions of words in the sequence. In this paper, a novel method named Frequent Itemset-based Clustering with Window (FICW) was proposed, which makes use of the semantic information for text clustering with a window constraint. The experimental results obtained from tests on three (hypertext) text sets show that FICW outperforms the method compared in both clustering accuracy and efficiency.