Research and Application of Clustering Algorithm for Text Big Data

Table 1

Comparison of the experimental results for several different clustering methods.

	The global K-means		Fast global K-means		The algorithm in this paper
	E	T (s)	E	T (s)	E	T (s)
Iris	78.9408	0.438	78.9451	0.078	78.9451	0.031
Wine	2.3707 × 10^6	0.765	2.707 × 10^6	0.094	2.3707 × 10^6	0.015
Soybean-small	96.3984	0.156	96.4702	0.047	96.4702	0
Segmentation	0.9398 × 10^6	3.765	1.0076 × 10^6	0.188	1.0198 × 10^6	0.125
Pima Indians diabetes	5.1363 × 10^6	3.625	5.1665 × 10^6	0.109	5.1363 × 10^6	0.047
Pen digits	9.9830 × 10^6	186.391	1.0480 × 10^7	6.844	1.0553 × 10^7	0.953