Table 3.

Comparison of the results on the lung cancer dataset using the proposed method of topic model-derived clustering based on feature selection and two conventional clustering methods of k-means and PCA.

Methods	k	Cluster ID	Adenocarcinoma	Squamous cell carcinoma	No. of misclassified samples	NMI
Topic model-derived clustering based on feature selection	2	1	42	11	22	0.2809
		2	11	47
	3	1	40	8	21	0.2417
		2	4	15
		3	9	35
	4	1	37	8	18	0.2926
		2	9	35
		3	0	14
		4	7	1
k-means	2	1	41	12	24	0.2461
		2	12	46
	3	1	8	35	31	0.1365
		2	27	17
		3	18	6
	4	1	6	14	25	0.1602
		2	22	6
		3	18	6
		4	7	32
PCA (10 features) + k-means	2	1	12	46	24	0.2461
		2	41	12
	3	1	8	35	31	0.1456
		2	22	6
		3	23	17
	4	1	16	5	25	0.1605
		2	6	14
		3	7	32
		4	24	7