A clustering-based topic model using word networks and word embeddings

Mu, Wenchuan; Lim, Kwan Hui; Liu, Junhua; Karunasekera, Shanika; Falzon, Lucia; Harwood, Aaron

doi:10.1186/s40537-022-00585-4

Journal of Big Data

Table 9 Comparison of ClusTop algorithm against various baselines, in terms of Precision (Pre), Recall (Rec) and F-score (FS) for the top 20 keywords/unigrams of each topic

From: A clustering-based topic model using word networks and word embeddings

Algorithm	Top 20 Keywords/Unigrams
	Dataset A			Dataset B			Dataset C			Average
	Precision	Recall	F-score	Precision	Recall	F-score	Precision	Recall	F-score	Rank@20
ClusTop-Word-NA	.617 ± .002 (24)	.034 ± .000 (6)	.064 ± .000 (6)	.709 ± .003 (23)	.055 ± .000 (2)	.100 ± .001 (3)	.669 ± .009 (22)	.038 ± .001 (4)	.070 ± .002 (4)	(10.4)
ClusTop-BiG-NA	.646 ± .002 (21)	.035 ± .000 (3)	.065 ± .000 (3)	.722 ± .004 (22)	.056 ± .001 (1)	.102 ± .001 (1)	.679 ± .010 (20)	.040 ± .001 (2)	.073 ± .002 (2)	(8.3)
ClusTop-TriG-NA	.636 ± .002 (22)	.035 ± .000 (3)	.065 ± .000 (3)	.726 ± .004 (20)	.055 ± .001 (2)	.101 ± .001 (2)	.662 ± .009 (24)	.041 ± .001 (1)	.075 ± .002 (1)	(8.7)
ClusTop-BiHa-NA	.635 ± .002 (23)	.035 ± .000 (3)	.065 ± .000 (3)	.702 ± .004 (24)	.054 ± .000 (4)	.099 ± .001 (4)	.679 ± .009 (20)	.040 ± .001 (2)	.073 ± .002 (2)	(9.4)
ClusTop-Hash-NA	.831 ± .004 (9)	.024 ± .000 (23)	.045 ± .000 (18)	.873 ± .008 (9)	.031 ± .001 (17)	.060 ± .001 (17)	.911 ± .012 (8)	.023 ± .001 (17)	.044 ± .001 (17)	(15.0)
ClusTop-Noun-NA	.666 ± .002 (17)	.033 ± .000 (9)	.061 ± .000 (11)	.748 ± .003 (18)	.051 ± .000 (5)	.094 ± .001 (5)	.719 ± .010 (15)	.036 ± .001 (6)	.067 ± .001 (6)	(10.2)
ClusTop-H2VG-NA	.891 ± .004 (2)	.024 ± .000 (23)	.046 ± .000 (17)	.952 ± .005 (1)	.031 ± .001 (17)	.060 ± .001 (17)	.962 ± .008 (2)	.023 ± .001 (17)	.044 ± .001 (17)	(12.6)
ClusTop-H2VW-NA	.869 ± .001 (5)	.027 ± .001 (20)	.027 ± .001 (22)	.927 ± .001 (4)	.030 ± .001 (20)	.030 ± .001 (20)	.983 ± .001 (1)	.020 ± .001 (23)	.020 ± .001 (23)	(15.3)
ClusTop-H2VF-NA	.856 ± .001 (7)	.028 ± .001 (16)	.028 ± .001 (19)	.894 ± .001 (7)	.030 ± .001 (20)	.030 ± .001 (20)	.942 ± .001 (5)	.021 ± .001 (20)	.021 ± .001 (20)	(14.9)
ClusTop-Word-AH	.660 ± .002 (19)	.029 ± .000 (14)	.055 ± .000 (14)	.724 ± .004 (21)	.044 ± .001 (12)	.080 ± .001 (13)	.703 ± .010 (17)	.033 ± .001 (7)	.061 ± .001 (8)	(13.9)
ClusTop-Hash-AH	.757 ± .002 (10)	.033 ± .000 (9)	.063 ± .000 (8)	.801 ± .004 (13)	.044 ± .000 (12)	.082 ± .001 (11)	.836 ± .009 (12)	.026 ± .001 (15)	.050 ± .001 (15)	(11.7)
ClusTop-Noun-AH	.700 ± .002 (14)	.029 ± .000 (14)	.054 ± .000 (15)	.774 ± .004 (16)	.042 ± .000 (15)	.079 ± .001 (15)	.743 ± .010 (14)	.033 ± .001 (7)	.062 ± .001 (7)	(13.0)
ClusTop-H2VG-AH	.743 ± .002 (12)	.030 ± .000 (13)	.058 ± .000 (13)	.809 ± .003 (11)	.043 ± .000 (14)	.080 ± .001 (13)	.858 ± .009 (11)	.028 ± .001 (10)	.053 ± .001 (11)	(12.0)
ClusTop-H2VW-AH	.894 ± .001 (1)	.027 ± .001 (20)	.027 ± .001 (22)	.902 ± .001 (6)	.029 ± .001 (22)	.029 ± .001 (22)	.962 ± .001 (3)	.020 ± .001 (23)	.020 ± .001 (23)	(15.8)
ClusTop-H2VF-AH	.879 ± .001 (3)	.028 ± .001 (16)	.028 ± .001 (19)	.905 ± .001 (5)	.030 ± .001 (20)	.030 ± .001 (20)	.939 ± .001 (6)	.021 ± .001 (20)	.021 ± .001 (20)	(14.3)
ClusTop-Word-AM	.660 ± .001 (19)	.036 ± .000 (1)	.067 ± .000 (1)	.778 ± .004 (15)	.050 ± .000 (6)	.091 ± .001 (6)	.664 ± .012 (23)	.027 ± .001 (13)	.052 ± .001 (13)	(10.8)
ClusTop-Hash-AM	.691 ± .002 (15)	.033 ± .000 (9)	.062 ± .000 (9)	.796 ± .004 (14)	.047 ± .000 (9)	.087 ± .001 (9)	.745 ± .011 (13)	.030 ± .001 (9)	.058 ± .001 (9)	(10.7)
ClusTop-Noun-AM	.841 ± .001 (8)	.027 ± .000 (18)	.051 ± .000 (16)	.883 ± .003 (8)	.041 ± .000 (16)	.078 ± .001 (16)	.859 ± .009 (10)	.026 ± .000 (15)	.049 ± .001 (16)	(13.7)
ClusTop-H2VG-AM	.736 ± .002 (13)	.032 ± .000 (11)	.061 ± .000 (11)	.856 ± .003 (10)	.045 ± .000 (10)	.084 ± .001 (10)	.875 ± .008 (9)	.027 ± .000 (13)	.052 ± .001 (13)	(11.1)
ClusTop-H2VW-AM	.876 ± .001 (4)	.027 ± .001 (20)	.027 ± .001 (22)	.936 ± .001 (2)	.029 ± .001 (22)	.029 ± .001 (22)	.955 ± .001 (4)	.021 ± .001 (20)	.021 ± .001 (20)	(15.1)
ClusTop-H2VF-AM	.859 ± .001 (6)	.027 ± .001 (20)	.027 ± .001 (22)	.930 ± .001 (3)	.028 ± .001 (24)	.028 ± .001 (24)	.928 ± .001 (7)	.021 ± .001 (20)	.021 ± .001 (20)	(16.2)
LDA-Orig	.752 ± .002 (11)	.032 ± .000 (11)	.061 ± .000 (11)	.802 ± .003 (12)	.044 ± .000 (12)	.082 ± .001 (11)	.706 ± .010 (16)	.028 ± .001 (10)	.054 ± .001 (10)	(11.6)
LDA-Hash	.689 ± .002 (16)	.035 ± .000 (3)	.065 ± .000 (3)	.751 ± .003 (17)	.048 ± .000 (8)	.088 ± .001 (8)	.702 ± .010 (18)	.038 ± .001 (4)	.070 ± .001 (4)	(9.0)
LDA-Ment	.666 ± .002 (17)	.034 ± .000 (6)	.064 ± .000 (6)	.730 ± .004 (19)	.049 ± .000 (7)	.090 ± .001 (7)	.698 ± .012 (19)	.027 ± .001 (13)	.052 ± .001 (13)	(11.9)

The rank of an algorithm’s performance for each metric are provided in brackets

Back to article page