EPJ Data Science

Table 6 Most relevant words for each class in dataset SENS3

From: Analysis and classification of privacy-sensitive content in social media posts

Sensitive				Non-sensitive
Overall rank	Word	Overall count	Relative frequency	Overall rank	Word	Overall count	Relative frequency
13	home	43.50 ± 4.33	92.82 ± 4.94	24	peopl	33.00 ± 3.13	70.34 ± 7.09
15	tomorrow	38.30 ± 3.83	90.45 ± 4.21	9	one	48.60 ± 6.59	66.78 ± 7.59
29	tonight	31.20 ± 6.25	86.73 ± 3.92	11	love	45.60 ± 5.04	64.11 ± 5.28
30	weekend	30.50 ± 4.79	85.68 ± 4.12	19	think	34.60 ± 3.92	63.46 ± 5.11
4	work	56.90 ± 6.59	84.55 ± 5.39	22	dont	33.50 ± 5.76	61.47 ± 5.66
5	back	56.10 ± 5.69	80.52 ± 3.67	7	like	55.60 ± 8.86	61.36 ± 6.06
1	go	110.00 ± 10.58	73.62 ± 2.63	18	make	34.60 ± 4.01	58.45 ± 8.25
0	propnam	149.10 ± 12.51	73.12 ± 5.17	23	happi	33.30 ± 3.68	57.76 ± 8.16
26	night	32.10 ± 5.47	70.74 ± 11.19	27	know	32.10 ± 6.89	55.75 ± 6.38
10	today	45.70 ± 3.80	69.28 ± 4.51	14	new	40.00 ± 6.46	49.41 ± 6.27
20	got	34.50 ± 2.42	67.46 ± 7.07	16	good	37.90 ± 7.77	47.78 ± 4.71
21	come	33.80 ± 5.63	67.43 ± 6.09	17	want	36.10 ± 6.12	46.69 ± 6.84
6	im	55.90 ± 6.59	67.15 ± 4.28	28	feel	31.40 ± 5.76	42.22 ± 10.66
3	day	72.50 ± 6.75	67.09 ± 3.33	8	time	54.60 ± 7.09	41.16 ± 6.60
12	see	44.60 ± 7.00	63.17 ± 5.80	25	cant	32.60 ± 4.17	38.58 ± 5.07
2	get	87.20 ± 8.20	62.64 ± 4.44	2	get	87.20 ± 8.20	37.36 ± 4.44
25	cant	32.60 ± 4.17	61.42 ± 5.07	12	see	44.60 ± 7.00	36.83 ± 5.80
8	time	54.60 ± 7.09	58.84 ± 6.60	3	day	72.50 ± 6.75	32.92 ± 3.33
28	feel	31.40 ± 5.76	57.78 ± 10.66	6	im	55.90 ± 6.59	32.85 ± 4.28
17	want	36.10 ± 6.12	53.31 ± 6.84	21	come	33.80 ± 5.63	32.57 ± 6.09

Back to article page