EPJ Data Science

Table 8 Most relevant words for each class in WH+TW

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
295	lesbian	41.70 ± 6.36	97.49 ± 2.75	190	ni**a	58.80 ± 8.23	100.00 ± 0.00
357	bi	33.20 ± 6.63	97.43 ± 2.41	269	rt	44.60 ± 9.19	99.78 ± 0.69
91	chat	101.10 ± 7.82	96.59 ± 1.49	194	tweet	57.80 ± 7.08	99.24 ± 1.36
281	whisper	43.20 ± 7.39	95.85 ± 2.63	219	da	52.90 ± 11.61	98.94 ± 1.53
73	boyfriend	117.30 ± 15.94	95.19 ± 1.96	376	kno	30.60 ± 5.17	98.60 ± 2.53
142	male	71.00 ± 7.54	95.09 ± 3.00	169	twitter	63.50 ± 6.59	98.18 ± 1.50
182	relationship	60.70 ± 11.44	93.27 ± 2.87	349	snow	34.30 ± 5.31	97.63 ± 2.29
249	18	47.30 ± 6.38	92.92 ± 3.63	314	wat	38.70 ± 7.56	97.20 ± 2.80
218	ex	53.20 ± 6.36	92.50 ± 3.08	121	lmao	79.80 ± 6.88	97.03 ± 1.39
237	girlfriend	49.40 ± 6.20	92.17 ± 3.65	287	jus	42.40 ± 6.72	96.90 ± 2.79
62	sex	136.80 ± 13.17	91.40 ± 2.78	159	wit	66.40 ± 6.47	96.79 ± 2.14
381	attract	30.30 ± 7.53	91.30 ± 3.81	289	yea	42.20 ± 8.42	96.30 ± 3.01
113	femal	86.00 ± 8.96	90.98 ± 3.01	257	smh	46.40 ± 7.52	95.63 ± 3.38
364	older	32.00 ± 5.72	89.81 ± 4.24	174	bout	62.30 ± 6.38	94.98 ± 2.96
288	f	42.30 ± 7.09	88.64 ± 5.65	144	ya	70.90 ± 5.26	94.43 ± 3.17
157	messag	66.80 ± 8.04	87.81 ± 4.99	3	u	613.90 ± 31.07	93.99 ± 0.93
167	gay	64.40 ± 5.78	86.50 ± 5.51	66	ur	125.10 ± 17.70	93.37 ± 2.98
373	bf	30.80 ± 5.47	86.11 ± 6.56	185	yall	59.20 ± 4.32	93.30 ± 2.87
374	cheat	30.80 ± 9.10	84.84 ± 9.58	263	lil	45.60 ± 6.38	92.81 ± 3.51
327	secret	37.20 ± 7.45	84.56 ± 6.29	6	lol	520.20 ± 30.12	92.36 ± 1.08

Back to article page