Data Mining Technology Application in False Text Information Recognition

Table 3

Feature sequence after feature selection.

Order	Feature name (feature number)
1	Number of Chinese characters (1)
2	Total number of characters (2)
3	different words (6)
4	Number of words (5)
5	Non-Chinese characters (4)
6	Auxiliary word (U) (73)
7	String punctuation (W) (76)
8	Numeral (m) (68)
9	“.” (11)
10	Adverb (d) (63)
11	Noun (n) (60)
12	Average sentence length (9)
13	Verb (V) (61)
14	“,” (10)
15	Temporal words (nt) (65)
16	Of (30)
17	Adjective (a) (62)
18	Total number of sentences (80)
19	Voluntary verb (vu) (78)
20	Past tense marker (33)
21	“;” (15)
22	Hapax dislegomena (8)
23	Preposition (p) (71)
24	Hapax legomena(7)
25	“/” (16)
26	Patient (84)
27	Pronoun (r) (66)
28	Quantifier (q) (70)
29	Conjunction (C) (72)
30	As soon as (43)
31	All (42)
32	Total number of numeric characters (3)
33	Descriptive word (vl) (64)
34	get (31)
35	Treatment (92)
36	Effect (91)
37	Also (40)
38	Improve (90)
39	“:” (14)
40	Live (34)
41	Than (24)
42	Make (27)
43	Side effect (83)
44	“ ” ” (17)
45	features (85)
46	Safety (88)
47	From (22)
48	Interjection (e) (74)
49	Four-word phrases (i) (79)
50	Be (19)
51	Relapse (94)
52	Period (87)
53	Like (26)
54	Effective (95)
55	“!” (13)
56	Locative words (nS) (67)
57	Symptom (82)
58	Prefix (nh) (75)
59	Auxiliary (used after an adverbial) (32)
60	Then (52)
61	Health (89)
62	Significant (93)
63	rare word (x) (77)
64	Treatment (81)
65	After (44)
66	“?” (12)
67	Later (50)
68	Include (25)
69	As (23)