Twitter ( ), ( ). i
|
|
- けいしょう ちゅうか
- 5 years ago
- Views:
Transcription
1 ( : A9TB2251)
2 Twitter ( ), ( ). i
3 A 25 ii
4 1 SNS( ) SNS SNS Twitter( ) SNS Twitter 140 SNS 1.4 ( ) Twitter 140 Twitter, Twitter Twitter [9, 10] HP 2 [1, 2] [3, 4] [6], 1
5 [7] Twitter [5] ( )
6 5 6 3
7 2 2.1 [1] Twitter [4] Twitter : RT RT URL web URL # # URL URL 4
8 URL URL, 2.2 [2] [1], [7] Twitter Twitter 5 Score = l(n ) n( ) (2.1) 8 5
9 3 2.3 [2] SVM URL Carlos [8] Carlos Twitter URL URL 6
10 3 3.1 (2011/3/11 9:00 201/3/18 9:00) ( 1 8 ) 3.2 [5].., 2.1 ( ) 7
11 3.1: ( ) ONEPEACE w 2.1 D P (w D w) = P (w D) P (w) = w w (3.1). 2.2 Score(s, t) = P MI(t, w) hist(len s t) (3.2) w C s w t PMI( ) s t Cs s PMI 2 lens s hist(l, t) s t l 14 ( 3.1) ( ) ( AND ) 8
12 1 3.3 ( ) ( ) ( ) Web URL ( ) ( ) ( 3.1) x x 9
13 [1, 2, 4, 5, 6, 7] 10
14 4, (TTC: Time To Correction) (TTS: Time To Suppress) (TTE: Time To End)
15 4.1: TTC( ) TTS( ) TTE( ) ONEPEACE ( )
16 4.1: 4 4.2: 4 ( ) 100 ONEPEACE
17 , 14
18 [5]. 5.1 [2, 5, 6, 7] (T): 121 ( A) Bag of words (B): ( ) URL (U): web URL URL 15
19 (R): (TW): 5 (D): ( ) ( ) 2.1 ( ) (SU SB): ( SU SB) ( 3.1) True Positive True Negative, False Positive False Negative ( 5.1) Precision, Recall, F1 5.1, 5.2, 5.3 P recision class = T ruep ositive class T ruep ositive class + F alsep ositive class (5.1) Recall class = T ruep ositive class T ruep ositive class + F alsenegative class (5.2) F 1 class = 2 P recision class Recall class P recision class + Recall class (5.3) 16
20 5.1: True Positive False Positive False Negative True Negative Accuracy ( 5.4) Precision Recall F1 ( 5.5, 5.6, 5.7) Accuracy = (T P + T N ) + (T P + T N ) + (T P + T N ) 3 AllT weet (5.4) P recision = P recision + P recision + P recision 3 (5.5) Recall = Recall + Recall + Recall 3 (5.6) F 1 = F 1 + F 1 + F 1 3 (5.7) (T) F1 ( 8 ) F ( ) Bag of words (7 ) F Bag of words URL 0.81, F URL (U) Bag of words 17
21 5.2: Accuracy Precision Recall F1 T( ) B(Bag of words) SB( ) SU( ) D( ) TW( ) (B ) T( ) R(RT ) U(URL ) Bag of Words (TW) URL, Web F1 F1 18
22 5.3: (Bag of Words 7 ) 2.73 (SU) 1.25 (T=True) 2.05 (SB) 0.79 (TW[0]= ) 0.82 (T=False) 0.69 URL (U=True) 0.71 (R=Ture) (R=False) (T=False) (T=True) (TW[1]= ) (TW[2]= ) (SU) 5.4: Precision Recall F1 T( ) ( Bag of Words) T( ) ( Bag of Words) T( ) ( Bag of Words) ( B(Bag of words))
23 5.1: 5.3 URL cc RT
24 URL RT ( ) ( ) ( ) 21
25 6 Bag of words URL URL 22
26 Twitter Japan 23
27 [1],,,,, Vol54, [2],,,, 18, [3],,,,, 10, [4],,,, Twitter RT,, [5],,,,, 19, [6],,,,,, Twitter, 26, [7],,,, Vol3, [8] Carlos Castillo, Marcelo Mendoza, Bardara Poblete, Insormation Crebidility on Twitter, International World Wide Web Conference, 2011 [9],,, [10] (ispp), 3.11,,
28 A A.1: 25
Twitter‡Ì”À‰µ…c…C†[…g‡ðŠŸŠp‡µ‡½…^…C…•…›…C…fi‘ã‡Ì…l…^…o…„‘îŁñ„�™m
27 Twitter 1431050 2016 3 14 1 Twitter,,.,.,., Twitter,.,,.,,. URL,,,. BoW(Bag of Words), LSI(Latent Semantic Indexing)., URL,,,,., Accuracy, AUC(Area Under the Curve), Precision, Recall, F,. URL,,,.,
More information,,, Twitter,,, ( ), 2. [1],,, ( ),,.,, Sungho Jeon [2], Twitter 4 URL, SVM,, , , URL F., SVM,, 4 SVM, F,.,,,,, [3], 1 [2] Step Entered
DEIM Forum 2016 C5-1 182-8585 1-5-1 E-mail: saitoh-ryoh@uec.ac.jp, terada.minoru@uec.ac.jp Twitter,, Twitter,,, Bag of Words, Latent Semantic Indexing,.,,,, Twitter,, Twitter,, 1. SNS, SNS Twitter 1,,,
More informationi
2012 2013 3 29 (: A9TB2133) i 1 1 2 5 2.1.............................. 5 2.2.............................. 6 3 1-7 3.1................................. 7 3.2.................................. 8 3.3.................................
More information=
2. 2.1 2.2 kuri@ice.uec.ac.jp ( 2007/10/30/16:46) 1 . 1. 1 + 2 = 5. 2. 180. 3. 3 3. 4.. 5.. 2 2.1 1.,,,,. 2., ( ) ( ).,,,, 3.,. 4.,,,. 3 1.,. 1. 1 + 2 = 5. (, ) 2. 180. (, ) 3. 3, 3. (, ) 4.. (, ) 5..
More informationアウトドアフェス2015企画書 公開版0811
2014 1 2,000/2,500() 2 3,000/3,500() 2900 5000 500 ID HP HP SNS 8 30 MTB 100 1921 WEB WEB WEB 91 60@1 500 300 4233000012 .. 1 350 500 2 350 500 11700 20034 () () 1 9:30 10:00 10:10 11:00 11:30 12:00
More informationTwitter 2016 3 201413127
Twitter Twitter Twitter Twitter 2 2 Twitter Twitter Twitter SVM(Support Vector Machine) Distant Supervision 2 1000 F. Twitter 2016 3 201413127 1 1 1.1....................................... 1 1.2.......................................
More informationuntitled
V http://ac-intelligence.jp/mystery/index.html HPhttp://ac-intelligence.jp/ http://blog.mag2.com/m/log/0000153671/ 2 ...4 155 800...10 155...10... 11...13...15 800...17...19 V 1,100...22...22...23...25...28...32...34...36...38...41...43...46...47...47...47...48
More informationuntitled
0120-888-089 0120-919-498 160-83381-26-1TEL. 03-3349-3111 URL http://www.sjnk.co.jp/ 160-83381-26-1 TEL. 03-3349-3111 URL http://www.sompo-japan.co.jp/ 100-89653-7-3 TEL. 0120-919-498 URL http://www.nipponkoa.co.jp/
More information表紙.PDF
2 3 4 2001 2010 33 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 H18 21 22 23 24 25 26 27 28 29 30 31 32 33 300 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 410-8601 055-934-4747 Fax 055-933-1412
More information農林金融2015年5月号
4 5 4 4 4 4 2014 5 14 7 2050 1412 15 3 19 1800 1 1 2 2 2 126 2 3 1 2 1 2 3 4 1 1 2 2 1 1 2014 3 2 126 2 2 3 1 40 30 20 10 0 2014 26 287 339 393 399 361 2 2 11 1 3 25 4 4 5 5 19898993 96 3 2000 1 1 1 5
More information4-3-1-2 2) 4-2 1974 10 824 46,000 4 8 4-2 1974 10 4-3-1-3 1) 2002 3 4-3-1-4 3) 4-3 2004 12 104 4-3 2004 12 26
4-1 3 4-2 WEB WEB 4-3 4-3-1 4-3-1-1 1) 4-1 9 50 7 4-1 25 4-3-1-2 2) 4-2 1974 10 824 46,000 4 8 4-2 1974 10 4-3-1-3 1) 2002 3 4-3-1-4 3) 4-3 2004 12 104 4-3 2004 12 26 4-3-1-5 2) 4-3-1-6 WEB 16 8 11 4-3-1-7
More information橡matufw
3 10 25 3 18 42 1 2 6 2001 8 22 3 03 36 3 4 A 2002 2001 1 1 2014 28 26 5 9 1990 2000 2000 12 2000 12 12 12 1999 88 5 2014 60 57 1996 30 25 205 0 4 120 1,5 A 1995 3 1990 30 6 2000 2004 2000 6 7 2001 5 2002
More informationO
11 2 1 2 1 1 2 1 80 2 160 3 4 17 257 1 2 1 2 3 3 1 2 138 1 1 170 O 3 5 1 5 6 139 1 A 5 2.5 A 1 A 1 1 3 20 5 A 81 87 67 A 140 11 12 2 1 1 1 12 22 1 10 1 13 A 2 3 2 6 1 B 2 B B B 1 2 B 100 B 10 B 3 3 B 1
More informationuntitled
0-1 1 1 2 2 3 3 4 4 5 5 6 6 7 7 8 8 9 9 10 10 11 11 12 12 13 13 14 14 15 15 16 16 17 17 18 18 19 19 20 20 21 21 22 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 0-2 1 1 2 23 3 46 4 714 5 15 0-3
More informationルール&マナー集_社内版)_修正版.PDF
WWW(World Wide Web) Web 12 WWW ID 2 1 2 3 4 WWW World Wide Web 5 5 A B 11 http://www.enc.or.jp/enc/code/rule/main.html 12 3 ... 2 1... 5 1.1... 5 1.2... 5 1.3... 6 1.4... 7 2... 9 2.1... 9 2.2 ID... 10
More informationVol..3 2010 2 10 2
1 Vol..3 2010 2 10 2 Vol..3 2010 2 10 3 Vol..3 2010 2 10 4 Vol..3 2010 2 10 5 Vol..3 2010 2 10 6 Vol..3 2010 2 10 7 Vol..3 2010 2 10 8 Vol..3 2010 2 10 9 Vol..3 2010 2 10 10 Vol..3 2010 2 10 11 Vol..3
More information09030549_001.図書館31-1
vol.31 NO.1 2 3 5 6 10 12 8 1947-1954- 1950-1838-1904 1952-1996 1 913-1954 1981-1958- 1 902-1992 1972-1946- 1 940- 1 2 3 3 5 6 6 8 8 12 8 8 1 2 2 http://library.hokkai-s-u.ac.jp/cgi-bin/tosyokan/index.cgi
More informationみさき_1
2 3 4 5 6 7 1F 2F 8 9 10 11 17 18 19 20 21 22 23 24 31 25 26 27 28 29 30 8 1 2 3 4 5 6 7 6 8 7 16 7 8 9 10 11 12 14 15 13 17 23 Vol.41 8 6 20 11 7 15 7 23 7 7 7 16 23 23 8 13 18:00 22:00 722
More informationH21_report
vol.4 1 2 6 10 14 18 20 22 24 25 1 2 2172 73 3 21925 926 21125 126 4 5 6 21629 630 7 21107 108 21127 128 8 9 10 21616 617 11 211026 1027 211213 1214 12 13 14 21713 714 15 2194 95 211031 111 16 17 18 19
More information2009-9-2.indd
Q No.1441 Q No.1442 Vol.33 NO.9 (2009) 30 (614) Q No.1443 Vol.33 NO.9 (2009) 31 (615) Q No.1444 Vol.33 NO.9 (2009) 32 (616) Vol.33 NO.9 (2009) 33 (617) Vol.33 NO.9 (2009) 34 (618) Q No.1445 Vol.33 NO.9
More informationuntitled
Vol.27 1 Vol.27 2 Vol.27 3 Vol.27 4 Vol.27 5 Vol.27 6 Vol.27 7 Vol.27 8 Vol.27 9 Vol.27 10 11 Vol.27 Vol.27 12 Vol.27 13 Vol.27 14 Vol.27 15 Vol.27 16 Vol.27 17 Vol.27 2007 10 29 18 http://www.nira.or.jp/index.html
More information