JMP WordMiner JMPer s Meeting 2005104 ohsumi@ss.iij4u.or.jp yasuda@hic.co.jp All rights reserved. Copyright by Noboru Ohsumi, ISM Professor Emeritus. JMPer s Meeting TM JMP JMP version 6.0 WordMiner JMP Web 2 1
JMP 3 DMTM 4 2
5 mine 6 3
7 DM TMDM 8 4
TM Qualitative Research Content Analysis CACA Computer-assisted Content Analysis open-ended questionsfgofg 9 TM vs vs RDBDWH E-mailWeb vs TM 10 5
11 12 6
7 13 14 SOM Kohonen mapping IMMIS OLAP CART, THAIDCHAID SVM EDA
TMTM text categorization document clusteringdocument classification rule mining from text conceptrelationship mining from text information integration topic detection summarization, analysis of text collection knowledge acquisition/capture & understanding text navigation, visualization WebWeb CRM full text documents searchdocument retrieval 15 data science WordMiner data-driven experimental design) data collection mode analyzing 16 8
qualitative information ( GIFGI OFG TM Web 17 2004 192135-159 HP WordMiner http://wordminer.comquest.co.jp/ 18 9
closed questions open-ended questions TM WordMiner III JMP JMP 19 Web 20 10
1 2 i n w w w w (1) (1) (1) (1) 1 2 j k (2) (2) (2) (2) 1 2 j k w w w w w w w w () i () i () i () i 1 2 j l w w w ( n) ( n) ( n) 1 2 j fragments 21 22 11
(WordMiner) 23 24 12
WordMiner 25 J 1 2 j n I ' 1 2 k k K I J 1 2 i m 1 2 k ' k K f f f f 11 12 1 j 1n f f f f 21 22 2 j 2n F=(f ij ) f f f f i1 i2 ij in f f f f m1 m2 mj mn z z z z * * * * 11 21 j1 n1 z z z z * * * * 12 22 j2 n2 z z z z * * * * 1 2 k k jk nk z z z z * * * * ' ' ' ' 1 2 k k jk nk z z z z * * * * 1 2 K K ik nk z z z z z z z z z z z z z z z z z z z z 11 12 1k ' 1k 1K 21 22 2k ' 2k 2K i1 i2 ik ' ik ik m1 m2 mk ' mk mk I J IJ2 K=min{m, n} 1 26 13
2 I 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. J 1. 2. 3. 27 7 8 2 28 14
9 2 10 210 93 2 29 91 2 JMP 30 15
16 31 i1 z 2 i z 1 * j z 2 * j z 2 1 K=min{m, n}1=2 2 k k λ 11 12 32 210 2 JMP 3
WordMiner 33 WordMiner 4 WordMiner 2 6 34 17
18 35 WordMiner JMP JMP WordMiner JMP 36 SOM WordMiner JMP WordMiner JMP
Web Q31. Q32. 37 WordMiner A JMP B 38 AB 19
20 39 WordMiner JMP C C2 C1 D CD 40 WordMinerJMP 13
21 41 WordMinerJMP 14 42
43 JMP - 3-1 -2-1 -2-4 - 3 WordMinerJMP 44 22
23 45 46 8 ( -3 WordMiner
24 47 9 ( -3 nuage 48 k-som JMP k-
25 49 4-1 896 15 15 50 5-2 248 15
26 51 6-2 chaining effect 52 k- 7 k- -1 k-15 WordMiner
WordMiner WordMiner JMP 4 53 10-420 WordMiner JMP 54 27
15 55 15 56 28
29 57 11 58 12 20 WordMiner
30 59 148 10 C - 3 C1-1 - 3-1 15 60 13
WordMiner 14 WordMiner 61 16 12WordMiner WordMiner JMP 14 62 31
16 63 JMP JMP6.0 WordMiner JMP 64 32