Chances and Marketing: On-line Conversation Analysis for Creative Scenario Creation

1,104 views

Published on

Chances and Marketing: On-line Conversation Analysis for Creative Scenario Creation

Published in: Business, Education
0 Comments
2 Likes
Statistics
Notes
  • Be the first to comment

No Downloads
Views
Total views
1,104
On SlideShare
0
From Embeds
0
Number of Embeds
59
Actions
Shares
0
Downloads
0
Comments
0
Likes
2
Embeds 0
No embeds

No notes for slide

Chances and Marketing: On-line Conversation Analysis for Creative Scenario Creation

  1. 1. Chances and Marketing: On-line Conversation Analysis for Creative Scenario Creation Xavier Llora’, David E. Goldberg’, Yukio Ohsawa-’, Kei Ohnishi", Hiroshi Tamura‘, Yuichi Washida‘ & Masataka Yoshikawae‘ [Illinois Genetic Algorithms Lab. National Center for Supercomputer Application, University of Illinois at Urbana—Champaign ‘ichance Discovery Consortium. University of Tsukuba & University of Tokyo 3Research and Development Division. Hakuhodo Inc. xllora@illigal. ge. uiuc. edu / _ ' ulirr-r: ('-e'w ' , n'| r.'vri<lri'n'ar. vy M. fi? ll‘| ‘lfr‘I‘. ‘lr'I2vVm' GITA-I7 9, . . . __ __ ' Do‘, ’1Q'I'Vlo"| I :1‘ n"w':1l E/ igrwwvig Nn: .’ir. ru: u' (rT'r‘1’, 'A'. v’ ii. r.wri'. mn. ‘71'~_z A . :‘: rr, .‘n'os ‘ i_ , Um--¢-. 'sv a, lllnzm at Uztxznz-C! n'1|pav:1 M'"Ifl'S<'. ‘V ax’ u‘r'rrt: n at Ul'c: n:-C ‘, I“ 1 up . ‘: r.'V£TiVr . -mu. 1'. ' R31 (rmvvpuwyn It. 61610. -'_Ll_
  2. 2. DISCUS Project Innovation technology for creative collaboration, and decision-making. Combine HHC (human-human collaboration) and HMC (human-machine collaboration) to form powerful system. Envision both synchronous and asynchronous collaboration for continuous innovation. Overcome superficiality of online interaction through augmented reflection EWCD-2004 2 Xavier Llor‘a I , Nfis/ it
  3. 3. DISCUS Overall Picture -" ‘x. , , ”’ '_ Globalstakeholders ' ~ -’ Population (GSP) ‘ - / [EC Collaborative Validation Community (CVC) , V‘ 4' * i .396 GPG/ :9 0,5: Core Cohe rent ? Innovation Team e (CCIT) Tek 01* 1' EWCD-2004 3 Xavier Llor‘a I , Nfis/ it
  4. 4. ~ Knowledge Extraction (i? .l. i.- T. -.'-J. C‘ i'n; .l; t‘ l. 'n: --_) - Knowledge Management - Chance Discovery o Distributed Cooperation - Consensus Broadcast - Summary of the Evolved Solutions o Feedback Collection EWCD-2004 Collect Donate The DISCUS Methodology Relate _ - Knowledge Sharing 1,- 0 Collaborative Refinement /0 r - Electronic Brainstorming ‘ - Goal Definition DISCUS l . Create . - Innovation-based ' Decision Support (IDS) V o Innovation-based Solution Creation (ISC) 4 Xavier Llora l. . NESA
  5. 5. Marketing Scenarios Marketing scenario creation Scenarios: short descriptions of future hypothetical situations Identifying chances -9 anticipate: — market need new product Some data is usually available — objective: surveys, questionnaires. .. subjective: discussions, brainstorming sessions. .. EWCD-2004 5 Xavier Llor‘a L , N§5A
  6. 6. On-line Scenario Analysis 4: A group of marketing researchers a Discus about a particular topic: “single women and cell phone usage” . A big amount of raw information was available (questionnaires) . . They want to bootstrap a discussion of women interests using KeyGraphs e Identify relevant chances on the scenario discussion EWCD-2004 6 Xavier Llor‘a L , N§5A
  7. 7. The Creative Chance Discovery Process 4: A first test of the methodology: (1) Data gathering (objective data) (2) Data processing and mining (objective data) (3) Initial scenario creation (objective data) (4) Free discussion and reflection using the initial scenario (subjective data) (5) Creation of new scenarios (subjective data) EWCD-2004 7 Xavier Llor‘a L , N§5A
  8. 8. Initial Scenario Creation . Approximately 900 questionnaires were available . Cover three major US cities (New York, Chicago, Los Angeles) . A data cube was generated using DZK . Relevant rules were extracted using D2K’s rule association miner . Marketing researchers examined the rules and selected the relevant data subset they were interested (single women group) . An initial KeyGraph was built using the filtered data m EWCD-2004 3 Xavier Llora fl Nfi§A
  9. 9. Initial Scenario O . ... ... .. ... .y c u. U . .,. .~ . .,- . , . x.(. ".4 I o 0 mm‘. , . .. .. . I.. .-. « I. .. I. . W , U__, ,__‘_, _ o o mw IL ALI; . ° 0 . ,.. ¢. V. . . . : .-r. ».: . V . .. . (. ...4. . .-. I., ..u. .«. ._m. .. ., ... .-. ,.. o . 1-: ~vrnn. 'ur-'| 'Yh. nrbrv- -pm. ]'hv1vI, iv0wv1,ru-vvn-IrvqrrfI17rIvvrI'| r'I . ‘ I-‘VII, -AL . o . — -Thu" . «u-1-. -inn-an‘ "K 9" ‘u. '.L LlL>lL'_ | LL )1: ’ —, . . ,.. ., ,.. .c. .. . . ,.. I,. 4.. .. u. ..M. lb 1 . .47vUI4}_L§-K‘ . ».; .. llI4.'. l.lv V u~. ..I. . O Innf.1u. ‘.l-NlI'in'A mu ‘mum ¢«, v.u. . . .. . 0 . _ ‘H, _ _, , _ _H__(, _‘ U . ... .o--. .-. . . ... ,. m. ..«. r.v-nu n . ~ 1; . .4.7.~p. . . nn: n Q 1.-lxcltn . . 4n. .t. ... ..«. ~: 6 . . III, .. «.. um» nu. .. IVII‘ . . n I IV! »- . ~—r—». .‘-n-: —.u. .—. -4» ---vi L L I EWCD-2004 9 Xavier Llora L ,
  10. 10. On-line Scenario Discussion . Using the DISCUS infrastructure, the marketing researchers engage a discussion about the chances in the initial scenario . The conversation was analyzed on-line . The analysis was available at two levels: _ Message level Discussion level . KeyGraphs of the current discussion was available m EWCD-2004 10 Xavier Llora fl ~ޤA
  11. 11. L; [. p {In 1». ;. I 5,. I., b. 1I. I, -1-. 43;. «V. 3 In. . mm- J‘: ; cm. ~'VVW‘_ . .., .n. .,. :I1Iz= :. Kb | ': r “V2,. -u. '~: . 1-4. .7.e’4'. -»1.t‘. Iv. ¢tI afiillv. . 1'. -‘I5 --,3». .. 42.. .“. .. ¢. ... .I. ,. ; ... ..I: u.. . .3 ‘hip Molnle plane 1 . ... ..I-. (NEH! .. (mm . ..I I: , | s c. '. Rang um: hmu-n >. IIIL . ~.I-. . I. ,1.. I II T , . . ImI r Oh» n, ;_ r; ;;I'(» »I. .», ,.. . : 7.7.1 ; .I, —(. my gI—Iw, ; I urzz I; Iv, I—. (II flu lngI. ’IV1I"J‘ In. (IP11. » c : .I. »II .1 . ;II.1vI. :.. e 1I‘. V.l/ ‘ 391.: HILL I ,1; L'I‘ -11- II E! Ruhmm: zc anulnnuo : I -, I.}*: w.x'. :-Io. -I A E z (2. r: .I. I-. «.. .. . ..: ... s.. ‘-: £-: .-. i:LL‘_rrx- ‘ IDJIXILI. ;. II1.. II I —. I.. LI! " I II. 7' I. Iyru "nu ‘ I. .. . I ‘ I. —" I ' ‘F-I1 d. II II. -I‘ “ P1 :1.» . — m[IIYJ', 'h. Vfi'jI'vr'Iilh'V mI-. IIII+I ‘ Z Rn-Iuuun as --mgr-uo~x. L.u. -=. ;u. . I I. Lu: u -. x.I LI L. -.v v: r.: p. :I. l.x . :.; ... u.j . ':. u_ _ r. '»'jr1V1I‘: rr. c1-ix’ mi-I -mu’ II» TH: ma‘; II .1 hr 4 IIIs'4r" I In , , ‘pp41Ir‘Il‘[Il'VI1f" I11‘. » . ~.. - IhIII: _m Ru]-um . u I: Hug lmu — vI. .I. A-w. .: IJHILX I. ._II-I .1‘: 4 AIIII J II! -rII. I I- 4 X“ Lm r, M II‘ I-5e; rLf' Ixex: Ilhnk : I:m: zzu ; I., ;;. II»: Kulumm auanmztona -rt-IIl». I-xII. ‘ --go-In LJ - 9 ! .|. .IJ‘Ir Ir. .. ( , _.. 4. 1 ‘TI «1 , —I'. --*. I r _ _ .5 _ A -_ ~, _= I. -3 V 473 34.1;-Ft’ . ~‘7I~ E Inn 43: ‘ um-
  12. 12. Chance Discovery and Creativity Support: Final Remarks . On-line analysis enables fast feedback . KeyGraphs provide: a visual map of the discussion an externalization of the main topics and relations _ provide a common tools for reasoning about chances . Fast feedback speedup the creative process — cluster consolidation (convergent thinking) — cluster connection (divergent thinking) . Fatigue fighting (game like approach) m EWCD-2004 12 Xavier Llora fl ~g§A
  13. 13. V ‘ I‘ ‘ EWCD-2004 Cast of 1000 Characters Powered by TRECC: This project is supported by TRECC, a program of the UIUC administered by the NCSA by the Office of Naval Research under Grant It: N00014-01-1-0175 | lliGAL: David E. Goldberg, Xavier Llora, Naohiro Matsumura. Chen-Ju Chao, Feipeng Li, Kei Ohnishi, Tian Li Yu, Martin Butz, Antonio Gonzales ALG: Michael Welge, Loretta Auvil, Andew Shirk, Tom Redman, Duane Searsmith, Bei Yu Knowledge and Learning System Group: Tim Wentling, Andrew Wadsworth, Luigi Marini, Raj Barnerjee Minsker Research Group: Barbara Minsker, Aniruddha Bhaqwat, Wesley Dawsey, Abhishek Singh, Meghna Babbar Takagi Labo: Hideyuki Takagi University of Tsukuba GSM: Yukio Ohsawa Hakuhodo Inc. : Hiroshi Tamura, Yuichi Washida, Massataka Yoshikawa Others: Ali Yassine (GE), Miao Zhuang (GE) 13 Xavier Llora L I NCSA
  14. 14. Chances and Marketing: On-line Conversation Analysis for Creative Scenario Creation Xavier Llora’, David E. Goldberg’, Yukio Ohsawa-’, Kei Ohnishi", Hiroshi Tamura‘, Yuichi Washida‘ & Masataka Yoshikawa‘ [Illinois Genetic Algorithms Lab. National Center for Supercomputer Application, University of Illinois at Urbana—Champaign 2Chance Discovery Consortium. University of Tsukuba & University of Tokyo 3Research and Development Division. Hakuhodo Inc. xllora@illigal. ge. uiuc. edu / _ ' uIl'rIr: ('-e'w ' g. v|r. 'vn<1a'n'Iar. vy A| .rI: mIir»'Iflr'I: IrIII- Gimp G, . . . __ __ ' Do‘, ’1Q'l'Vlo"| I .1‘ n"w':1I E/ igrwwvig NI: .’iI: ru: I' (in'»‘I' , '.'. V h. r.wrI'. c:'In. ‘7I'~_z A . :‘: rI'. .‘7I'os ‘ l_ , Um--¢-. 'sv a, lllnzm .3! Uztxznz-C! n'1|pav:1 M'"Ifl'S<'. ‘V cx’ u‘I'rrtxs at Ul'(TC-C ‘, I“ 1 up . ‘: r.'i£TtVr tvxu. I". ' R31 (rg. Im, uIgI: II. 61610. -'_Ll_

×