59
DCLA meet CIDA Collec&ve Intelligence Delibera&on Analy&cs Simon Buckingham Shum & Anna De Liddo Mark Klein DCLA14: 2 nd Interna2onal Workshop on DiscourseCentric Learning Analy2cs at LAK14: hAp://dcla14.wordpress.com

DCLA meet CIDA: Collective Intelligence Deliberation Analytics

Embed Size (px)

DESCRIPTION

DCLA14: 2nd International Workshop on Discourse-Centric Learning Analytics at LAK14: http://dcla14.wordpress.com Abstract: This discussion paper builds a bridge between Discourse-Centric Learning Analytics (DCLA), whose focus tends to be on student discourse in formal educational contexts, and research and practice in Collective Intelligence Deliberation Analytics (CIDA), which seeks to scaffold quality deliberation in teams/collectives devising solutions to complex problems. CIDA research aims to equip networked communities with deliberation platforms capable of hosting large scale, reflective conversations, and actively feeding back to participants and moderators the ‘vital signs’ of the community and the state of its deliberations. CIDA tends to focus not on formal educational communities, although many would consider themselves learning communities in the broader sense, as they recognize the need to pool collective intelligence in order to understand, and co-evolve solutions to, complex dilemmas. We propose that the context and rationale behind CIDA efforts, and emerging CIDA implementations, contribute a research and technology stream to the DCLA community. The argument is twofold: (i) The context of CIDA work connects with the growing recognition in educational thinking that students from school age upwards should be given the opportunities to engage in authentic learning challenges, wrestling with problems and engaging in practices increasingly close to the complexity they will confront when they graduate. (ii) In the contexts of both DCLA and CIDA, different kinds of users need feedback on the state of the debate, and the quality of the conversation: the students and educators served by DCLA are mirrored by the citizens and facilitators served by CIDA. In principle, therefore, a fruitful dialogue could unfold between DCLA/CIDA researchers and practitioners, in order to better understand common and distinctive requirements.

Citation preview

Page 1: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

DCLA  meet  CIDA  Collec&ve  Intelligence  Delibera&on  Analy&cs    

Simon  Buckingham  Shum  &  Anna  De  Liddo    Mark  Klein  

DCLA14:  2nd  Interna2onal  Workshop  on  Discourse-­‐Centric  Learning  Analy2cs  at  LAK14:  hAp://dcla14.wordpress.com  

Page 2: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

Complex  societal  challenges  

Page 3: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

Interest  in  the  poten2al  of  plaGorms  to  harness  Social  Innova2on  Collec2ve  Intelligence  

Page 4: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

EU  Collec2ve  Awareness  PlaGorms  for  Sustainability  &  Social  Innova2on  hAp://caps2020.eu  

Page 5: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

CI  means  many  things  to    many  people…  

5  

Page 6: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

CATALYST  Project  hAp://catalyst-­‐fp7.eu  

Page 7: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

Flat  commen2ng/Threaded  discussion  what  everyone  uses  now    

Page 8: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

Idea2on  plaGorms  Intui2ve  for  scaleable  brainstorming    But  studies  show  that  the  explora2on  of  he  problem  space  is  poor,  a  lot  of  repe22on,  and  weak  knowledge  building.  Labour  intensive  to  sort  through  thousands  of  ideas.  Facilitators  play  a  key  role  in  ensuring  that  ideas  get  connected.    Hard  for  analy2cs  to  gauge  quality  of  discourse  

e.g.    hAp://www.spigit.com  hAp://ideascale.com  

Page 9: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

Idea2on  plaGorms  

hAp://www2.mitre.org/public/jsmo/call-­‐for-­‐papers-­‐lg-­‐scale-­‐idea2on%20.html  

Page 10: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

Pain  Points  in  Social  Innova2on  PlaGorms  

Page 11: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

Pain  Points  priori2sed  by  orgs  who  run  social  innova2on  plaGorms  

!   Hard  to  visualise  the  debate    !   Poor  summarisa2on  !   Poor  commitment  to  ac2on    !   Sustaining  par2cipa2on  !   Shallow  contribu2ons  and  unsystema2c  coverage  !   Poor  idea  evalua2on    

Effec2ve  visualisa2on  of  concepts,  new  ideas  and  delibera2ons  is  essen2al  for  shared  understanding,  but  suffers  both  from  a  lack  of  efficient  tools  to  create  them  and  from  a  lack  of  ways  to  reuse  them  across  plaGorms  and  debates      “As  a  user,  visualisa2on  is  my  biggest  problem.  It  is  o_en  difficult  to  get  into  the  discussion  at  the  beginning.  As  a  manager  of  these  plaGorms,  showing  people  what  is  going  on  is  the  biggest  pain  point.”    

Page 12: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

Pain  Points  priori2sed  by  orgs  who  run  social  innova2on  plaGorms  

!   Hard  to  visualise  the  debate    !   Poor  summarisa2on  !   Poor  commitment  to  ac2on    !   Sustaining  par2cipa2on  !   Shallow  contribu2ons  and  unsystema2c  coverage  !   Poor  idea  evalua2on    

Par2cipants  struggle  to  get  a  good  overview  of  what  is  unfolding  in  an  online  community  debate.  Only  the  most  mo2vated  par2cipants  will  commit  a  lot  of  2me  to  reading  the  debate  in  order  to  iden2fy  the  key  members,  the  most  relevant  discussions,  etc.  The  majority  of  par2cipants  tend  to  respond  unsystema2cally  to  s2mulus  messages,  and  do  not  digest  earlier  contribu2ons  before  they  make  their  own  contribu2on  to  the  debate,  such  is  the  cogni2ve  overhead  and  limited  2me.    

Page 13: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

Pain  Points  priori2sed  by  orgs  who  run  social  innova2on  plaGorms  

!   Hard  to  visualise  the  debate    !   Poor  summarisa2on  !   Poor  commitment  to  ac2on    !   Sustaining  par2cipa2on  !   Shallow  contribu2ons  and  unsystema2c  coverage  !   Poor  idea  evalua2on    

Bringing  mo2vated  audiences  to  commit  to  ac2on  is  difficult.  Enthusiasts,  those  who  have  an  interest  in  a  subject  but  have  yet  to  commit  to  taking  ac2on,  are  le_  behind.      Need  to  prompt  ac2on  in  community  members    Reaching  a  consensus  was  considered  less  important  than  being  enabled  to  act.    

Page 14: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

Pain  Points  priori2sed  by  orgs  who  run  social  innova2on  plaGorms  

!   Hard  to  visualise  the  debate    !   Poor  summarisa2on  !   Poor  commitment  to  ac2on    !   Sustaining  par2cipa2on  !   Shallow  contribu2ons  and  unsystema2c  coverage  !   Poor  idea  evalua2on    

Mo2va2ng  par2cipants  with  widely  differing  levels  of  commitment,  exper2se  and  availability  to  contribute  to  an  online  debate  is  challenging  and  o_en  unproduc2ve.      Sustaining  par2cipa2on  more  important  than  enlarging  par2cipa2on.      “It  is  beAer  to  have  quality  input  from  a  small  group  than  a  lot  of  members  but  very  liAle  content”.    

Page 15: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

Pain  Points  priori2sed  by  orgs  who  run  social  innova2on  plaGorms  

!   Hard  to  visualise  the  debate    !   Poor  summarisa2on  !   Poor  commitment  to  ac2on    !   Sustaining  par2cipa2on  !   Shallow  contribu2ons  and  unsystema2c  coverage  !   Poor  idea  evalua2on    

Open  innova2on  systems  tend  to  generate  a  large  number  of  rela2vely  shallow  ideas.  Poor  collabora2ve  refinement  of  ideas  that  could  allow  the  development  of  more  refined,  deeply  considered  contribu2ons.      No  easy  way  to  see  which  problem  facets  remain  under-­‐covered.  Very  par2al  coverage  of  the  solu2on  space.  

Page 16: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

Pain  Points  priori2sed  by  orgs  who  run  social  innova2on  plaGorms  

!   Hard  to  visualise  the  debate    !   Poor  summarisa2on  !   Poor  commitment  to  ac2on    !   Sustaining  par2cipa2on  !   Shallow  contribu2ons  and  unsystema2c  coverage  !   Poor  idea  evalua2on    

Patchy  evalua2on  of  ideas    Poor  quality  jus2fica2on  for  ideas.      Hard  to  see  why  ra2ngs  have  been  given.      Unclear  which  ra2onales  are  evidence  based.  

Page 17: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

CI  Delibera2on  PlaGorms:  the  addi2on  of  seman2c  structure  

Page 18: DCLA meet CIDA: Collective Intelligence Deliberation Analytics
Page 19: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

ODET website: slides, movies, papers, tools

19

olnet.org/odet2010  

Page 20: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

bCisive  online:  product  grade  argument  mapping  

20  

Page 21: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

DebateGraph  

Page 22: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

DebateGraph  

Page 23: DCLA meet CIDA: Collective Intelligence Deliberation Analytics
Page 24: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

MIT’s  Deliberatorium  

Page 25: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

OU’s  Evidence  Hub  

25  

Page 26: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

OU’s  Evidence  Hub  

Page 27: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

OU’s  Cohere  

Page 28: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

OU’s  Cohere  

Page 29: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

29  

OU’s  Cohere  

Page 30: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

Consider.It  

Page 31: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

Consider.It  

Page 32: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

YourView  

Page 33: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

Can  we  see  such  tools  in  educa2on?  

Page 34: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

CI  Discourse  and  Formal  Educa2on  Discourse  —  a  mee2ng  of  minds  

Collec&ve  Intelligence  for  Social  Innova&on   Formal  Educa&on  

Ci2zen   Student  

Moderator   Teacher  

Seeking  strong  voluntary  par2cipa2on   Voluntary/required  par2cipa2on  

Seeking  good  explora2on  of  the  problem,  building  on  peers’  ideas  

Seeking  collec2vely  owned  solu2on   May  also  be  seeking  the  correct  solu2on  

Civil  discourse,  ideally  well  argued  

Ideas  from  all  stakeholders  

Page 35: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

CI  vs  Educa2onal  Discourse  Tools  

CI  Delibera&on  PlaDorms   Educa&onal  Argumenta&on  PlaDorms  

simple,  professional  interfaces   efforGul,  more  amateur  interfaces  

authen2c,  complex  problems     ar2ficial  problems  untrained  users  (ci2zens)  who  

choose  to  use  the  tools  (possibly  trained)  students  who  are  required  to  use  the  tools  

mul2ple,  engaging  visualiza2ons  

argument  networks  

Page 36: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

Approaches  to  Discourse  Analy2cs  

Page 37: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

DCLA  strategies  from  AIED/CSCL  Scheuer  O,  McLaren  BM,  Loll  F  and  Pinkwart  N.  (2012)  Automated  Analysis  and  Feedback  Techniques  to  Support  Argumenta2on:  A  Survey.  In:  McLaren  BM  and  Pinkwart  N  (eds)  Educa-onal  Technologies  for  Teaching  Argumenta-on  Skills.  Bentham  Science  Publishers,  71–124    

37  

Analysis  Approach Descrip&on

Syntac2c  analysis Rule-­‐based  approaches  that  find  syntac2c  paAerns  in  argument  diagrams  

Systems:  Belvedere,  LARGO

Problem-­‐  specific  analysis Use  of  a  problem-­‐specific  knowledge  base  to  analyze  student  arguments  or  synthesize  new  arguments  

Systems:  Belvedere,  LARGO,  Rashi,  CATO

Simula2on  of  reasoning  and  decision  making  processes

Qualita2ve  and  quan2ta2ve  approaches  to  determine  believability  /  acceptability  of  statements  in  argument  models  

Systems:  Zeno,  Hermes,  ArguMed,  Carneades,  Convince  Me,  Yuan  et  al.  (2008)

Assessment  of  content  quality Collabora2ve  filtering,  a  technique  in  which  the  views  of  a  community  of  users  are  evaluated,  to  assess  the  quality  of  the  contribu2ons’  textual  content  

Systems:  LARGO

Classifica2on  of  the  current  modeling  phase Classifica2on  of  the  current  phase  a  student  is  in  according  to  a  predefined  process  model  

Systems:  Belvedere,  LARGO

Page 38: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

Argunaut  Moderator  Tool  

38  

Page 39: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

Catalyst  Project:  CI  Analy2cs  Concept  

Page 40: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

Or  use  a  Na2ve    IBIS  PlaGorm  

Page 41: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

Discourse  Analy2cs:  Visualiza2on  

Page 42: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

DCLA  analy2cal  ques2ons  

Page 43: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

CIDA  Visualiza2on  storyboarding  

Page 44: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

CIDA  Visualiza2on  storyboarding  

Page 45: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

CI  Dashboard  mockups  

Page 46: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

Discourse  Analy2cs:  Rhetorical  Parsing  of  Discussion  Forum  

Simsek  D,  Buckingham  Shum  S,  Sándor  Á,  De  Liddo  A  and  Ferguson  R.  (2013)  XIP  Dashboard:  Visual  Analy&cs  from  Automated  Rhetorical  Parsing  of  Scien&fic  Metadiscourse.  1st  Interna-onal  Workshop  on  Discourse-­‐Centric  Learning  Analy-cs,  at  3rd  Interna-onal  Conference  on  Learning  Analy-cs  &  Knowledge.  Leuven,  BE  (Apr.  8-­‐12,  2013).  Open  Access  Eprint:  hAp://oro.open.ac.uk/37391  

Page 47: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

Rhetorical  discourse  analy2cs  to  what  extent  do  comments  display  the  hallmarks  of  reasoned  wri2ng  which  makes  thinking  visible?  

<IMPORTANT  SUMMARY>  The  argument  is  that  the  consumer  has  benefited  because  technology  has  increasesd  consumer  access  to  markets  and  has  forced  brands  to  become  more  open  and  transparent.    Likewise,  organisa2ons  benefit  as  technology  allows  them  greater  access  to  consumer  informa2on.  So  it  seems  that  we  have  all  gained  from  the  impact  of  technology.    The  strongest  arguments  seemed  to  lean  towards  the  consumer  as  benefi2ng  most.    I  am  not  convinced.    I  think  that,  as  brands  become  more  sophis2cated  and  knowledgeable  in  their  approach,  consumer  resistance  becomes  more  difficult.      <IMPORTANT  SUMMARY  CONTRAST>  Really  good  thoughts  -­‐  I  hadn't  considered  the  other  stakeholders.  I’m  thinking  of  local  brands  ,  which  are  small  now  ,  but  have  ambi2on  to  get  bigger.  SMEs  are  not  going  to  create  huge  brand  value  overnight  ,  but  I  think  lessons  can  be  taken  from  what  the  big  brands  are  doing  and  employed  by  SMEs    

Page 48: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

Rhetorical  discourse  analy2cs  to  what  extent  do  comments  display  the  hallmarks  of  reasoned  wri2ng  which  makes  thinking  visible?  

Page 49: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

Rhetorical  discourse  analy2cs  to  what  extent  do  comments  display  the  hallmarks  of  reasoned  wri2ng  which  makes  thinking  visible?  

Page 50: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

Discourse  Analy2cs:  Process-­‐Goal-­‐Excep2on  Analysis  

Klein  M.  (2003)  A  Knowledge-­‐Based  Methodology  for  Designing  Reliable  Mul2-­‐Agent  Systems.  In:  Giorgini  P,  Mueller  JP  and  Odell  J  (eds)  Agent-­‐Oriented  SoIware  Engineering  IV.  Springer-­‐Verlag,  85  -­‐  95.    Klein  M.  (2012)  Enabling  Large-­‐Scale  Delibera2on  Using  AAen2on-­‐Media2on  Metrics.  Computer  Supported  Coopera-ve  Work  21:  449-­‐473    

Page 51: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

Process-­‐Goal-­‐Excep2on  analysis  

identify normative process model

identify ideal goals for each subtask

identify possible exceptions for each goal

processdecomposition

process model with goals

process model with goals and exceptions

identify handlers for each exception

Page 52: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

Process-­‐Goal-­‐Excep2on  (PGE)  analysis  

process

exception

goal

requires

has-part

is-violated-by

is-handled-by

has-part

is-caused-by

Page 53: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

Deliberatorium  PGE  analy2cs  modelling  

Page 54: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

PGE  analysis  of  author  diversity  

Page 55: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

PGE  analysis  of  diversity  of  ideas  

Page 56: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

PGE  analysis  of  IBIS  syntax  checking  for  impoverished  argumenta2on  

Page 57: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

Implemen2ng  handlers  using  PQL  graph  queries  

Currently  hardwired  to  Deliberatorium,  but  will  work  on  data  compliant  with  a  new  interchange  format  for  cross-­‐

plaMorm  interoperability  

Page 58: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

Deliberatorium  recommender  agent  priori2ses  areas  for  moderator  aAen2on  

Page 59: DCLA meet CIDA: Collective Intelligence Deliberation Analytics

CIDA—DCLA  synergies  

DCLA  

New  kinds  of  UX  for  

structured  argumenta2on  

New  kinds  of  visualiza2on  of  argumenta2on  

+  domain  

Authen2c  use  contexts  

Moderator  tools  

CIDA  Analy2cs  

Taxonomies  

AI  techniques  

Small  scale  prototypes  

AAen2on  to  quality  

discourse