DCLA14: 2nd International Workshop on Discourse-Centric Learning Analyticsat 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.
1. 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
8. 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
11. 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.”
12. 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.
13. 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.
14. 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”.
15. 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.
16. 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.
34. 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
35. 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
37. 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
46. 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
47. 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
48. Rhetorical
discourse
analy2cs
to
what
extent
do
comments
display
the
hallmarks
of
reasoned
wri2ng
which
makes
thinking
visible?
49. Rhetorical
discourse
analy2cs
to
what
extent
do
comments
display
the
hallmarks
of
reasoned
wri2ng
which
makes
thinking
visible?
50. 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
51. Process-‐Goal-‐Excep2on
analysis
identify normative
process model
identify ideal goals for
each subtask
identify possible
exceptions for each goal
process
decomposition
process model
with goals
process model with
goals and exceptions
identify handlers for
each exception
56. PGE
analysis
of
IBIS
syntax
checking
for
impoverished
argumenta2on
57. 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
59. 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