How can we use Artificial Intelligence (AI) for improving reading comprehension of poor comprehenders? This presentation, delivered at AI*IA 2014 for Society in Pisa (Italy), explain how natural language processing and constraint-based reasoning technologies can help in automating the generation of serious games for improving the reading comprehension of poor comprehenders. The slides also analyse the results in terms of pedagogical effectiveness of the system for improving reading comprehension and quality of the automated generation process for delivering games.
40. SemiRautomated*generation
story
enriched annotations
text
Educator
Expert
Expert GUI
Learner GUI
text text text
annotations
GUI Layer ALS Layer
Learner
Reasoning
Module
Reasoner
Adaptive
Engine
NPL
illustrations
Visualisation
Annotation
44. SemiRautomated*generation
template
visual
image
text
image image image
text text text
story
annotations
enriched annotations
GUI Layer Educator
Expert
Learner
Expert GUI
Learner GUI
Visualisation
Module
Visualisation
www.terenceproject.eu
45. games
SemiRautomated*generation
template
visual
image
text
image image image
text text text
story
annotations
enriched annotations
GUI Layer Educator
Expert
Learner
Expert GUI
Learner GUI
Visualisation
Module
Visualisation
www.terenceproject.eu
55. EXAMPLE%METHODS%IN%TERENCE
APPROACH WITH%WHOM EXAMPLE%METHODS HOW
analytical HMI%experts%or%
domain%experts
heuristic%evaluation
formative,%
expert%evaluation summative
cognitive%walk`through
small`scale learners
observations
formative
think%aloud
large`scale learners aield%studies summative
From D4.2 and D4.3 technical annex
56. From D4.2 and D4.3 technical annex
overall%generation
story
text
games
text$+$visual
revision%of%Natural%
Language%Processing%
(NLP)%generation%of%
texts%for%labelling%%%%%%%%%
text
text text text
Evaluation*tasks
revision%of%Automated%
Reasoning%selection%of%
central%events%and%
solutions
57. From D4.2 and D4.3 technical annex
overall%generation
story
text
games
text$+$visual
revision%of%Natural%
Language%Processing%
(NLP)%generation%of%
texts%for%labelling%%%%%%%%%
text
text text text
Evaluation*tasks
revision%of%Automated%
Reasoning%selection%of%
central%events%and%
solutions
63. Overall*generation
Overall*generation:%development%
times:%%
Results2for2revision2time:22
` 12’6”%per%game%instance:%
↑%12’8”%for%time%games%
↓%10’6”%for%who%games%
Results2for2creation2time:22
` avg.%23”%per%game%instance
story
text
games
text$+$visual
From D4.2 and D4.3 technical annex
64. Overall*generation
Overall*generation:%development%
times:%%
Results2for2revision2time:22
` 12’6”%per%game%instance:%
↑%12’8”%for%time%games%
↓%10’6”%for%who%games%
Results2for2creation2time:22
` avg.%23”%per%game%instance
story
text
games
text$+$visual
Implications2for2the2generation2process:%the%AI`based%semi`automated%
development%process%seems%to%be%promising%for%optimising%development%
times
From D4.2 and D4.3 technical annex
68. Merry Seasonal Festivities
with
TERENCE books
for remembering
its best stories
THE TERENCE BOOK 2
THE TERENCE BOOK 1
Dedicated to:TERENCE team, reviewers, teachers, children, and all those
that contributed to the TERENCE stories. Our sincere “grazie”.