The document describes an interactive and personalized eBook system called iRead for learning to read. iRead was developed as part of an EU Horizon-2020 funded project. It includes a software infrastructure for personalized and adaptive technologies, applications like NAVIGO and AMIGO, and an authoring tool to create personalized eBooks. The system architecture includes an eBook synthesizer, Android eBook reader, local server, and content provider. It uses domain models represented as directed acyclic graphs to model reading skills and user profiles to track competencies and unlocked features. The system personalizes content and activities based on the user's model, profile, and progress to adapt learning over time.
Interactive and Personalized Activity eBooks for Learning to Read: The iRead Case
1. Interactive and Personalized
Activity eBooks
for Learning to Read
The iRead Case
Panagiotopoulos Dionysios
National and Technical University of Athens
authors: N. Deligiannis, D. Panagiotopoulos, P. Patsilinakos, C. Raftopoulou, A. Symvonis
2. What is the iRead?
“Infrastructure and integrated tools for personalized learning
of reading skills”
*European Union Horizon-2020 funded project
• Software infrastructure of personalized, adaptive technologies
(domain models, profiles, linguistic resources)
• A set of applications (NAVIGO, AMIGO)
• “Authorless” eBook creator
2The 20th International Conference on Artificial Intelligence in
Education - AIED'2019
June 2019, Chicago
3. User’s typical work-flow
3The 20th International Conference on Artificial Intelligence in
Education - AIED'2019
June 2019, Chicago
Register/Login
Select games
and size
eBook
synthezither
ContentActivity*
Download
and play
4. System architecture
4The 20th International Conference on Artificial Intelligence in
Education - AIED'2019
June 2019, Chicago
eBook synthesizer UI
Android device –
eBook reader
Local save/sync server
Content provider
iRead
5. Domain Model
Features
(vertices)
Prerequisite
relations
(edges)
English 279 4,457
Greek 446 17,552
German 316 748
Spanish 326 17,290
5
Models
• English, Greek, Spanish, German, EFL,
Dyslexia (English, Greek)
• Directed Acyclic Graph (DAG)
Features (separate linguistic objects)
Relations (prerequisites)
The 20th International Conference on Artificial Intelligence in
Education - AIED'2019
June 2019, Chicago
Profiles
• Competence
• Unlocked
6. Profile – Feature unlocking
6The 20th International Conference on Artificial Intelligence in
Education - AIED'2019
June 2019, Chicago
Not enough
incoming flow
Unlock from
incoming flow
Unlock by
competence
7. Adaptivity and Personalization
7The 20th International Conference on Artificial Intelligence in
Education - AIED'2019
June 2019, Chicago
Model
Profile
Content
Synthesizer
eBook
Update Profile
8. Thank you
8The 20th International Conference on Artificial Intelligence in
Education - AIED'2019
June 2019, Chicago
9. Game description
Mini Game Description
Cleomatchra The learner connects pairs of strings to form full words. (PickaTale)
Perilous paths The learner selects a string (word or sentence) over three possible options, to complete a
sentence or answer a question. (PickaTale)
Remove the runes The learner selects all words from a given list that have a specific property. (Patakis).
Sliceophagus The learner splits compound words to stems. (PickaTale)
Bridgyptian The learner selects a number of string compounds from an available set and places them
in an appropriate order so that she creates a word or a sentence. Some string compounds
may have a predefined, fixed placement. (PickaTale)
Raft rapid re Words appear sequentially on the learner's screen. The learner must “hit” all words that
have a predefined property. (Patakis)
9The 20th International Conference on Artificial Intelligence in
Education - AIED'2019
June 2019, Chicago