Toward optimal pedagogical action patterns by means of Partially Observable Markov Decision Process

Manuel Mejía-Lavalle; Hermilo Victorio; Alicia Martínez; Grigori Sidorov; Enrique Sucar; Obdulia Pichardo-Lagunas

doi:10.1007/978-3-319-62428-0_38

Toward optimal pedagogical action patterns by means of Partially Observable Markov Decision Process

Manuel Mejía-Lavalle, Hermilo Victorio, Alicia Martínez, Grigori Sidorov, Enrique Sucar, Obdulia Pichardo-Lagunas

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Good pedagogical actions are key components in all learning-teaching schemes. Automate that is an important Intelligent Tutoring Systems objective. We propose apply Partially Observable Markov Decision Process (POMDP) in order to obtain automatic and optimal pedagogical recommended action patterns in benefit of human students, in the context of Intelligent Tutoring System. To achieve that goal, we need previously create an efficient POMDP solver framework with the ability to work with real world tutoring cases. At present time, there are several Web available POMDP open tool solvers, but their capacity is limited, as experiments showed in this paper exhibit. In this work, we describe and discuss several design ideas toward obtain an efficient POMDP solver, useful in our problem domain.

Original language	English
Title of host publication	Advances in Soft Computing - 15th Mexican International Conference on Artificial Intelligence, MICAI 2016, Proceedings
Editors	Obdulia Pichardo-Lagunas, Sabino Miranda-Jimenez
Publisher	Springer Verlag
Pages	473-480
Number of pages	8
ISBN (Print)	9783319624273
DOIs	https://doi.org/10.1007/978-3-319-62428-0_38
State	Published - 2017
Event	15th Mexican International Conference on Artificial Intelligence, MICAI 2016 - Cancun, Mexico Duration: 23 Oct 2016 → 28 Oct 2016

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	10062 LNAI
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	15th Mexican International Conference on Artificial Intelligence, MICAI 2016
Country/Territory	Mexico
City	Cancun
Period	23/10/16 → 28/10/16

Keywords

Automatic pattern generation
Intelligent Tutoring Systems
Optimal pedagogical actions
Partially Observable Markov Decision Process
Statistical & structural pattern

Access to Document

10.1007/978-3-319-62428-0_38

Cite this

Mejía-Lavalle, M., Victorio, H., Martínez, A., Sidorov, G., Sucar, E., & Pichardo-Lagunas, O. (2017). Toward optimal pedagogical action patterns by means of Partially Observable Markov Decision Process. In O. Pichardo-Lagunas, & S. Miranda-Jimenez (Eds.), Advances in Soft Computing - 15th Mexican International Conference on Artificial Intelligence, MICAI 2016, Proceedings (pp. 473-480). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10062 LNAI). Springer Verlag. https://doi.org/10.1007/978-3-319-62428-0_38

Mejía-Lavalle, Manuel ; Victorio, Hermilo ; Martínez, Alicia et al. / Toward optimal pedagogical action patterns by means of Partially Observable Markov Decision Process. Advances in Soft Computing - 15th Mexican International Conference on Artificial Intelligence, MICAI 2016, Proceedings. editor / Obdulia Pichardo-Lagunas ; Sabino Miranda-Jimenez. Springer Verlag, 2017. pp. 473-480 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{1f10cc9a34bf4b93b50ac55367e06d34,

title = "Toward optimal pedagogical action patterns by means of Partially Observable Markov Decision Process",

abstract = "Good pedagogical actions are key components in all learning-teaching schemes. Automate that is an important Intelligent Tutoring Systems objective. We propose apply Partially Observable Markov Decision Process (POMDP) in order to obtain automatic and optimal pedagogical recommended action patterns in benefit of human students, in the context of Intelligent Tutoring System. To achieve that goal, we need previously create an efficient POMDP solver framework with the ability to work with real world tutoring cases. At present time, there are several Web available POMDP open tool solvers, but their capacity is limited, as experiments showed in this paper exhibit. In this work, we describe and discuss several design ideas toward obtain an efficient POMDP solver, useful in our problem domain.",

keywords = "Automatic pattern generation, Intelligent Tutoring Systems, Optimal pedagogical actions, Partially Observable Markov Decision Process, Statistical & structural pattern",

author = "Manuel Mej{\'i}a-Lavalle and Hermilo Victorio and Alicia Mart{\'i}nez and Grigori Sidorov and Enrique Sucar and Obdulia Pichardo-Lagunas",

note = "Publisher Copyright: {\textcopyright} Springer International Publishing AG 2017.; 15th Mexican International Conference on Artificial Intelligence, MICAI 2016 ; Conference date: 23-10-2016 Through 28-10-2016",

year = "2017",

doi = "10.1007/978-3-319-62428-0_38",

language = "Ingl{\'e}s",

isbn = "9783319624273",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Verlag",

pages = "473--480",

editor = "Obdulia Pichardo-Lagunas and Sabino Miranda-Jimenez",

booktitle = "Advances in Soft Computing - 15th Mexican International Conference on Artificial Intelligence, MICAI 2016, Proceedings",

address = "Alemania",

}

Mejía-Lavalle, M, Victorio, H, Martínez, A, Sidorov, G, Sucar, E & Pichardo-Lagunas, O 2017, Toward optimal pedagogical action patterns by means of Partially Observable Markov Decision Process. in O Pichardo-Lagunas & S Miranda-Jimenez (eds), Advances in Soft Computing - 15th Mexican International Conference on Artificial Intelligence, MICAI 2016, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 10062 LNAI, Springer Verlag, pp. 473-480, 15th Mexican International Conference on Artificial Intelligence, MICAI 2016, Cancun, Mexico, 23/10/16. https://doi.org/10.1007/978-3-319-62428-0_38

Toward optimal pedagogical action patterns by means of Partially Observable Markov Decision Process. / Mejía-Lavalle, Manuel; Victorio, Hermilo; Martínez, Alicia et al.
Advances in Soft Computing - 15th Mexican International Conference on Artificial Intelligence, MICAI 2016, Proceedings. ed. / Obdulia Pichardo-Lagunas; Sabino Miranda-Jimenez. Springer Verlag, 2017. p. 473-480 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10062 LNAI).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Toward optimal pedagogical action patterns by means of Partially Observable Markov Decision Process

AU - Mejía-Lavalle, Manuel

AU - Victorio, Hermilo

AU - Martínez, Alicia

AU - Sidorov, Grigori

AU - Sucar, Enrique

AU - Pichardo-Lagunas, Obdulia

N1 - Publisher Copyright: © Springer International Publishing AG 2017.

PY - 2017

Y1 - 2017

N2 - Good pedagogical actions are key components in all learning-teaching schemes. Automate that is an important Intelligent Tutoring Systems objective. We propose apply Partially Observable Markov Decision Process (POMDP) in order to obtain automatic and optimal pedagogical recommended action patterns in benefit of human students, in the context of Intelligent Tutoring System. To achieve that goal, we need previously create an efficient POMDP solver framework with the ability to work with real world tutoring cases. At present time, there are several Web available POMDP open tool solvers, but their capacity is limited, as experiments showed in this paper exhibit. In this work, we describe and discuss several design ideas toward obtain an efficient POMDP solver, useful in our problem domain.

AB - Good pedagogical actions are key components in all learning-teaching schemes. Automate that is an important Intelligent Tutoring Systems objective. We propose apply Partially Observable Markov Decision Process (POMDP) in order to obtain automatic and optimal pedagogical recommended action patterns in benefit of human students, in the context of Intelligent Tutoring System. To achieve that goal, we need previously create an efficient POMDP solver framework with the ability to work with real world tutoring cases. At present time, there are several Web available POMDP open tool solvers, but their capacity is limited, as experiments showed in this paper exhibit. In this work, we describe and discuss several design ideas toward obtain an efficient POMDP solver, useful in our problem domain.

KW - Automatic pattern generation

KW - Intelligent Tutoring Systems

KW - Optimal pedagogical actions

KW - Partially Observable Markov Decision Process

KW - Statistical & structural pattern

UR - http://www.scopus.com/inward/record.url?scp=85028449794&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-62428-0_38

DO - 10.1007/978-3-319-62428-0_38

M3 - Contribución a la conferencia

SN - 9783319624273

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 473

EP - 480

BT - Advances in Soft Computing - 15th Mexican International Conference on Artificial Intelligence, MICAI 2016, Proceedings

A2 - Pichardo-Lagunas, Obdulia

A2 - Miranda-Jimenez, Sabino

PB - Springer Verlag

T2 - 15th Mexican International Conference on Artificial Intelligence, MICAI 2016

Y2 - 23 October 2016 through 28 October 2016

ER -

Mejía-Lavalle M, Victorio H, Martínez A, Sidorov G, Sucar E, Pichardo-Lagunas O. Toward optimal pedagogical action patterns by means of Partially Observable Markov Decision Process. In Pichardo-Lagunas O, Miranda-Jimenez S, editors, Advances in Soft Computing - 15th Mexican International Conference on Artificial Intelligence, MICAI 2016, Proceedings. Springer Verlag. 2017. p. 473-480. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-319-62428-0_38

Toward optimal pedagogical action patterns by means of Partially Observable Markov Decision Process

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this