Intelligent Tutoring Systems can successfully complement and substitute other instructional models in many contexts. However, it is very common to students to become bored or disengaged using ITS. The inclusion of gami cation capabilities (e.g., level, points and so on) in ITS design aims to engage students and to drive desired learning behaviors. Researchers have been noting that teachers are increasingly demanding to act as active users of systems with such features. In this context, the main challenge of this project is contributing to the actively participation (i.e., design) of teachers in the use of gami ed intelligent tutoring systems. This challenge leads to the following research questions: (i) \how could we enable teachers to customize the construction of gami ed ITS in a simple way and without requiring technical capabilities from them?"; and (ii) \how could we also provide good design principles in order to aid teachers in the customization of gami ed ITS?". Thus, our aim is to develop an intelligent authoring platform to enable teachers for customizing gami ed ITSs. In this way, we describe in this text a set of speci c objectives that must be completed to achieve this general aim.
Empirical evidences suggest that Intelligent Tutoring
Systems (ITSs) can successfully complement and substitute other
instructional models in many situations [
In general, the traditional development of ITSs do not
make efforts to engage and motivate students. On the other
hand, motivated, challenged and intrigued students tend to
have better learning results [
Meanwhile, teachers are increasingly demanding to act as
active users of systems with such features. For instance, a
recent survey [
In this context, the main challenge of this work is contributing to the actively participation of teachers in the use of intelligent tutoring systems that consider motivational aspects of the students using gami cation. However, since teachers have different expectations and/or methodologies as well as could use ITSs in several domains and different educational levels, they should be able to customize them according to their preferences. Thus, by actively participation we mean that teachers may be primary actors of gami ed ITSs, for example, by selecting which functionalities they are interested to incorporate in ITSs, by de ning which gami cation behaviors they expect from their students, by choosing which pedagogical strategies they may consider or by creating and/or reusing content.
The design of ITSs is very complex. It should take into
account the four classic ITS models (domain, student,
pedagogical and interface models) [
On the one hand, there is an increasing interest by teachers to actively use (i.e., customize) systems with such features, but on the other hand it is very complex to build them. Thus, \how could we enable teachers to customize the construction of Gami ed Intelligent Tutoring Systems in a simple way and without requiring technical capabilities from them?". However, only enabling teachers to customize these systems without providing some kind of support for their decision-making is not enough, because it is likely that they would build ineffective tutors both from performance and motivational aspects. In this way, \how could we also provide good design principles in order to aid teachers in the customization of gami ed ITS?"
Aiming to contribute to the challenges previously mentioned, we present in this section the objectives of this PhD project. However, before presenting our objectives, we will brie y discuss some theoretical concepts and important technologies that are used in this work.
Researchers have been investigating the use of authoring
tools for building intelligent tutoring systems since the
beginning of ITS research [
However, the development of authoring tools to aid the
construction of ITS with game elements may be considered
an open problem/challenge in computers and education
research. So far, to the best of our knowledge, we could not
nd any work in the literature that propose to use authoring
tools for enabling teachers to build gami ed ITS.
Furthermore, as mentioned by Sottilare et al [
In this way, to address the questions raised in the end of Section 1, and taking into account the potential bene ts of using authoring tools in the context of Gami ed ITSs, our aim in this PhD project is to develop an intelligent authoring platform to enable teachers for building Gami ed ITSs.
As part of the \intelligent" aspect of our authoring platform, we represent the knowledge about gami cation theories and ITS models as well as good gami cation behavior practices in a way that it can be used to aid the authoring process. For instance, some pre-de ned gami cation behaviors (e.g., performance, participation and so on) with prede ned game elements choices may support a better and more simple decision from the teacher. In order to represent such knowledge in a way that could be processable by the authoring platform, we are relying on the concept of ontologies. Ontologies can be logically reasoned and shared within a speci c domain. Thus, ontologies are a standard form for representing the concepts within a domain, as well as the relationships between those concepts in a way that allows automated reasoning.
Additionally, taking into account the high variability of gami ed ITSs (i.e., there are several technological, ITS and gami cation features that could be combined in different tutors) as well as the need for representing the knowledge about con guration choices of a teacher, we also intend to create a con guration model that could be automatically reasoned by a gami cation ITS platform to deliver a speci c system according to author's choices. In this context, we are relying on the concept of feature modeling to manage the variability of gami ed ITS.
Moreover, enabling the automatic analysis of feature
models and hence providing recon guration of a gami ed ITS is
also required. Achieving these characteristics could allow
for example, to monitor learner's motivational levels at the
time they are interacting with the ITS and to recon gure
the system with a different gami cation behavior that could
improve the engagement of students. Thus, in comparison
to other mechanisms for automatic analysis of features
models, description logic (DL) based methods (e.g., ontologies)
promise to provide improved automated inconsistency
detection, reasoning efficiency, scalability and expressivity [
Figure 1 presents an overview of the intelligent authoring process for building gami ed ITS from teacher's perspective. In order to achieve our general objective, we intend to reach the following speci c objectives: (a) Investigate and select proper ITS models from the literature to be represented in ontologies; (b) Investigate and select proper models of motivation and human behaviors from the literature to be represented in a gami cation ontology; (c) De ne a set of gami cation behavior good practices from empirical research papers in the context of education online to be incorporated in the gami cation ontology; (d) Specify an integrated ontology relating gami cation and
ITS ontologies; (e) Design and implement the authoring platform taking into account the gami cation and ITS knowledge represented in ontologies. This platform must consider authoring of content as well as authoring for customizing the design of gami ed ITS by teachers; (f ) De ne an ontology-based feature modeling approach to represent the con guration knowledge of the authoring platform which can be used to instantiate a speci c gami ed ITS. 3.
The literature review about the use of authoring tools to
build gami ed ITS was conducted in three different ways:
(i) analysis of the papers that propose authoring tools and
that are included in a recent book [
After performing these three steps, we have found seven
authoring tools that can be considered related to our
platform: ASSISTments [
In this section, we explain the speci c objectives that we have already performed and which activities we still need to execute. In addition, we will further explain how we intend to validate our proposal.
To perform the objective (a) we have rst studied several ITS theories and models (i.e., domain, student and pedagogical) and then we have adapted ITS ontologies available in the literature.
In order to complete the objective (b), mentioned in
Section 2, we have rst studied several theories related to
gami cation (e.g., Fogg's behavior model, Self-Determination
Theory, Reinforcement theory, ow theory and so on) to
understand the gami cation domain. Then, we speci ed a
domain ontology according to the Self-Determination
Theory (SDT) { since one of the main de nitions [
Afterwards, in order to achieve objective (c), we have rst
searched for empirical studies that report positive effects
on the use of gami cation in education from published
systematic literature reviews on the topic (e.g., [
To achieve the objective (d) we have connected gami cation concepts to ITS concepts in an integrated ontology (named Gami ed Tutoring Ontology - GaTO) based on the gami cation ontology and on the ITS ontologies speci ed in the execution of objectives (a), (b) and (c).
In order to achieve the objective (e) we have conducted the requirements engineering and architectural design phases for the authoring platform. We are currently implementing the speci ed architecture (75% of the implementation is already developed). It is noteworthy that some non-functional requirements are crucial to the design of the authoring platform, for instance, usability. As we are designing an authoring platform for teachers, this requirement drives several decision-makings, since it is of utmost importance to the effectiveness of the authoring platform.
The output of the authoring platform is a con guration
model that represents which features of a gami ed ITS should
be incorporated in the tutor authored by teachers. In this
way, to achieve objective (f) we de ned and validated an
ontology-based feature modeling (OntoSPL) approach [
Finally, after concluding the implementation of the authoring platform we will further validate our platform in three different ways. First, we intend to evaluate the usability (i.e, based on Nielsen's heuristics) for customizing gami ed ITS from teachers perspective in academic settings. Second, we intend to analyse the authoring platform integrated with a gami ed ITS platform aiming to characterize the effort for creating gami ed ITSs with respect to the time of creation and ease of use from a teacher viewpoint in industry settings (i.e., MeuTutor). Finally, we also intend to analyse gami ed ITSs designed by teachers using our authoring platform to characterize them with respect to motivation and learning performance from a student viewpoint in academic settings.
This PhD student is supervised by Professor Ig Ibert Bittencourt (Federal University of Alagoas). This work has been supported by the Brazilian institutions: Conselho Nacional de Desenvolvimento Cient co e Tecnologico (CNPq), Coordenaca~o de Aperfeicoamento de Pessoal de N vel Superior (CAPES).