STSIM: Semantic-web Based Tool to Student Instruction Monitoring
H
´
ector Yago Corral
1
and Julia Clemente P
´
arraga
2
1
Escuela T
´
ecnica Superior de Ingenier
´
ıa Inform
´
atica, Universidad de Alcal
´
a, Campus Universitario,
Alcal
´
a de Henares, Spain
2
Departamento de Autom
´
atica, Universidad de Alcal
´
a, Campus Universitario, Alcal
´
a de Henares, Spain
Keywords:
Semantic Web, Student Ontology, Student’s Learning, Monitoring.
Abstract:
In this article, a tool so-called STSIM is presented. It is able to monitor the student’s progress along learning
experiences. This tool, based on semantic web, allows students and teachers to monitor the knowledge
student state including, among others, the learning objective state -achieved and not achieved- in different
types of activities with psychomotors, cognitives or affective competences, and the efficiency accomplished
in activity execution to facilitate the tutor or student the supervision of learning in a more adaptive way
according to the individual characteristics and student knowledge state in each moment. To achieve this
goal, STSIM uses a flexible student model supported by an ontology network, the Student Ontology. The tool
has been developed to be multiplatform, multilingual, based on current and open-source web technology and
characterized by its usability. STSIM is built on the UML-based web engineering (UWE) methodology and
the Model-View-Controller (MVC) pattern.
1 INTRODUCTION
The technological advances have encouraged a bright
evolution in many areas including the educational
arena. In this way, one of the mostly researched
aspects is the monitoring which, in general, consists
of observing a situation, process, event, etc. through
a receptor in order to check the quality and discover
anomalies. In education, monitoring is a method
that constantly analyses the student’s evolution in
correspondence with the proposed objectives. It
allows teachers to ensure the process’s direction; the
software/human tutor can have a more accurate vision
of the student’s knowledge and, therefore, can take
more informed and personalized tutoring decisions.
In this sense, it is important to highlight
the difference between monitoring and supervision
because both terms are often used interchangeably.
Monitoring tries to find out if the student’s
evolution corresponds to the proposed objectives
and, otherwise, the tool allows the responsible to
take an adequate decision to correct the variations.
Supervision is responsible for monitoring and making
proper decisions.
A monitoring system is usually made up of the
following components:
• Indicators. Measures summarizing the data.
• Record. Tools and methods to collect
information.
• Interpretation. It analyses the stored data and
• Visualization. It shows the information in a
specific instant.
The indicators, record and interpretation are
necessary in all monitoring system. Nevertheless,
the visualization is an optional component in all
monitoring systems (Sampieri, 2008).
The tool explained in this article tries to progress
in improving the quality of the educational system
through the student’s learning monitoring taking
into account so many information sources such as
the interaction possibilities derived from the use
of Virtual Environments (VEs) to provide valuable
information on the student’s kwnowlede state, the
student’s behavior along different learning sessions,
etc., so that the tutor can make sensible tutoring
decisions or provide the most suitable feedback to the
student in each moment. Thanks to it, STSIM may
be easily used for differents learning environments
and contexts in the short term. Furthermore, it is
based on the semantic web process, specifically in
the use of ontologies. There are multiple definitions
about ontologies, but one of the most popular is the
following: ”An ontology is a formal and explicit
276
Yago Corral H. and Clemente Párraga J..
STSIM: Semantic-web Based Tool to Student Instruction Monitoring.
DOI: 10.5220/0004841402760284
In Proceedings of the 6th International Conference on Computer Supported Education (CSEDU-2014), pages 276-284
ISBN: 978-989-758-020-8
Copyright
c
2014 SCITEPRESS (Science and Technology Publications, Lda.)
specification of shared conceptualization” (Gruber
et al., 1993).
This tool is supported by the Student Ontology
(Clemente, 2011) consisting of an ontology
network or, in other words, a collection of
ontologies connected by some relationships such us
mapping, modularization, version and dependency
(Su
´
arez-Figueroa et al., 2008). Student Ontology
contains a wide range of types for modelling student
in an Intelligent Tutoring system (ITS) and other
complex learning environments such as the so-called
Intelligent Virtual Environments for Training and/or
Instruction (IVETs). The data registered in terms
of the Student Ontology allows to carry out a
pedagogical-cognitive diagnosis with non-monotonic
reasoning capacities, that is able to infer the state of
the learning objectives encompassed by the ITS and
correspondingly infer the student’s knowledge state
(Clemente et al., 2013).
The article begins with a brief description of
some important related work on the student’s learning
monitoring. The paper continues with a description
of the adopted solution including a general overview
of its architecture, design based on MVC pattern
and technologies involved. Besides, some details on
both the goals of present work and Student Ontology
structure are given. Next, an application example is
described. The paper ends with the main future work
lines and conclusions.
2 APPROACHES TO
MONITORING STUDENTS’
LEARNING
The student’s learning monitoring has been a highly
researched topic since the last 20
th
century when
12 modules for validating new technologies were
identified including the monitoring (Zelkowitz et al.,
1998).
Currently, there are some works closely related
to student’s learning monitoring which are worth
emphasizing such as: a) the theorical study and
the tool about the progress of student’s learning
(Sampieri, 2008). The tool consists of two modules
supported by a database that can be used by teachers
and student providing feedback to them through
different graphs about the mark and efficiency during
the course. b) the approach of OeLE platform
(S
´
anchez-Vera et al., 2012). OeLE tries to evaluate
the answers to open questions and to give feedback
to teachers and students. Some of the most important
characteristics of OeLE are the ability of monitoring
the learning objective state and the use of ontologies.
Another important research line in this study
is the feedback since it is essential that teachers
and students receive suggestions to improve the
teaching/learning process. From this perspective,
we should highlight the work about data mining
(Dyckhoff et al., 2011) involving the tool eLat. This
tool offers support to teachers in the process of
improving the efficiency in the group. It examines the
Table 1: Analysis of student’s learning monitoring tools.
Tools Author Year Technology Objectives
Monitoring & feedback Supervision Indicators
Future
potential
Student Teacher Human Software Mark Efficiency Objectives
eLat Dyckhoff,
A.L. et
al.
2012 Data
mining
Improve
the course
effectiveness
NO YES YES NO Participation in forums and
number of request to content.
Medium
OeLE S
´
anchez,
M.d.M.
et al.
2012 Ontologies Evaluate the
answers to
open questions
YES YES YES YES NO NO YES High
Check
my
activity
Fritz, J. 2011 Blackboard Provide
comparative
reports to
students
YES NO YES NO YES NO NO Medium
LiMS Sorenson,
P. et al.
2010 Web Extract the
achieved
objectives by
the student
NO YES YES NO NO NO YES Medium
ETR Sampieri,
M.
2008 Database Monitor the
student’s
learning
YES YES YES NO YES YES NO Medium
STSIM:Semantic-webBasedTooltoStudentInstructionMonitoring
277
number of mails sent by a student or the number of
times that a student access to content of the subject
and gives feedback to the teacher concerning the
student’s results. In this line, Check my activity was
created (Fritz, 2011). This tool offers the students
some comparative reports about the answers given by
their classmates in an anonymous way. The aim of
this tool is that the students receive feedback and ask
for help when they need it.
Other interesting proposal that has been
researched is LiMS (Sorenson and Macfadyen,
2010). This tool consists on the extraction of the
objectives achieved by the students through contents
published on the web.
In the table 1 a comparative analysis of some
main features (year, used technology, objectives,
monitoring and feedback, supervision, different
indicators, future potential) of the mentioned tools
related to student’s learning monitoring as well as
several main projects developed in the last five years
(2008-2013) are shown.
As seen in Table 1, the related tools use different
technology in order to profit from the advantages of
each one. eLat use data mining in order to extract
information to improve the course effectiveness.
OeLE is the only tool based on ontologies in order
to evaluate the answers to open questions. Check
my activity uses the platform Blackboard to provide
information to students. LiMS searches on the web to
get the objectives achieved by students and ETR uses
common databases to monitor the student’s learning.
Moreover, the analysed tools have different
monitoring and feedback targets. OeLE and ETR
provide monitoring and feedback to teachers and
students. Nevertheless, eLat and LiMS only provide
the information to teachers and Check my activity
gives information only to students.
Regarding supervision, it is important to
emphasize that all tools allow the human supervision.
However, only OeLE allows also the software
supervision. This idea is essential because it offers a
great potential in the future because mistakes can be
detected and corrected more easily.
Another important aspect is the indicators which
help teachers to understand the students. The
mark is only measured in ETR and Check my
activity. However, the efficiency is more exceptional
because it is only measured in ETR. The objectives
are considered in the OeLE platform and LiMS,
respectively. Finally, eLat includes other indicators
like participation in forums or number of requests to
content.
3 PROPOSED SOLUTION
Our proposed solution to a monitoring tool is a
Semantic-web based Tool to Student Instruction
Monitoring (STSIM). It is a Java web application
using a Model-View-Controller pattern. MVC
facilitates the application’s development (Leff and
Rayfield, 2001), dividing the tool into three
components:
• Model. It communicates the application with the
Student Ontology through the Jena framework
1
and SparQL query language (Prud’Hommeaux
et al., 2008).
• Controller. It contains the application logic
communicating the Model with the View. It
1
http://jena.apache.org/.
Figure 1: STSIM general architecture.
CSEDU2014-6thInternationalConferenceonComputerSupportedEducation
278
is implemented in ZK
2
, a event-driven and
component based pattern framework.
• View. It lets the user (teacher or student) request
information from the Model and later, it generates
an output representation to the user in several
visual formats (graphics, tables, or plain text).
In the Figure 1, the different application
components of STSIM and their connections are
presented. The user interacts with the application
Views through a web navigator. The Controller
catches different actions from the users and requests
to Model the required information. The Model
consults the Student Ontology about the information
which is sent to a new view in order to be visualised
by the user.
STSIM is a web application built using UWE
methodology. UWE (Koch and Kraus, 2002) is a web
extension of the UML modelling standard.
The developed application tries to monitor the
student’s learning process in a subject matter based
on an instructional design. It implies defining a
group of activities and the objectives that the student
should achieve in each activity. The relationship
between learning objectives and the knowledge
objects involved in a course is stored in Student
Ontology. This representation is fundamental because
it will allow a monitoring with different granularity
levels; a monitoring of the reached or not reached
learning objective states (coarse-grained monitoring)
or a monitoring of student’s specific knowledge state
(fine-grained monitoring). In this way, monitoring
the student’s learning evolution provides greater
assistance in the generation of a personalized plan for
each student.
In STSIM, a instructional design of a course is
used, which implies defining a group of activities for
2
http://www.zkoss.org/.
the subject matter to be taught and the objectives that
the student should achieve in each activity.
As well as the objectives, other aspects can be
monitored such as the mark in an activity or the
efficiency in a course or activity. All these options
are offered in the tool in order to provide teachers and
students a more complete and better feedback about
the student’s learning.
3.1 The Student Ontology
The Student Ontology is, in fact, an ontology network
composed originally by seven ontologies (Clemente
et al., 2011). It was developed using the Prot
´
eg
´
e
editor
3
and the ontology language OWL to be used,
among others, in IVETs. It has been extended
as support of the monitoring tool presented here
with a new ontology so-called Tutoring Information
which contains information about teachers, activities,
subjects and their relationships. Therefore, this
ontology network is composed by the following main
ontologies:
• Tutoring Information. It includes, among
others, the information about the student groups
created in a course for a certain subject, teachers,
modules or activities belonging to a subject,
etc. In the Figure 2, the conceptual model
of Tutoring Information can be observed. It is
composed by seven classes: (a) Activity contains
the activity weight, required and achieved
objectives and the module where it is located. (b)
Course provides information about the associated
curriculums and subjects. (c) Curriculum stores
the syllabus it belongs to, its courses, etc. (d)
Module offers information about its activities,
objectives required and achieved by the module,
3
http://protege.stanford.edu/.
Figure 2: Conceptual model of Tutoring Information.
STSIM:Semantic-webBasedTooltoStudentInstructionMonitoring
279
etc. (e) Student Group contains the data about
timetable, teacher, students, etc. (f) Subject
includes type, modules, etc. (g) Teacher provides
some personal data, student groups who he/she
teaches, etc.
• Learning Objective. It represents the learning
objectives defined in an educational process
at cognitive, psychomotor or affective level.
It is divided into Didactic Objective and
Specific Objective.
• Knowledge
Object. It depicts a knowledge
element which can be learned in a particular
educational process. This ontology does not
depend on any other.
• Student State. It describes the student’s
knowledge, the acquired learning objectives,
the degree of completion of the instructional
design of the course, and an assessment of
the student’s performance throughout different
learning sessions.
• Student Information. It is created as an
aggregate of all the information specific for
each student. It includes the Student Profile,
Student Monitoring, Student State and
Student Trace ontologies. This ontology is also
related to the new ontology Tutoring information.
• Student Monitoring. This single ontology
allows us to define varied monitory strategies
for the different variables that the tutor may be
interested in monitoring (position of a student in
the Virtual Environment, student’s gaze direction,
etc.).
• Student Profile. It contains some personal
information about the students (demographic
data, preferences, physical and psychological
features, etc.).
• Student Trace. It describes the temporal registry
of the student’s activity during his learning
experience in a subject.
A more detailed description of the above
ontologies can be found in (Clemente et al.,
2011). Additionally, a new simple ontology
and the relationships with previous ontologies
have been added to the ontology network:
Tutoring Information.
Furthermore, it is important to highlight that
the Student Ontology was built using a modular
network with the methodology NeOn (G
´
omez-P
´
erez
and Su
´
arez-Figueroa, 2009) because currently it is
the only one that allows the development of ontology
modular networks.
In the Figure 3, the ontology relationships
between the different components of Student
Ontology can be observed. The new ontology and
its relationships with the previous already existing
one, created in this work, (Tutoring Information,
Learning Objective and Student Information)
appears in red color. The original ontology network
is shown in blue color.
Figure 3: Ontology modular network in STSIM.
CSEDU2014-6thInternationalConferenceonComputerSupportedEducation
280
3.2 Objectives
The use of this tool is intended to provide teachers
and students with some information
4
which lets them
prepare a specific learning plan for each student
according to their current knowledge state and their
particular characteristics.
Furthermore, other important objectives pursued
with the development are:
• Extensibility. The application is characterized
by the extensibility in the used student modelling
(Student Ontology) and in the application
development.
• Multiplatform. It represents the possibility of
using the tool in different Operating Systems and
web browsers.
• Multilingual. The tool should be provided in
different international languages and it offers the
possibility to add new languages easily.
• User-friendly. The users of this platform will
only log into the web application. Moreover,
the application will offer different alternatives of
help such as pop-ups and an user’s manual in
order to show the user all the possibilities to take
advantage of the monitoring application.
• Based on Semantic Web. It is intended to use the
semantic web technology, in particular, ontology
technology since it offers reusability, extensibility
and the possibility to infer knowledge that, in the
future, could help in the supervision task.
4
In the current initial state of STSIM tool, the large
amount and diversity of stored information in the used
ontology network is not being exploited more than at very
low level for monitoring student’s learning evolution.
4 TOOL EXAMPLE
In this sense, the learning objectives associated with
the activity Development of a minishell have
been previously defined within a pedagogical design.
The initial state for the objectives in each learning
activity depends on several factors such as tutoring
strategy, student’s background, whether the objectives
have already been reached in previous activities, etc.
We suppose that the student has not already acquired
the required theoretical knowledge to do the practice
so, we assume the initial state acquired=false for the
objectives. Besides, the Student Ontology provides
instances about the knowledge objects involved in the
course Operative System; the dependencies between
the activities objectives and the knowledge objects;
students’ profile; subject information (modules,
activities, teacher(s), student’s groups, etc.). The
students answer questions and, consequently, some
rules are fired and the ontology content is updated
with the new objective states (achieved or not
achieved) (Clemente et al., 2013). Thereafter, we can
see in Figure 4 and Figure 5, STSIM allows users
(student/subject teachers) to see student’s learning
monitoring data though graphics, tables and plain
text. Also, STSIM presents a great potential in the
near future taking advantage of ontology inference
capability, specifically, inferring from the information
stored in the Student Ontology.
The Figure 4 shows four groups of bars. The
first is the mark percentage, the second is the correct
answers percentage, the third is the incorrect answers
percentage and the forth is weight percentage which
has been obtained by the student Antonio Mart
´
ın
P
´
erez in activity Development of a minishell of
Operating Systems subject on a degree course in
Computer Science. The red bars represent the student
attributes, the blue bars show the average values of the
Figure 4: Mark obtained by a student.
STSIM:Semantic-webBasedTooltoStudentInstructionMonitoring
281
Figure 5: Learning objective states of a particular student.
student group, the green bars depict the average of all
the students that have carried out the activity and the
white bar stands for the activity weight on the subject.
The Figure 5 represents the objectives achieved
and not achieved by the student Antonio Mart
´
ın
P
´
erez in the activity Development of a minishell of
Operating Systems subject on a degree course in
Computer Science. At the top of the figure, a pie
chart with the achieved and not achieved objectives is
shown and at the bottom of the figure, a table indicates
the specific objectives achieved and not achieved by
the student in the above-mentioned activity.
Despite teachers and students can monitor
objective state, the teacher has more options for
monitoring because it can monitor a student, a student
group or all groups of a subject taught by him.
However, the student can only monitor himself and
obtain the averages of his student’s class.
5 FUTURE WORK
With the tool presented in this article, we intend to
open several lines of future work. The first and
most important line of research consists on using
automated tools such as planners (Plaza et al., 2008)
may provide support for the planning and supervision
of the student’s learning evolution taking to support
monitoring output of STSIM tool.
Another working lines are related to enhancing the
tool development:
• From a functional point of view: a) the
monitoring tasks can be extended with other
key indicators to monitor the student’s learning
process. It includes, information such as
the relationship between the objective states
(achieved or not achieved) and its associated
learning objects. Likewise, from this focus,
using the tool in Intelligent Virtual Environments
(IVE) provides teachers much information about
student’s knowledge states based on information
related to these environment types already
registered in terms of Student Ontology. b)
Using semantic technology to infer additional
knowledge from the information stored in the
ontology allows the teacher to adopt tutoring
decisions more adaptable to the particular
characteristics and knowledge states of each
student at every moment of their learning. c)
Carring out a survey of accessibility and usability
of the tool developed using standard techniques
and tools. From the previous analysis, adequately
improve in the tool (Slatin and Rush, 2003). d)
Extending the multilingual capability of STSIM
(currently it is offered in English, French, German
and Spanish). e) Adapting the web application
to be used in different environments and types of
activities like forums, physical tests, etc.
• From a structural point of view: extending the
ontology of the student in some weak points.
For example, in learning objective ontology,
student profile ontology (personal characteristics
that influence student’s learning), or other aspects
not yet covered as tutoring strategies. In this
line, perhaps other ontological or not ontological
resources currently existing in repositories, etc,
could be reused.
CSEDU2014-6thInternationalConferenceonComputerSupportedEducation
282
The great potential offering this project is the
future line implementation because the technological
advances in semantic web are able to provide
improvements in the educational field. In a future,
this tool will be able to be implemented into different
environments and work in a collaborative way with
other ontologies and monitoring tools.
6 CONCLUSIONS
This article has described a solution to monitor
the student’s learning process. The general goal of
this work has been the development of a monitor
tool so-named STSIM, based on web and ontology
technology with the following main characteristics:
available for teachers and students, multilingual,
multiplatform, easily extensible, user-friendly
and developed using the framework ZK, a Web
application framework based on patterns and events,
and Jena framework.
Besides, it is worth mentioning the importance
of monitoring as information source to the human
supervision (tutor or the student throughout his
learning) or software supervision because it has a
great potential to detect weaknesses in the student’s
learning process using ontological inference and
monitoring information.
We should emphasize the importance of the
use of ontologies and its advantages, including the
the ability of inference from their knowledge. It
can benefit and enrich greatly the monitoring and
supervision of student’s learning and, ultimately,
encourage advance towards the improvement of
educational processes, essential goal of our work.
A wide representation of information relating
to complex environments such as the Virtual
Environments for Training/Instruction, whose
benefits have been proven in the field of education
(Mantovani, 2001) and, specifically, the IVETs,
already exists in the ontology network used in
STSIM tool that can be exploited and extended in the
future to achieve the final goal.
REFERENCES
Clemente, J. (2011). Una propuesta de modelado
del estudiante basada en ontolog
´
ıas y diagn
´
ostico
pedag
´
ogico-cognitivo no mon
´
otono. PhD thesis,
Universidad Polit
´
ecnica de Madrid.
Clemente, J., Ram
´
ırez, J., and De Antonio, A. (2011). A
proposal for student modeling based on ontologies
and diagnosis rules. Expert Systems with Applications,
38(7):8066–8078.
Clemente, J., Ram
´
ırez, J., and De Antonio, A. (2013).
Applying a student modeling with non-monotonic
diagnosis to intelligent virtual environment
for training/instruction. Expert Systems with
Applications.
Dyckhoff, A. L., Zielke, D., Chatti, M. A., and Schroeder,
U. (2011). elat: An exploratory learning analytics tool
for reflection and iterative improvement of technology
enhanced learning. In EDM, pages 355–356.
Fritz, J. (2011). Classroom walls that talk: Using online
course activity data of successful students to raise
self-awareness of underperforming peers. The Internet
and Higher Education, 14(2):89–97.
G
´
omez-P
´
erez, A. and Su
´
arez-Figueroa, M. C. (2009).
Neon methodology for building ontology networks: a
scenario-based methodology.
Gruber, T. R. et al. (1993). A translation approach
to portable ontology specifications. Knowledge
acquisition, 5(2):199–220.
Koch, N. and Kraus, A. (2002). The expressive
power of uml-based web engineering. In Second
International Workshop on Web-oriented Software
Technology (IWWOST02), volume 16. CYTED.
Leff, A. and Rayfield, J. T. (2001). Web-application
development using the model/view/controller design
pattern. In Enterprise Distributed Object Computing
Conference, 2001. EDOC’01. Proceedings. Fifth
IEEE International, pages 118–127. IEEE.
Mantovani, F. (2001). VR learning: Potential and
challenges for the use of 3d environments in
education and training. Towards CyberPsychology:
mind, cognition and society in the internet age,
2(1ntroduction):207.
Plaza, J., MD, R., Moreno, B. C., Carbajo, M., and Moreno,
A. (2008). PIPSS: Parallel integrated planning and
scheduling system. In The 27th Annual Workshop
of the UK Planning and Scheduling Special Interest
Group (PLANSIG08), Edinburgh (UK).
Prud’Hommeaux, E., Seaborne, A., et al. (2008). Sparql
query language for rdf. W3C recommendation, 15.
Sampieri, M. (2008). Monitorizaci
´
on del progreso en
el aprendizaje. PhD, Polythecnic University of
Cata–lonia.
S
´
anchez-Vera, M. d. M., Fern
´
andez-Breis, J. T.,
Castellanos-Nieves, D., Frutos-Morales, F., and
Prendes-Espinosa, M. P. (2012). Semantic web
technologies for generating feedback in online
assessment environments. Knowledge-Based
Systems, 33:152–165.
Slatin, J. and Rush, S. (2003). Maximum accessibility:
making your web site more usable for everyone.
Addison-Wesley Professional.
Sorenson, P. and Macfadyen, L. P. (2010). The learner
interaction monitoring system (lims): Capturing
online learner behaviour. In World Conference on
E-Learning in Corporate, Government, Healthcare,
and Higher Education, volume 2010, pages
2725–2725.
Su
´
arez-Figueroa, M. C., Fern
´
andez-L
´
opez, M.,
G
´
omez-P
´
erez, A., Dellschaft, K., Lewen, H.,
and Dzbor, M. (2008). Neon d5. 3.2 revision and
STSIM:Semantic-webBasedTooltoStudentInstructionMonitoring
283
extension of the neon development process and
ontology life cycle. NeOn project. http://www.
neon-project. org.
Zelkowitz, M. V., Wallace, D. R., and Binkley, D. (1998).
Culture conflicts in software engineering technology
transfer. In NASA Goddard Software Engineering
Workshop, page 52. Citeseer.
CSEDU2014-6thInternationalConferenceonComputerSupportedEducation
284
