Villegas Ch2019
Villegas Ch2019
Villegas Ch2019
university learning
William Villegas-Ch Xavier Palacios-Pacheco Sergio Luján-Mora
Facultad de Ingenierías y Ciencias Departamento de sistemas, Department of Software and
Aplicadas, Universidad Internacional del Ecuador, Computing Systems,
Universidad de Las Américas, Quito, Ecuador Alicante, Spain
Quito, Ecuador [email protected] [email protected]
[email protected]
Abstract— Currently, universities seek to improve The proposed ES captures the knowledge of experts on
academic methods that come from a traditional model. different topics, as well as experts in education. This
Traditional models base learning on the experience of teachers functionality allows us to recommend suitable strategies that
and the development of activities. Studies carried out in the facilitate the scope and objectives proposed for a specific
educational field consider that not all activities generate university population [3]. The ES is able to interact with each
knowledge in students. The activities must adapt to the needs student and integrate into various computer systems to
of each student to build an efficient model that contributes to extract information to reach a conclusion. Students develop
learning. To solve this problem, education experts support different activities in learning management systems (LMS)
academic management in the use of information and
based on the resources provided by each teacher [4].
communication technologies. The inclusion of techniques such
as the analysis of educational data and artificial intelligence The information that the ES obtains from the LMS added
identify the deficiencies and needs of students. This work to the information obtained from the interaction with the
proposes the design of an expert system that interacts with student allows him to recommend specific activities to each
students and evaluates their responses with those of data student. Reaching this level of learning management
analysis systems to reach a conclusion and recommend guarantees effectiveness and can even nullify other problems
activities that align with the needs of each student. linked to student performance such as desertion. The paper is
divided as follows: Section II contains the theoretical
Keywords— artificial intelligence, expert systems, active
foundations that contribute to the design and implementation
learning, intelligent systems
of the proposed system. Section III contains the method in
which the entire development is systematically explained.
I. INTRODUCTION Finally section IV contains the conclusions obtained.
The contributions of information and communication
technologies (ICT) enrich the educational work. The II. THEORETICAL FOUNDATION
development of ICT demands a change in traditional learning
environments. The use of software and computerized A. Expert systems
educational materials as a resource to support teaching and ES use artificial intelligence techniques to respond
learning processes has become a necessity. In addition, they automatically to questions raised as if they were human
constitute a response to the problem that revolves around the experts. These systems are applied in many fields of
cognitive understanding of concepts and notions in knowledge due to their own nature and that are increasingly
classrooms [1]. Previous works analyze the advances that being perfected. In the educational field, they play a leading
help to learn by integrating new technological concepts in role due to the trends related to personalized education
education. A large percentage of research focuses especially through technological means [5].
on the integration of artificial intelligence (AI) in educational
processes. The inclusion of computer systems in universities
is increasingly necessary due to its flexibility and scalability B. Active learning
in educational management. An advantage of using ICT is Active learning involves the use of a set of more effective
the analysis of this data. This process allows detecting the and interesting experimental methods. With active learning,
academic problems that each student presents. Once the students assume greater responsibility for their own
problems are identified, methods and models can be education. It is characterized by very well structured and
generated to improve learning. challenging activities, with enough flexibility to adapt them
to the characteristics of the learning group and even at the
Several techniques allow the student to reflect on what individual level. The activities are organized so that the
they have learned. These techniques can be adapted to satisfy students develop them in face-to-face and virtual spaces or
diverse needs, everything depends on the tasks or activities the combination of them [6].
proposed to the student [2]. On the revealed ideas, it can be
considered that AI systems such as the expert systems (ES),
created for pedagogical and instructional purposes can III. METHOD
diagnose, debug and correct the development of student The process begins with the identification of the problem
learning. In addition, the system determines the cognitive that lies in the fact that not all students learn in the same way.
level of the student and helps him improve his weaknesses to There are several types of activities that are tailored to the
reach a higher level of learning. This work proposes the needs of each student, thus personalizing education. This
development of an ES that represents a response-oriented to educational method is known as active learning. The
the efficient decision making on teaching models and objective is to identify the needs of each student and
didactic activities. recommend a type of activity that meets these needs [7]. The
identification of needs, as well as the recommendation of
Developm
Playful
control
ent of
cases
the designer.
The validity of the results obtained allows highlighting
the role of the experts who contributed with their knowledge
Little time of 0 1 0.75 0 0 to deduce the relevant variables of each subject and its
development of importance in learning. With this contribution, it is possible
activities to carry out a multicriteria classification for the
recommendation of activities in a timely and efficient
Solving
manner. The result of the use of the ES allows us to
problems 0 0 0.75 1 1 recommend a variety of activities that contribute to active
improves your learning, using the multicriteria concept; it presents the result
learning of the optimal parameters of each activity based on the
Discussion on behavior of the students.
0 0 0 0.5 0
specific topics
REFERENCES
What I do I 0 0.25 0.75 1 0.75
understand [1] J. López, S. Martín, M. Castro and R. Meier, “Web-based platform
for the Information and communications technology (ICT) research in
What I teach to 0 0 0.25 0.75 0 engineering education,” IEEE Global Engineering Education
another I master Conference (EDUCON), Marrakech, 2012, pp. 1-6.
[2] H. Chenn-Jung, L. Ming-Chou, C. San-Shine and C. Chin-Lun,
“Application of machine learning techniques to Web-based intelligent
What I
0 0 0 1 0 learning diagnosis system,” International Conference on Hybrid
understand, I Intelligent Systems (HIS), Kitakyushu, 2004, pp. 242-247.
understand [3] E. El Haji, A. Azmani and M. El Harzli, “Expert system design for
educational and vocational guidance, using a multi-agent system,”
International Conference on Multimedia Computing and Systems
To identify the tasks that are recommended to students (ICMCS), Marrakech, 2014, pp. 1018-1024.
with little availability at the time of development of activities
[4] N. H. M. Ariffin, N. A. Alias, H. A. Rahman and J. Sardi,
and learn to solve problems, the system eliminates activities “Assessment of the students' utilization of a Learning Management
that do not meet these criteria by recommending activities System in a Malaysian higher education,” IEEE Conference on e-
with greater weight. For students who meet the conditions Learning, e-Management and e-Services (IC3e), Hawthorn, 2014, pp.
indicated after carrying out the process and calculation, the 18-23.
system will recommend the activities of games and [5] V. A. Suzdaltsev, I. V. Suzdaltsev and N. Y. Bogula, “Fuzzy rules
roundtables with the second being recommended with a formation for the construction of the predictive diagnostics expert
system,” IEEE International Conference on Soft Computing and
lower weight. Measurements (SCM), St. Petersburg, 2017, pp. 481-482.
[6] X. Jin-Peng, “Active learning for person re-identification,”
IV. CONCLUSIONS International Conference on Machine Learning and Cybernetics,
Xian, 2012, pp. 336-340.
Although universities have assigned areas that are [7] C. Wei-Fan, “Work in progress - creating an active learning
responsible for academic monitoring, however, educational environment for a computer networking class: expert systems
trends require greater use of ICT and automation of systems.
construction,” Frontiers in Education 35th Annual Conference,
Indianopolis, 2005, pp. S3C-15.
[8] M. Andergassen et al., “The evolution of e-learning platforms from
content to activity based learning: The case of Learn@WU,”
International Conference on Interactive Collaborative Learning
(ICL), Florence, 2015, pp. 779-784.
[9] R. Liu, T. Chen and L. Huang, “Research on human activity
recognition based on active learning,” International Conference on
Machine Learning and Cybernetics, Qingdao, 2010, pp. 285-290.
[10] M. Hayat, R. Hasan, S. I. Ali and M. Kaleem, “Active learning and
student engagement using Activity Based Learning,” International
Conference on Infocom Technologies and Unmanned Systems (Trends
and Future Directions) (ICTUS), Dubai, 2017, pp. 201-204.
[11] J. P. Held and J. V. Carlis, “Conceptual data modeling of expert
systems,” IEEE Expert, 1981, vol. 4, no. 1, pp. 50-61.
[12] S. WEISS, C.A. KULIKOWSKI, “Computer systems that learn:
classification and prediction methods from statistics, neural nets,
machine learning, and expert systems,” Morgan Kaufmann Publishers
Inc., 1991.
[13] Z. Yanqin, L. Fanzhang and H. Yuemei, “The design and application
of dynamic fuzzy expert database system,” IEEE International
Conference on Granular Computing, Beijing, 2005, pp. 378-381 Vol.
1.
[14] Y. Miao, “Modelling dynamic causal relationship in fuzzy cognitive
maps,” IEEE International Conference on Fuzzy Systems (FUZZ-
IEEE), Beijing, 2014, pp. 1013-1020.
[15] S. Thirumuruganathan and M. Huber, “Building Bayesian Network
based expert systems from rules,” IEEE International Conference on
Systems, Man, and Cybernetics, Anchorage, 2011, pp. 3002-3008.
[16] Y. Hoi-Kau and T. J. Richards, “GTKAT: a grounded theory based
knowledge acquisition tool for expert systems,” First New Zealand
International Two-Stream Conference on Artificial Neural Networks
and Expert Systems, Dunedin, 1993, pp. 152-155.
[17] G. Yanqiang, W. Xiangzheng and W. Aimin, “The research and
design of knowledge base on oracle expert system,” International
Conference on Transportation, Mechanical, and Electrical
Engineering (TMEE), Changchun, 2011, pp. 472-475.
[18] R. Ahmad and S. Rahimi, “A Perception Based, Domain Specific
Expert System for Question-Answering Support,” International
Conference on Web Intelligence (WI 2006 Main Conference
Proceedings)(WI), Hong Kong, 2006, pp. 893-896.