You are on page 1of 5

Advance Online Learning System with the Incorporation of

Deterministic Finite Automata in an LMS

As-il, Jhaeus Lubrin,Allann Jake Villanueva,Raymart


University of the Cordilleras University of the Cordilleras University of the Cordilleras
+639472476464 +639274866937 +639213135488
jhaeus8@gmail.com allann.nirbul@gmail.com raymartvillaneuva11@gmail.co
m

ABSTRACT that have served as a channel to keep certain


areas available and functional that are necessary
Adaptive Online learning contributes to changing
for the development and subsistence of society.
the way instructional materials are presented
These tools are information and communication
based on the interests of the students. Since each
technologies (ICT), which have allowed most
student's comprehension ability is different, the
activities to be carried out remotely and securely
learning path should be tailored to their specific
[2]. It's worth noting that what happened has
needs. Few students can grasp concepts quickly,
altered our perspective on life.
while others may need further instruction in a
different manner.This paper takes a similar subject
With the use of ICT and new technologies, the
and introduces it in a variety of ways to different
world is experiencing a new normalcy that brings
groups of students of varying levels of
with it new challenges that must be overcome
comprehension. The proposed solution in this
and assumed by all sectors.Higher education is
paper is based on two concepts: Deterministic
one of the sectors of society that for several years
Finite Automata (DFA) and Case Based Study
has integrated these technologies into its
(CBS), with DFA providing adaptive nature and
activities.Despite the significant disadvantages,
displaying state transitions based on efficiency.
this integration has allowed education to
For a given subject, the learning path differs
continue. [3] However, it is necessary to recognize
depending on the student; this function is used to
the issues that have arisen and to implement
provide adaptive nature in E-learning. CBS was
broad education models that incorporate new
once known for offering state-specific research
and better technologies that allow students to
materials. CBS selects research materials from a
continue learning in any situation [4]. To achieve
case library.
this goal, it is necessary to revisit some previously
Keywords
overlooked concepts and tools.

Adaptive Online learning, Deterministic finite


For example, learning management systems
automata (DFA).
(LMSs) have waned in popularity in recent years
because some institutions regard them as simple
1. INTRODUCTION
repositories. Due to current circumstances, where
Currently, society is dealing with a health crisis that LMS is the medium that allows students to
has altered how people live. The Coronavirus maintain interaction with their institutions, this
disease 2019 (COVID-19) has revealed the fragility vision has taken a drastic turn [5].
of all areas, be they health, education, industrial,
etc. There is no part of society that has not been It is important to note that due to current
affected; however, it is the duty of universities and circumstances, the face-to-face educational
their research departments to work on all these model has been forced to transition to a virtual or
weaknesses and create robust models based on online educational model.Although this factor
what has been learned from this emergency [1]. appears to be easily absorbed by students, it
For this, it is necessary to take into account the makes a significant difference in a deeper
tools that have allowed us to combat this disease analysis. A change in study modality has a
and significant impact on students' performance, such
as a loss of interest and an inability to adapt, both
of which are factors that directly affect learning. solution[10]. The number of states in deterministic
Because he is the owner of knowledge, the finite automata is fixed, and students can only be
teacher is expected to be the main actor in in one state at a time. The transition of states in
DFA is determined by the input and state.DFA can
learning in a traditional education model [6].The
be represented by 5-tuples (Q, Σ, δ, q0, F) where
problems begin when this shifts and the student
becomes the primary actor in their education. ● Q is a finite set of states.
To address them, it is critical to create an ideal ● Σ is a finite set of input symbols called the
learning environment for students, one in which alphabet .
● δ is a transition function (δ : Q × Σ → Q).
they can access all of the resources they require.
● a start state (q0 Q).
In addition, systems that continuously monitor their
● a set of accepted states (F Q).
performance must be included [7], as well as the
inclusion of systems in charge of assigning Now Q is a set of states through which student’s
activities that are aligned with each student's pass in the learning process. Each state defines
characteristics. a student's level of knowledge in a specific
subject. Σ is a finite set of inputs, where the input is
the student's score, which is calculated in each
The integration of artificial intelligence (AI),
state based on the state's examination. Transition
machine learning, and a learning management δ represents change of knowledge or state. It is
system (LMS) is proposed in this paper to improve based on the previous state and the state's
an online education model, increasing the input.The initial state, q0, is when they have little
efficiency of machine learning through knowledge of the chapter. This process will be
deterministic finite automata and, as a result, repeated until the student has a sufficient
understanding of the subject and has reached
student learning.
the accepted state F. Score is calculated
depending upon two parameters:
Learning will become more popular as a result of
the increased use of educational material in web 1) Accuracy (A).
form in a web-based educational system.It 2) Marks of the particular Examination (M)
encourages students to investigate alternative
navigational paths using domain knowledge and A= 𝑛𝑜. 𝑜𝑓 𝑞𝑢𝑒𝑠𝑡𝑖𝑜𝑛𝑠 𝑎𝑡𝑡𝑒𝑚𝑝𝑡𝑒𝑑 * 100 (1)
𝑛𝑜. 𝑜𝑓 𝑞𝑢𝑒𝑠𝑡𝑖𝑜𝑛𝑠 𝑐𝑜𝑟𝑟𝑒𝑐𝑡𝑒𝑑

resources from around the world. It is impossible to M=


𝑚𝑎𝑟𝑘𝑠 𝑜𝑏𝑡𝑎𝑖𝑛𝑒𝑑
* 100 (2)
𝑡𝑜𝑡𝑎𝑙 𝑚𝑎𝑟𝑘𝑠
care for each and every student individually in our
S= % (3)
𝐴+𝑀

educational system, but online learning systems


2

provide individualized attention to each student


Inputs are divided into three ranges (R) according
based on their abilities.This paper employs
to Score(S).
Deterministic Finite Automata (DFA) to aid in the
R=
implementation of online learning in the LMS.They
{1 𝑖𝑓 0% ≤ 𝑆 ≤ 34% , 2 𝑖𝑓 35% ≤ 𝑆 ≤ 64% , 3 𝑖𝑓 𝑆 > 65%
taught the same topic in a different way unless
the students were completely prepared for the
(4)
chapter. It also looks after the students'
self-esteem by demonstrating progress over time.
The initial state q0 represents the start of online
A different DFA is used for each chapter. More learning for a specific topic. The initial state system
transitions are required for weaker students to examines the curriculum and depth of knowledge
reach the goal or final state, whereas two of students by administering an initial examination.
transitions are sufficient for fast learners. The system then calculates the Score of the
particular student and decides the student's next
state based on that score.
2. METHODS
(1) The first activity is that they are given study
material and video of lectures for a specific
It is important to define the environment in which
chapter based on their score. This decision can be
the various frameworks will be implemented in
made by teachers (if no previous cases have
order to progress with the project. It is possible to
been stored in the case library) or by referring to
decide the best way to integrate systems by
previously stored cases. CBS (Case Base Study)
evaluating the exact current conditions. In an
saves a lot of information for future use.It
online education model, the AI system that deals
compares similarities to find the solution method
with the LMS, and students is tweaked to maximize
for the most similar case. A CBS's task is to find the
learning.Deterministic Finite Automata (DFA) are
most similar case from the case library based on
used in this paper to implement E-Learning. Each
their score. CBS examines the Score and
chapter of the subject has its own DFA in this
compares it to previous cases. The best of these
cases is chosen, and the student is given the same 3. FINDINGS AND DISCUSSIONS
study materials and lecture videos. How do I pick
a case? The completeness of the students' It is possible to determine the best way to
learning and its assessment results will be integrate technologies by determining the exact
influenced by an important factor in case current conditions. In an online education model,
evaluation. The probability of success should be the AI system that works with the LMS and students
used to make a choice (SP). is tweaked to improve learning. The objective in
this paper is that curriculum sequencing is done
SP for i th Case and k th input range is represented by a subject expert and that the learning path is
determined by the students' skill, knowledge, and
𝑛
∑ 𝑆𝑗
by 𝑆𝑃𝑖𝑘= 𝑗 =𝑛1 where n is number of times the dedication.
particular Case 𝐶𝑖is used for kth range 𝑆𝑗 is the
When students enter the final state F⊆Q, it
Score of a particular student when the Case 𝐶𝑖is
indicates that they have finished learning for that
used for jth time. subject. For each chapter, the same activity will
be done. Each state activity is represented by a
(2) The corrective evaluation is the second action. control flow diagram. The activity done on each
There are two parts to this activity: state is represented in the control flow diagram
below.
a) Solving queries and doubts of the students.
b) Problem-solving exercises given to them, it Fig. 1 : Diagram of the control flow for each state's
covers the same concepts and material as activities
the first activity but asks questions in a
slightly different way or format.

(3) With the previous step, the system also


executes a motivational activity by displaying the
student's progress over time. Previous
accomplishment is stored state-by-state in the
database. The previous data was used to
determine the performance of a specific student.

(4) After completing all of the aforementioned


tasks, the system conducts an examination, from
which a score is computed. This score determines
the state's transition. However, if a student's
performance deteriorates, research documents
are re-verified to meet the needs of the student.

Figure 2 depicts DFA for an online learning on any


subject, where q0 is the initial state. At this level,
the system conducts one examination to
determine a student's depth of understanding in
terms of Score using formula 1, 2, 3. The range
given in formula 4 is then determined based on
the calculated score. The feedback to the state
q0 is range.

Fig 2: DFA for a particular topic


4. CONCLUSIONS

Using Deterministic Finite Automata, this paper


strengthens the adaptive design of an online
learning system. It assigns varying levels of effort to
various students. This system further motivates
students by praising them for their contributions. It
displays an overview of each student's success at
any given time.

This dissertation can be seen as a modern


perspective on Interactive Adaptive Environment
from a theoretical viewpoint. Where multiple
interacting states gradually build and modify the
learning to a subject. The key aim is to provide
students with in-depth awareness of each subject
before going on to the next.

The manual sequencing of the curriculum is


The above state transition diagram works as performed by a topic specialist. Curriculum
follows: sequencing is based on the concept of designing
an individualized course for each student by
Step 1: if input(𝑞0)= 1
dynamically choosing what will be achieved in
the future.
goto state 𝑞1
5. ACKNOWLEDGEMENTS
Perform all of the steps outlined in fig. 1’s
control flow diagram and quantify range, The researchers would like to thank the College of
then repeat activity q0 according to fig. Information Technology and Computer Science
2's state transition diagram until it enters staff, students and seniors of the University of the
the final state.
Cordilleras for their online assistance and support.

Step 2 : else if input (𝑞0 )= 2

goto state 𝑞2

Perform all of the steps in fig. 1's control


flow diagram and compute range, then
repeat the activity as q0 according to fig.
2's state transition diagram until it enters
the final state.

Step 3 : else goto stare 𝑞3

Perform all of the steps in fig. 1's control


flow diagram and compute range, then
repeat the activity as q0 according to fig.
2's state transition diagram until it enters
the final state.
https://www.mdpi.com/2071-1050/11/4/985

[8 ]Mario Muñoz-Organero, Member, IEEE, Gustavo


A. Ramírez, Pedro Muñoz Merino, and Carlos
Delgado Kloos, Senior Member, IEEE ” Analyzing
Convergence in e-Learning Resource Filtering Based
on ACO Techniques: A Case Study With
Telecommunication Engineering Students” IEEE
Transactions on Education, Vol. 53, No. 4, November
2010.

[9]Dr. Latha Parthiban and Saranya Sekar”Ant


Colony Optimization for Adaptive Learning “Indian
Journal of Computational Intelligence & Systems
REFERENCES Sciences,Volume 1, Number 1, January - June 2013,
pp.19-24

[1] Li, H.; Liu, S.M.; Yu, X.H.; Tang, S.L.; Tang, C.K. [10]Mario Muñoz-Organero, Member, IEEE, Gustavo
Coronavirus disease 2019 (COVID-19): Current status A. Ramírez, Pedro Muñoz Merino, and Carlos
and future perspectives. Int. J. Antimicrob. Agents 2020, Delgado Kloos, Senior Member, IEEE ” Analyzing
55, 105951. Convergence in e-Learning Resource Filtering Based
on ACO Techniques: A Case Study With
https://www.sciencedirect.com/science/article/abs/pii Telecommunication Engineering Students” IEEE
/S0924857920301011?via%3Dihub Transactions on Education, Vol. 53, No. 4, November
2010.
[2] Beldarrain, Y. Distance education trends: Integrating
new technologies to foster student interaction and [11]Herman Dwi Surjono “The Evaluation of a
collaboration. Distance Educ. 2006, 27, 139–153. Moodle Based Adaptive e-Learning System”
International Journal of Information and Education
https://www.tandfonline.com/doi/abs/10.1080/0158791
Technology, Vol. 4, No. 1, February 2014
0600789498
[12]Floriana Esposito, Oriana Licchelli, Giovanni
[3] Hssina, B.; Bouikhalene, B.; Merbouha, A. Europe and
Semeraro, “Discovering Student Models in e-learning
MENA Cooperation Advances in Information and
Systems,” Journal of Universal Computer Science, vol.
Communication Technologies; Rocha, A., Mohammed,
10, 2004,pg,no.47-57
S., Felgueiras, C., Eds.; Springer International Publishing:
Saidia, Marocco, 2017; Volume 520, ISBN [13]Manju Bhaskar, Minu M Das, Dr. T. Chithralekha, Dr.
978-3-319-46567-8. S. Sivasatya “Genetic Algorithm Based Adaptive
Learning Scheme Generation For Context Aware
[4] Villegas-Ch, W.; Lujan-Mora, S.; Buenano-Fernandez,
E-Learning” International Journal on Computer
D. Application of a Data Mining Method in to LMS for
Science and Engineering Vol. 02, No. 04, 2010,
the Improvement of Engineering Courses in Networks. In
1271-1279
Proceedings of the 10th International Conference of
Education, Research and Innovation (Iceri2017), Seville, [14]S. Gutierrez, G. Valigiani, Y. Jamont, P. Collet, and
Spain, 16–18 November 2017; pp. 6374–6381. C. Delgado, “A Swarm Approach for Automatic
Auditing of Pedagogical Planning,” in Proc. IEEE Int.
[5] Comendador, B.E.V.; Rabago, L.W.; Tanguilig, B.T. An
Conf. Adv. Learning Technol., Niigata, Japan, 2007,pp.
educational model based on Knowledge Discovery in
136–138.
Databases (KDD) to predict learner’s behavior using
classification techniques. In Proceedings of the IEEE
International Conference on Signal Processing,
Communications and Computing, Shanghai, China,
20–25 March 2016; pp. 1–6.

[6] Lee, S.J.; Lee, H.; Kim, T.T. A study on the instructor
role in dealing with mixed contents: How it affects
learner satisfaction and retention in e-learning.
Sustainability 2018, 10, 850.

https://www.mdpi.com/2071-1050/10/3/850

[7]Lee, J.; Song, H.D.; Hong, A.J. Exploring factors, and


indicators for measuring students’ sustainable
engagement in e-learning. Sustainability 2019, 11, 985.

You might also like