User Interactions Analysis on a Moodle-based Online Learning

Management System during Pandemic

Bernard Renaldy Suteja

and Wilfridus Bambang Triadi Handaya

Department of Information Technology, Universitas Kristen Maranatha, Surya Sumantri 65, Bandung, Indonesia

Department of Informatics, Universitas Atma Jaya Yogyakarta, Babarsari 44, Yogyakarta, Indonesia

Keywords: Academic, Moodle, Learning Management System, Pandemic.

Abstract: The Covid19 pandemic in Indonesia has provided a concise overview and practical understanding of how

technology might aid educational continuity. This pandemic presents a unique challenge for academic

innovation in establishing the optimum technology-based distribution strategy. Moodle was one of the most

popular LMS sites for providing an online learning experience during the pandemic. However, not all

educational institutions are subject to the government's online learning model. This study uses a sample

dataset from Moodle, a learning management system (LMS) platform used in the odd semester of 2020/2021.

Start on September 21, 2020, until February 5, 2021. During the study, data was collected from only three

departments at Universitas Kristen Maranatha. According to the statistics, the Informatics Engineering

undergraduate department student interacts with LMS for 17200 hours and 11455 times throughout a semester.

1 INTRODUCTION

In March 2020, most universities and educational

institutions shuttered to limit the COVID-19

outbreak, which was quickly followed by the launch

of online education (Alghamdi, 2021). However, as

previously reported by Abidah et al. (Abidah,

Hidaayatullaah, Simamora, Fehabutar, & Mutakinati,

2020), not all educational institutions are prepared to

face the government's online learning model.

It has been a year since all the learning completed

working online. Universities have adjusted all

restrictions to the use of technology as an alternative

learning technique, according to prior research by

Abidah et al. (Abidah et al., 2020). Online learning

processes are regulated by higher education

institutions, which include lecturers' growing grasp of

online learning technologies. Both asynchronous and

synchronous implementations should be usable and

well-assembled in the online learning environment.

In synchronous e-Learning, the educational and

learning process that students go through in a natural

context is virtualized. Technology such as electronic

board apps, live chat applications, and video

conferencing systems, all available through an LMS,

https://orcid.org/0000-0002-1178-8513

https://orcid.org/0000-0002-4394-7056

are used in the learning process. The learning activity

is timed, and everyone must be logged in at the same

time. Asynchronous e-Learning, on the other hand, is

a type of online learning. Students take part in learning

exercises that the teacher fictionalizes at their leisure,

and they have access to the resources that have been

provided to them (Ülker & Yılmaz, 2016).

As indicated in prior research by Deepak Kc (KC,

2017), students and lecturers gain from accessing the

LMS, the primary platform for lectures, directly

during learning activities, also known as an online

learning management system. Furthermore, the

system would serve as the principal means of

disseminating books, teaching, learning services, and

other online information. As stated in table 1, both

teaching techniques are based on Daniel Stanford's

quadrant of Internet bandwidth usage.

Using learning platforms, particularly Moodle,

can be driven from various viewpoints from an

academic standpoint. Individuals will, for example,

customize the appearance of the course, resulting in a

more pleasant and inspiring learning environment.

(Sayco Evale, 2017). On the other hand, formal online

education offers enormous potential to expand access

to higher education and diversify the student

314

Suteja, B. and Handaya, W.

User Interactions Analysis on a Moodle-based Online Learning Management System during Pandemic.

DOI: 10.5220/0010753100003113

In Proceedings of the 1st International Conference on Emerging Issues in Technology, Engineering and Science (ICE-TES 2021), pages 314-319

ISBN: 978-989-758-601-9

population. Furthermore, different students and levels

of education are transforming the learning

environment by including digital content and

activities that stimulate cooperation and

connectedness (Mohd Kasim & Khalid, 2016).

Table 1: Bandwidth Immediacy Matrix.

No Spectrum of

Internet connection

stren

Categories of Activities

1 Low immediacy and

high bandwidth

• Pre-recorded video

• Pre-recorded audio

• Asynchronous

Discussions with Video

• Asynchronous

Discussions with Audio

2 High immediacy

and high bandwidth

• Video conferences

• Audio conferences

3 Low immediacy and

high bandwidth

• Discussion boards with

text and images

• Readings with text and

images

• Email

4 High immediacy

and low bandwidth

• Collaborative

documents

• Group chat and

messa

This epidemic period will train and instill the habit

of becoming independent learners who continue to

build and improve online activities through various

online classes or webinars attended by students

(Sadeghi, 2019). In addition, students will work

together to solve problems and address real-world

difficulties. When it comes to imparting education,

this circumstance is difficult for both students and

lecturers, as lecturers must ensure that students

understand the topic. As a result, institutions

frequently employ open-source learning management

systems such as Moodle (Modular Object-Oriented

Dynamic Learning Environment) (Juárez Santiago et

al., 2020).

This study uses a sample dataset from Moodle, a

learning management system (LMS) platform used in

the odd semester of 2020/2021. Start on September

21, 2020, until February 5, 2021. This study examines

how MOODLE LMS's extensive data analysis may

be utilized to analyze the degree of interaction

between academic civitas in a department during

complete online learning.

2 METHODS

Any exceptionally vast and complex piece of data is

referred to as "big data." Because of their size, all of

these enormous data constitute a high-value

perspective on student behavior for numerous

education research domains, according to Fischer et

al. in a previously published (Fischer et al., 2020). It

is, therefore, critical to include it in the earliest and

continuous stages of learning (Ruiz-Palmero,

Colomo-Magaña, Ríos-Ariza, & Gómez-García,

2020).

The characteristics of Big Data are as follows:

a. Volume

There are many numbers to go through.

b. Velocity

Data is received at a rapid speed and can be used

instantly.

c. Variety

Traditional data styles are often standardized.

The amount of unstructured data grows in

tandem with the amount of structured data.

There are 185 tables in the Moodle Learning

Management System. One table, for example, is

utilized as log data for all activities performed when

using the LMS in online learning activities. As seen

in Figure 1, log data will become increasingly

valuable as LMSs are increasingly employed for

online learning. As a result, data can be collected,

aggregated, analyzed, categorized, and learned so that

it can be used to study the behavior of digital

technology users.

Figure 1: Data Log for Moodle.

2.1 Entity Relationship

It is essential to log in to analysis, reporting, and

analytics. Moodle can keep track of all user activity

data, allowing for intelligent, evidence-based

decisions. The researcher, for example, will look to

check if a user has accessed a specific course, signed

in, or logged out of the system.

This study analyzed learning data using Moodle

log data (Rapanta, Botturi, Goodyear, Guàrdia, &

Koole, 2020). Most of this study is focused on entity

relationships. The entity-relationship model is used

User Interactions Analysis on a Moodle-based Online Learning Management System during Pandemic

315

for logging data. As seen in figure 2, a user, who in

this situation could be an instructor, student, or

administrator, will be registered numerous times in

the log table.

Figure 2: Relationship between the Log Entity and the User.

The log table has ten fields: id, time, userid, ip,

path, module, cmid, action, url, and data. The

components that can be employed in this study are

time, userid, ip, and activity. The userid will be

associated with the user table to determine the

identity of the access viewer. The IP will be used to

determine the location of the user's access, and the

behavior involving the operation or contact from the

user to the LMS will all be used to determine the

access case. Figure 3 shows the log table structure of

the Moodle LMS database.

Figure 3: Structure of a Log Table.

2.2 Data Iteration

A UNIX-timestamp time field in the log table can be

utilized to read the log data. The Unix-time stamp

represents the number of seconds since Unix was

created on January 1, 1970. This Unix time must first

be converted before it can be used to compare the time

conditions of log data retrieval. In the analysis, IP

addresses were also employed as a component of

time. As a result, the IP address and time of the user

are utilized to group them. The overall number of user

interactions and their length will be calculated using

this classification. However, first and foremost, a

method for retrieving user interaction data from the

Moodle LMS is required:

1. View all log data recordings from September 21,

2020, until February 5, 2021.

Here is the SQL command that was used:

SELECT mdl_log.userid,

mdl_user.username,

mdl_user.firstname, mdl_user.lastname

FROM mdl_log, mdl_user WHERE

mdl_log.userid = mdl_user.id and

FROM_UNIXTIME (mdl_log.time,'%Y-%m-%d

%H:%i:%s') between :starttime and

:endtime group by mdl_log.userid order

by mdl_log.time

2. Each user id is utilized to construct time and IP

groupings from each iteration, allowing the

LMS system to calculate the number of

interactions from that IP.

Here is an example of a SQL command:

SELECT FROM_UNIXTIME

(mdl_log.time,'%Y-%m-%d'),

mdl_log.ip, count(mdl_log.ip) FROM

mdl_log WHERE

FROM_UNIXTIME(mdl_log.time,'%Y-%m-%d

%H:%i:%s') between :awal and :akhir

and mdl_log.userid=:id group by

FROM_UNIXTIME(mdl_log.time,'%Y-%m-

%d'),mdl_log.ip order by mdl_log.time,

mdl_log.ip

3. Perform the shelter's stage over a while and with

various user activities or interactions—formulas

for calculating time differences.

Δt = t1 – t0, as seen in figure 4.

Figure 4: Interaction time detection.

Where Ta is the amount of Δt:

First Iteration: Δt = t1 – t0 (1)

Second Iteration: Δt = t2 – t1 (2)

Where Tb is the sum of Δt:

First Iteration: Δt = t1 – t0 (3)

The total interaction time:

Ta + Tb (4)

The SQL command that was used:

SELECT mdl_log.action,

FROM_UNIXTIME(mdl_log.time,'%Y-%m-%d

%H:%i:%s') as wkt FROM mdl_log WHERE

FROM_UNIXTIME(mdl_log.time,'%Y-%m-

%d')=:wkt and mdl_log.ip=:ip and

mdl_log.userid=:id order by

mdl_log.time

Log User

has

ICE-TES 2021 - International Conference on Emerging Issues in Technology, Engineering, and Science

316

3 RESULTS AND DISCUSSION

This study aims to determine the number of

interactions and periods users communicate with the

LMS system. Detection can be used to evaluate the

interactions of administrators, instructors, students, or

a combination of these groups, all the way up to the

entire user. A visualization illustration from a

computer science postgraduate department is shown

in Figure 5.

Figure 5: Output samples for interaction time detection.

The instructor interacted with the device 11 times

during the odd semester 2020/2021. The most

interactions 71 times occurred on November 5, 2020,

for 15638 seconds, equivalent of 4.34 hours on the

day, and from an IP address of 36.71.232.99. Figure

6 displays a more detailed visualization.

Figure 6: Examples of user login activity logs.

Figure 7 shows the lecturer's involvement time

with the LMS system based on log data acquired from

the Computer Science postgraduate department.

Figure 7: Time spent interacting with the lecturer.

Figure 8 displays the length of time students spent

interacting with the LMS system.

Figure 8: Time spent by students interacting.

User Interactions Analysis on a Moodle-based Online Learning Management System during Pandemic

317

Figure 9 depicts the comparison of research from

the Computer Science postgraduate, Informatics

Engineering undergraduate, and Information Systems

undergraduate departments as a description of data

processing in the final stage of data comparison of

three undergraduate and postgraduate related to the

university LMS. The results acquired reveal valid

data from the Universitas Kristen Maranatha

Learning Management System for a total of 17200

hours and 11455 times, the majority of which were

conducted by students from the Informatics

Engineering undergraduate department.

Time Count

s2ilkom 4955695 637

s1if 62026251 11455

s1si 17546200 4167

Figure 9: Three departments' duration of time and activity.

4 CONCLUSIONS

In the odd semester of the 2020/2021 academic year,

the research was only performed in three

departments. Nonetheless, using analytical data from

MOODLE Learning Management System log files to

predict user interaction activities can help educational

institutions detect and improve LMS effectiveness.

This study's findings can be used to help higher

education institutions integrate online learning.

Internally managed LMS across departments can help

implement these policies by integrating all

knowledge resources into a single system. In addition,

the strategy to permit hybrid learning during the

pandemic will undoubtedly influence educational

institutions in selecting the optimum approach to

organize the educational process in their respective

institutions based on policy readiness, supporting

infrastructure, and human resources.

REFERENCES

Abidah, A., Hidaayatullaah, H. N., Simamora, R. M.,

Fehabutar, D., & Mutakinati, L. (2020). The Impact of

Covid-19 to Indonesian Education and Its Relation to

the Philosophy of "Merdeka Belajar". Studies in

Philosophy of Science and Education, 1(1), 38-49.

doi:10.46627/sipose.v1i1.9

Alghamdi, A. A. (2021). Impact of the COVID-19

pandemic on the social and educational aspects of Saudi

university students' lives. PLoS One, 16(4), e0250026.

doi:10.1371/journal.pone.0250026

Fischer, C., Pardos, Z. A., Baker, R. S., Williams, J. J.,

Smyth, P., Yu, R., . . . Warschauer, M. (2020). Mining

Big Data in Education: Affordances and Challenges.

Review of Research in Education, 44(1), 130-160.

doi:10.3102/0091732x20903304

Juárez Santiago, B., Olivares Ramírez, J. M., Rodríguez-

Reséndiz, J., Dector, A., García García, R., González-

Durán, J. E. E., & Ferriol Sánchez, F. (2020). Learning

Management System-Based Evaluation to Determine

Academic Efficiency Performance. Sustainability,

12(10). doi:10.3390/su12104256

KC, D. (2017). Evaluation of Moodle Features at Kajaani

University of Applied Sciences – Case Study. Paper

presented at the 2nd International Conference on

Computer Science and Computational Intelligence

2017, Bali, Indonesia.

Mohd Kasim, N. N., & Khalid, F. (2016). Choosing the

Right Learning Management System (LMS) for the

Higher Education Institution Context: A Systematic

Review. International Journal of Emerging

Technologies in Learning (iJET), 11(06).

doi:10.3991/ijet.v11i06.5644

Rapanta, C., Botturi, L., Goodyear, P., Guàrdia, L., &

Koole, M. (2020). Online University Teaching During

and After the Covid-19 Crisis: Refocusing Teacher

Presence and Learning Activity. Postdigital Science

and Education, 2(3), 923-945. doi:10.1007/s42438-

020-00155-y

Ruiz-Palmero, J., Colomo-Magaña, E., Ríos-Ariza, J. M.,

& Gómez-García, M. (2020). Big Data in Education:

Perception of Training Advisors on Its Use in the

Educational System. Social Sciences, 9(4).

doi:10.3390/socsci9040053

Sadeghi, M. (2019). A Shift from Classroom to Distance

Learning: Advantages and Limitations. International

Journal of Research in English Education, 4(1), 80-88.

doi:10.29252/ijree.4.1.80

Sayco Evale, D. (2017). Learning Management System

with Prediction Model and Course-content

Recommendation Module. Journal of Information

ICE-TES 2021 - International Conference on Emerging Issues in Technology, Engineering, and Science

318

Technology Education: Research, 16, 437-457.

doi:10.28945/3883

Ülker, D., & Yılmaz, Y. (2016). Learning Management

Systems and Comparison of Open Source Learning

Management Systems and Proprietary Learning

Management Systems. Journal of Systems Integration,

18-24. doi:10.20470/jsi.v7i2.255

User Interactions Analysis on a Moodle-based Online Learning Management System during Pandemic

319