Upload
others
View
8
Download
0
Embed Size (px)
Citation preview
Volume 5 | Number 1 | October 2020
E-ISSN : 2541-2019P-ISSN : 2541-044X
JURNAL DAN PENELITIAN TEKNIK INFORMATIKA
Dipublikasi oleh :
PREFACE
Praise the presence of Allah SWT, for blessings and mercy, in the successful
publication of the Journal SinkrOn Volume 5 Number 1 for the October 2020 Period which
refers to the rule for journal writing determined by Polytechnic Ganesha Medan.
SinkrOn as a media publication and a place to share scientific work in the field of
Informatics Engineering which is published twice in a year, namely in October and April,
and this edition is filled with contribution from a various institution in parts of Indonesia
with research fields related to the field of Informatic Engineering from lectures and graduate
students.
In this edition and beyond, SinkrOn is committed to continuously improving the
quality of journals to be published. SinkrOn substantially tightens the entire review process
to publish quality journals. Currently, SinkrOn has been accredited by RistekBRIN Sinta 3
and indexed DOAJ. SinkrOn has the vision to continue to improve accreditation to reach
Sinta 1 in the next few years. Besides, SinkrOn also has the vision to enter the realm of
international journals with the hope of reaching writers and readers from all over the world.
In this edition, Sinkon has also involved several editors and reviewers from various countries
such as Indonesia, Malaysia, India, Iraq, and Italy.
Our deepest gratitude to the authors, reviewers, editors, and all parties involved in
publishing this journal. Hopefully, this journal can provide good benefits for all academics
in the field of Informatics Engineering and still waiting for your brilliant work in the next
edition.
Medan, 27 November 2020
Editor in Chief
Thanks to Reviewers
Mohammad Andri Budiman (Universitas Sumatera Utara, Indonesia)
Dian Rachmawati (Universitas Sumatera Utara, Indonesia)
Sriadhi (Medan of University, Medan, Indonesia)
Dicky Apdillah (Asahan Univercity, Indonesia)
Ida Bagus Irawan Purnama (Politeknik Negeri Bali, Indonesia)
Oman Somantri (Politeknik Negeri Cilacap, Indonesia)
Syahrul Mauluddin (Universitas Komputer Indonesia, Indonesia)
Addin Aditya (STIKI Malang, Indonesia)
Anjar Wanto (STIKOM Tunas Bangsa)
Anak Agung Gde Satia Utama (Universitas Airlangga)
Silvester Dian Handy Permana (Universitas Trilogi)
Alyaa Abdul-Moneim Mohammed (University of Babylonia, Iraq)
I Ketut bayu Yogha Bintoro (Universitas Trilogi, Indonesia)
Castaka Agus Sugianto (Politeknik TEDC, Bandung, Indonesia)
Nitin Jain (Galgotias University, Greater Noida, India)
Siti Mutrofin (Universitas Pesantren Tinggi Darul Ulum, Indonesia)
Nihad Ibrahim Abdullah (Universitas Tekniknal Malaysiam Melaka)
Mesran (Sekolah Tinggi Manajemen Informatika dan Komputer Budi Darma, Medan, Indonesia)
Tengku Mohd Diansyah (Universitas Harapan, Medan, Indonesia)
Table of Contents
Preface
Thanks to Reviewers
Accreditation Certificate
Table of Contents
Business Development Management Model at Samo-Samo Recycling House Based
on SWOT Analysis
Rusdiansyah Rusdiansyah, Harun Al Rasyid, Suryanto Sosrowidigdo
1-6
Expert System for Diagnosing Pregnancy Complaints by Forward Chaining
Embun Fajar Wati, Anggi Puspitasari
7-16
Implementation of Apriori Algorithm Data Mining for Increase Sales
Reza Alfianzah, Rani Irma Handayani, Murniyati Murniyati
17-25
Drug Stock Prediction at Balige HKBP Hospital Using Adaptive Neuro-Fuzzy
Inference System
Arie Satia Dharma, Lily Andayani Tampubolon, Daniel Somanta Purba
26-34
The Implementation of Artificial Intelligence in Charity Box at Mosque and
Musholla as RFID Based Security System
Defnizal Defnizal, Risa Nadia Ernes
35-42
Facial Recognition Implementation using K–NN and PCA Feature Extraction in
Attendance System
Juliansyah Putra Tanjung, Bayu Angga Wijaya
43-50
Data Mining Model For Designing Diagnostic Applications Inflammatory Liver
Disease
Omar Pahlevi, Amrin Amrin
51-57
Prediction of Netizen Tweets Using Random Forest, Decision Tree, Naïve Bayes, and
Ensemble Algorithm
Yan Rianto, Antonius Yadi Kuntoro
58-71
Expert System for Monitoring Elderly Health Using the Certainty Factor Method
Linda Marlinda, Widiyawati Widiyawati, Reni Widiastuti, Wahyu Indrarti
72-77
An Expert System Diagnosis Human Eye Diseases Using Certainty Factor Method
Web-Based
Bayu Angga Wijaya, Juliansyah Putra Tanjung
78-83
Assosiation Rules for Product Sales Data Analysis Using The Apriori Algorithm
Jarseno Pamungkas, Yopi Handrianto
84-91
Designing an Arduino-based Automatic Cocoa Fermentation Tool
Balkhaya, Dirja Nur Ilham, Rudi Arif Candra, Hasbaini Hasbaini, Fera Anugreni
92-99
Design of an Automatic Water Pump on a Traditional Boat
Ihsan Ihsan, Dirja Nur Ilham, Rudi Arif Candra, Amsar Yunan, Hardisal Hardisal
100-106
Sentiment Analysis Of Full Day School Policy Comment Using Naïve Bayes
Classifier Algorithm
Miftahul Kahfi Al Fath, Arini Arini, Nashrul Hakiem
107-115
The K-Medoids Clustering Method for Learning Applications during the COVID-19
Pandemic
Samudi Samudi, Slamet Widodo, Herlambang Brawijaya
116-121
Credit Loan Selection During the Pandemic Recommendation MCDM-Promethee
Method
Akmaludin, Erene Gernaria Sihombing, Linda Sari Dewi, Rinawati Rinawati, Ester
Arisawati
122-128
The Infusion of Notification Design With an Application of Social Media Based on a
Internet of Things (IOT)
Rudi Arif Candra, Devi Satria Saputra, Dirja Nur Ilham, Herry Setiawan, Hardisal
129-137
Decision Support System For Determining Exemplary Students Using SAW Method
Adjat Sudradjat, Henny Destiana, Aprilah Amira Sefenizka
138-145
A Geographic Information System for Managing and Mapping Irrigation
Infrastructure
Rachmat Wahid Saleh Insani, Syarifah Putri Agustini
146-151
Web-Based Desktop Support Trouble Ticket System Design In PT. Mnc Mediacom
Cable
Besus Maulana Shulton, Eva Zuraidah
152-163
Period Study Accuracy Prediction using Sequential Minimal Optimization Algorithm
Hendri Noviyanto, Bayu Mukti
164-169
Twitter Comment Predictions on Dues Changes BPJS Health In 2020
Riza Fahlapi, Yan Rianto
170-183
Sinkron : Jurnal dan Penelitian Teknik Informatika Volume 5, Number 1, October 2020
DOI : https://doi.org/10.33395/sinkron.v5i1.10649
e-ISSN : 2541-2019 p-ISSN : 2541-044X
*Corresponding Author
This is an Creative Commons License This work is licensed under a Creative
Commons Attribution-NonCommercial 4.0 International License. 116
The K-Medoids Clustering Method for
Learning Applications during the COVID-19
Pandemic
Samudi1)*, Slamet Widodo2), Herlambang Brawijaya3) 1)STMIK Nusa Mandiri,Indonesia, 2,3)Bina Sarana Informatika University, Indonesia
1)[email protected], 2)[email protected], 3)[email protected]
Submitted : Sep 21, 2020 | Accepted : Oct 8, 2020 | Published : Oct 8, 2020
Abstract: The central government took a policy to carry out social distancing with
this social distancing impact on all activities such as impacting the learning process
that usually takes place in class turning into online learning which uses several
supporting applications in the learning process during the Covid 19 pandemic. Such
as using WhatsApp, Moodle, and Zoom for the learning process. Of the three
applications commonly used by students and lecturers for the learning process, they
can be grouped using the K-Medoids method which will get the preferred and
disliked application clusters. Researchers took data from questionnaires made with
Google Form which were distributed to 100 students who often did the learning
process with these three applications. Can produce groupings of applications that are
liked and disliked by students in the learning process. By looking at the level curve
in fig 7. There are two clustering in red and blue. If the curve line to the right is
increasing, many things are not liked in the application used. Conversely, if the line
from the curve towards the right decreases, the less preferred online applications.
The results of this study can be seen from the curve image that learning is the most
preferred in the learning process during the Covid 19 pandemic using the Moodle
and WhatsApp applications, and for the use of zoom is not preferred in the learning
process.
Keywords: Data Mining, K-Medoid Algorithm, Clustering, Learning, Covid 19
INTRODUCTION
All countries are currently experiencing a phenomenon that is very dangerous to health, namely the attack of
COVID 19 or commonly called the Corona Virus, one of which is the State of Indonesia which consists of several
islands that cannot be separated from the attack of COVID 19. COVID 19 is one of the diseases that first appeared
in Wuhan City, Hubei Province, China at the end of December 2019 (Etikasari, Puspitasari, Kurniasari, &
Perdanasari, 2020). Until now, there has not been found a drug or vaccine that can be used to overcome the spread
of the COVID 19 pandemic (Arianto & P, 2020). In Indonesia, the addition of new cases As of August 20, 2020,
an increase of 2,266 cases, while the number of patients recovering from COVID 19 has exceeded the number
100,674 cases or 68.3% (Tim Komunikasi Komite Penanganan Corona Virus Disease 2019 (Covid-19) dan
Pemulihan Ekonomi Nasional, 2020).
For this reason, the central government continues to urge everyone to implement social distancing to minimize
the transmission of COVID 19. Social distancing itself is an action where everyone is required not to be close to
one another (Haqien & Rahman, 2020). With social distancing, all activities hurt all fields (Dwitri, Tampubolon,
Prayoga, Zer, & Hartaman, 2020), one of which is the learning process at the school to college-level must be done
at home. Therefore, the learning process carried out at home requires learning media that is affordable online
(Chesilia, Oktaviany, & Dewi, 2012), the challenge is determining the right platform (Yulianto, Cahyani, &
Silvianita, 2020) to support the learning process during the COVID pandemic. 19.
The learning process that takes place in tertiary institutions that researchers have done Online, is what
researchers often do for the learning process with three applications such as WhatsApp, zoom, and Moodle. With
the three applications online, researchers feel.
Based on the constraints in determining applications to support learning during a pandemic, it can be grouped
using the K-Medoids algorithm, grouping or what is commonly called clustering based on previous researchers is
a technique used for statistical data analysis in various fields, one of which is data mining (Madhulatha TS, 2012)
Sinkron : Jurnal dan Penelitian Teknik Informatika Volume 5, Number 1, October 2020
DOI : https://doi.org/10.33395/sinkron.v5i1.10649
e-ISSN : 2541-2019 p-ISSN : 2541-044X
This is an Creative Commons License This work is licensed under a Creative
Commons Attribution-NonCommercial 4.0 International License. 117
and aims to share data in clusters (Mustakim, 2017). In making clusters, K-Medoids can use because they have a
lot of information on the cluster creation process (Tamura & Miyamoto, 2014).
In this study, the researchers applied the K-Medoids algorithm in grouping the use of applications for learning
during the COVID 19 pandemic from various applications that could support the learning process, so researchers
wanted to find out information about the media used from various platforms such as zoom, WhatsApp, and Moodle
applications which is often used for learning media is it acceptable during the learning process.
LITERATURE REVIEW Medoid is better than K-Means because in K-Medoids we find k as a representative object to minimize the
number of inequalities of data objects, whereas in K-Means we use the total Euclidean distance for data objects
(Marlina, Putri, Fernando, & Ramadhan, 2018). Clustering is a technique used for statistical data analysis, in
various fields, one of which is data mining (Madhulatha TS, 2012) and aims to divide into clusters (Sabzi A,
Farjami Y, 2011) with a similarity level (Irwansyah & Faisal, 2015). Grouping using the K-Medoids method is
widely proposed by several researchers and can be applied such as clustering the spread of COVID 19 in Indonesia
(Sindi et al., 2020), which results in grouping the spread of COVID 19 for the community. And there is another
researcher for the K-Medoids algorithm for disease grouping in Pekanbaru Riau (Juninda, Mustakim, & Andri,
2019) which produces 4 clusters. The steps for the K-Medoids algorithm are as follows (Pramesti, Lahan, Tanzil
Furqon, & Dewi, 2017)
1. Initialize k cluster centers (number of clusters)
2. Allocate each data (object) to the nearest cluster using the Euclidian Distance measurement equation with the
equation:
3. Randomly select objects in each cluster as candidates for the new medoid.
4. Calculate the distance of each object in each cluster with the new medoid candidate.
5. Calculate the total deviation (S) by calculating the value of the new total distance - the old total distance. If S
< 0, then swap objects with cluster data to form a set of k new objects as medoids.
6. Repeat steps 3 to 5 until there is no change in medoids so that you get the cluster and the members of each
cluster.
METHOD
The method used in this study is from identifying problems that exist at this time, then by collecting data, after
the data is collected, the initial data processing is then divided into several categories, then the application of the
K-Medoids algorithm by entering data into the rapid miner application the last is the evaluation of the results of
the data that has been inputted to the rapid miner, then a diagram will appear that can provide information related
to data grouping using K-Medoids. You can also see the sequence based on the previous researcher diagram in Fig
1 below (Juninda et al., 2019).
Start
Identification of problems
Data collection
Initial Data Processing
Application of the K-Medoids Algorithm
Results Evaluation
Finish
Source : (Juninda et al., 2019)
Fig. 1. Research methods
Sinkron : Jurnal dan Penelitian Teknik Informatika Volume 5, Number 1, October 2020
DOI : https://doi.org/10.33395/sinkron.v5i1.10649
e-ISSN : 2541-2019 p-ISSN : 2541-044X
This is an Creative Commons License This work is licensed under a Creative
Commons Attribution-NonCommercial 4.0 International License. 118
RESULT
Identification Problem
Researchers took three applications that would be grouped based on the learning process that was liked and
disliked in the learning process during the COVID 19 pandemic. The three applications that the researchers used
in this study were Moodle, Zoom, and Whatsapp.
Data Collection
In collecting data the researchers used an online questionnaire using a google form which was addressed to
100 students who used to carry out the teaching and learning process online with WhatsApp, Zoom, and Moodle.
Initial Data Processing
The data processing that the researchers did was divided into two categories of likes and dislikes from the three
Zoom, Moodle, and Whatapp applications used in the learning process during the COVID 19 pandemic as follows:
Table 1.
Application Users
Application Like Dislike
Moodle 11 16
Zoom 36 68
WhatsApp 53 16
Source : Research Result (2020)
Application of the K-Medoids Algorithm
The data that has been obtained, the researcher enters the Rapid Miner software by selecting New Process as
follows.
Fig. 2. Rapid Miner
After that, a worksheet consisting of the left middle, and right columns will appear. To enter data from Excel,
select the lower left part where there is a sign that the import data into an existing repository is as follows:
Fig. 3. Entry Data
After entering data from Excel then click next so that it will display the type attribute that will be used in the
grouping process in Rapid Miner. Several things must be changed in determining attributes, including changing
the application attribute to nominal and changing it to id as follows:
Sinkron : Jurnal dan Penelitian Teknik Informatika Volume 5, Number 1, October 2020
DOI : https://doi.org/10.33395/sinkron.v5i1.10649
e-ISSN : 2541-2019 p-ISSN : 2541-044X
This is an Creative Commons License This work is licensed under a Creative
Commons Attribution-NonCommercial 4.0 International License. 119
Fig 4. Attribute Selection
Click next, then save it with the name Learning Data With Applications in the COVID Period. After saving,
the next step is to enter the K-Medoids into the worksheet on the top left in the modeling, clustering, and
segmentation section. Then move the worksheet, then the Excel data set is also moved to the worksheet, then draw
a line so that it forms as below:
Fig. 5. Inserting K-Medoids
After that, the next step is the kiln run which has the blue triangle symbol above, a data view, a metadata view,
a plot view, advanced charts, and annotations will appear. So that the formation of cluster 0 and cluster 1 will
appear, for cluster 0 consists of moodle and WhatsApp, for cluster 1 consists of zoom. The display is as follows:
Fig 6. Cluster formation
To be clearer in grouping likes or dislikes in the learning process using K-Medoids, it can be seen in Figure 7
to show it, you can choose the Cluster Model (Clustering) so that it will be clear that the ascending and descending
curves will be seen, such as cluster 0 which consists of Moodle and WhatsApp with a blue stripe showing that the
more to the right, the smaller is not liked, unlike cluster 1 which consists of zoom, the higher is not liked.
Sinkron : Jurnal dan Penelitian Teknik Informatika Volume 5, Number 1, October 2020
DOI : https://doi.org/10.33395/sinkron.v5i1.10649
e-ISSN : 2541-2019 p-ISSN : 2541-044X
This is an Creative Commons License This work is licensed under a Creative
Commons Attribution-NonCommercial 4.0 International License. 120
Fig 7. Cluster Model (Clustering)
5. Evaluation of Results In the evaluation and the results in this fifth stage, it can be seen the results of the clustering model that is formed in the rapid miner which has been described above that grouping with K-Medoids can form 2 clusters.
DISCUSSION
By clustering online applications such as WhatsApp, zoom, and moodle which are often used in the learning
process for 100 students, it can produce a grouping of applications that students like and dislike in the learning
process. By looking at the level curve in fig 7. There are 2 clusterings in red and blue. If the curve line to the right
is increasing, many things are not liked in the application used. Conversely, if the line from the curve tow
CONCLUSION
Based on the results of the research, the authors can conclude that during the COVID-19 pandemic that spreads
throughout the world, wherein the learning process that is usually delivered directly meeting one another must be
replaced by a social distancing system The lessons that are delivered must be presented online with a variety of
applications used, such as zoom, WhatsApp or moodle. In this study, the results show that the clustering or
grouping that has been carried out on the rapid miner results that learning is the most preferred in the learning
process during the current COVID 19 pandemic using the Moodle and WhatsApp applications, for the use of zoom
is not preferred in the learning process. For future research, we can examine the factors that influence the
application in learning during a pandemic which are the most liked and disliked.
REFERENCES
Arianto, F. S. D., & P, N. (2020). PREDIKSI KASUS COVID-19 DI INDONESIA MENGGUNAKAN METODE
BACKPROPAGATION DAN FUZZY TSUKAMOTO. (Jurnal Teknologi Informasi, 4(1), 120–127.
Retrieved from http://jurnal.una.ac.id/index.php/jurti/article/view/1265%0A
Chesilia, S., Oktaviany, D., & Dewi, D. (2012). Sistem Informasi Manajemen Penjualan dan Persediaan Barang
Berbasis Web pada CV. Matrik Jaya. Jurnal Sistem Informasi STMIK GI MDP, (x), 1–10. Retrieved from
http://eprints.mdp.ac.id/1738/1/Jurnal-2012240058.pdf
Dwitri, N., Tampubolon, J. A., Prayoga, S., Zer, P. P. P. A. N. W. F. I. R. H., & Hartaman, D. (2020).
PENERAPAN ALGORITMA K-MEANS DALAM MENENTUKAN TINGKAT PENYEBARAN
PANDEMI COVID-19 DI INDONESIA. Jurnal Teknologi Informasi, 4(1), 128–132. Retrieved from
http://jurnal.una.ac.id/index.php/jurti/article/download/1266/1104%0A
Etikasari, B., Puspitasari, T. D., Kurniasari, A. A., & Perdanasari, L. (2020). Sistem Informasi Deteksi Dini Covid-
19. Jurnal Teknik Elektro Dan KOmputer, 9(2). Retrieved from
https://ejournal.unsrat.ac.id/index.php/elekdankom/article/view/28278%0A
Haqien, D., & Rahman, A. A. (2020). PEMANFAATAN ZOOM MEETING UNTUK PROSES
PEMBELAJARAN PADA MASA PANDEMI COVID-19. Susuan Artikel Pendidikan, 5(1). Retrieved from
https://journal.lppmunindra.ac.id/index.php/SAP/article/view/6511%0A
Irwansyah, E., & Faisal, M. (2015). Advanced Clustering: Teori dan Aplikasi. Deepublish.
Juninda, T., Mustakim, & Andri, E. (2019). Penerapan Algoritma K-Medoids untuk Pengelompokan Penyakit di
Pekanbaru Riau. Seminar Nasional Teknologi Informasi, Komunikasi Dan Industri (SNTIKI) 11,
(November), 42–49. Retrieved from http://ejournal.uin-
suska.ac.id/index.php/SNTIKI/article/view/8002/4465
Madhulatha TS. (2012). An Overview on Clustering Methods. IOSR Journal of Engineer, 2(4), 719–725.
Marlina, D., Putri, N. F., Fernando, A., & Ramadhan, A. (2018). Implementasi Algoritma K-Medoids dan K-
Means untuk Pengelompokkan Wilayah Sebaran Cacat pada Anak, 4(2), 64–71.
Mustakim. (2017). Effectiveness of K-means clustering to distribute training data and testing data on K-nearest
neighbor classification. Journal of Theoretical and Applied Information Technology, 95(21), 5693–5700.
Pramesti, D. F., Lahan, Tanzil Furqon, M., & Dewi, C. (2017). Implementasi Metode K-Medoids Clustering Untuk
Pengelompokan Data. Jurnal Pengembangan Teknologi Informasi Dan Ilmu Komputer, 1(9), 723–732.
https://doi.org/10.1109/EUMC.2008.4751704
Sabzi A, Farjami Y, Z. M. (2011). An Improved Fuzzy K-Medoids Clustering Algorithm with Optimized Number
of Clusters. IEEE International Conference on Hybrid Intelligent System, 206–210.
Sindi, S., Ratnasari, W., Ningse, O., Sihombing, I. A., Zer, F. I. R. H., & Hartama, D. (2020). Analisis algoritma
k-medoids clustering dalam pengelompokan penyebaran covid-19 di indonesia, 4(1), 166–173.
Tamura, Y., & Miyamoto, S. (2014). Two-Stage Clustering Using One-Pass K-Medoids and Medoid-Based
Agglomerative Hierarchical Algorithms. In IEEE International Conference on Soft Computing and
Intelligent Systems and 15th International Symposium on Advanced Intelligent System (pp. 484–488).
Kitakyushu.
Tim Komunikasi Komite Penanganan Corona Virus Disease 2019 (Covid-19) dan Pemulihan Ekonomi Nasional.
Sinkron : Jurnal dan Penelitian Teknik Informatika Volume 5, Number 1, October 2020
DOI : https://doi.org/10.33395/sinkron.v5i1.10649
e-ISSN : 2541-2019 p-ISSN : 2541-044X
This is an Creative Commons License This work is licensed under a Creative
Commons Attribution-NonCommercial 4.0 International License. 121
(2020). Kesembuhan Covid-19 di Indonesia Tembus 100.000. Retrieved from
https://covid19.go.id/p/berita/kesembuhan-covid-19-di-indonesia-tembus-100000
Yulianto, E., Cahyani, P. D., & Silvianita, S. (2020). Perbandingan Kehadiran Sosial dalam Pembelajaran Daring
Menggunakan Whatsapp group dan Webinar Zoom Berdasarkan Sudut Pandang Pembelajar Pada Masa
Pandemic COVID-19. Jurnal Riset Teknologi Dan Inovasi Pendidikan, 3(2), 331–341. Retrieved from
https://journal-litbang-rekarta.co.id/index.php/jartika/article/view/277%0A