p–ISSN: 2723 - 6609 e-ISSN: 2745-5254
Vol. 5, No. 9 September 2024 http://jist.publikasiindonesia.id/
Jurnal Indonesia Sosial Teknologi, Vol. 5, No. 9, September 2024 3222
Implementation of the Smart Indonesia Card Scholarship
(KIP) Acceptance Using the K-NN Method (Case Study:
Politeknik Siber Cerdika Internasional)
Abi Surya Wijaya
1*
, Lena Magdalena
2
, Chairun Nas
3
Universitas Catur Insan Cendekia Cirebon, Indonesia
Email:
1*
, [email protected].id
2
3
*Correspondence
ABSTRACT
Keywords: IT scholarship
acceptance; smart
indonesia card (KIP); K-
nearest neighbor (K-NN).
This study discusses the implementation of the K-Nearest
Neighbor (K-NN) method in the process of receiving the
Smart Indonesia Card (KIP) scholarship at the Cerdika
International Cyber Polytechnic. The main purpose of this
study is to improve accuracy and efficiency in the selection
of KIP scholarship recipients. The K-NN method was
chosen because of its ability to classify data based on the
proximity of features between samples. This research
involves analyzing data on prospective scholarship
recipients which includes variables such as academic
achievement, economic conditions, and extracurricular
activities. The results of the implementation of the K-NN
method show that this method can be used as an effective
tool in the selection process of KIP scholarship recipients,
with a fairly high level of accuracy compared to traditional
methods. This finding is expected to help the Polytechnic in
increasing transparency and fairness in providing
scholarships.
Introduction
Higher education is one of the important aspects in the development of a country,
playing a vital role in improving the quality of human resources. However, access to
higher education is still a challenge for some people, especially those from low economic
backgrounds. To overcome this inequality, various scholarship programs have been
introduced, one of which is the Smart Indonesia Program (PIP) better known as the KIP
Scholarship (Smart Indonesia Card) (Maulida & Sari, 2015).
The Indonesia Smart Program through the Indonesia Smart Card (KIP) is the
provision of education cash assistance to all school-age children (6-21 years old) and one
of the national programs (listed in the 2015-2019 RPJMN) in government regulations
since the end of 2014. The Smart Indonesia Program through KIP is part of the
improvement of the Poor Student Assistance Program (BSM).
SCI Polytechnic under the Mansyur Al-Makki Foundation is the first Polytechnic
in Cirebon that has Digital Business, Network Computer Engineering Engineering, and
Implementation of the Smart Indonesia Card Scholarship (KIP) Acceptance Using the K-NN
Method (Case Study: Politeknik Siber Cerdika Internasional)
Jurnal Indonesia Sosial Teknologi, Vol. 5, No. 8, August 2024 3223
Community Economic Rural Empowerment study programs. SCI Polytechnic carries the
tagline Skillfull College which ensures that each graduate has the best skills in their study
program. All Study Programs have been accredited by LAMEMBA, LAM TEKNIK, and
BAN-PT in 2023.
Law Number 12 of 2012 concerning Higher Education has given a mandate to the
government to realize affordability and equitable distribution in obtaining access to
quality higher education that is relevant to the interests of the community for progress,
independence, and welfare. The government is obliged to increase access and learning
opportunities and prepare intelligent and competitive Indonesian people. (Law (UU) No.
12 of 2012 concerning Higher Education, 2012). One of the government's efforts to
increase access to learning for the community is through the provision of scholarships.
The Smart Indonesia Card Program (KIP) is an initiative of the Indonesian
government that aims to ensure that all Indonesian children have access to a proper
education. (Amadi et al., 2023). Through this program, students from underprivileged
families are assisted in the form of scholarships that cover tuition fees and other needs.
The implementation of this program is expected to help reduce the dropout rate and
improve the quality of human resources in Indonesia. (Zainal, Joesyiana, Zainal,
Wahyuni, & Adriyani, 2023).
However, as the number of KIP scholarship recipients increases every year, an
accurate and fair recipient selection process is a challenge in itself. (NEGARA, n.d.). In
practice, several obstacles are often faced, such as invalid recipient data, a selection
process that takes a long time, and the potential for human error in determining
scholarship recipients. Therefore, a system is needed that can help the selection process
of scholarship recipients efficiently and on target.
Cerdika International Cyber Polytechnic as one of the educational institutions
participating in the KIP program, also experienced challenges in the selection process of
scholarship recipients. This institution needs a system that can process scholarship
applicant data quickly and accurately so that it can select prospective scholarship
recipients who are truly entitled more efficiently. (Maryaningsih, Siswanto, & Mesterjon,
2013).
In this context, the K-Nearest Neighbor (K-NN) method can be applied as a solution
to solve the problem. K-NN is one of the methods in machine learning used for
classification and regression (Bugis, Cakra, Patombongi, & Suarna, 2024). This method
works by comparing new data with existing data and determining the class of the new
data based on proximity (similarity) with several nearby data.
By applying the K-NN method in the selection of KIP scholarship recipients at the
Cerdika International Cyber Polytechnic, it is hoped that the selection process can be
faster, more efficient, and more accurate. This system can help reduce errors in
determining scholarship recipients and ensure that scholarships are awarded to those who
are truly in need and meet the criteria.
This study will examine how the implementation of the K-NN method can be
applied in the selection process of KIP scholarship recipients at the Cerdika International
Abi Surya Wijaya, Lena Magdalena, Chairun Nas
Jurnal Indonesia Sosial Teknologi, Vol. 5, No. 9, September 2024 3224
Cyber Polytechnic, as well as analyze the effectiveness and efficiency of this method in
solving existing problems. It is hoped that the results of this study can make a positive
contribution to efforts to improve the quality and fairness of KIP scholarship distribution
in Indonesia (Hisyam, Khotimah, Dewi, & Virdi, 2024).
The Indonesia Smart Card Scholarship (KIP) Lecture is one of the scholarship
pathways offered by the government to increase access to higher education for people
who are outstanding and economically disadvantaged. (K. Religion, 2020) The KIP
Lecture Scholarship used to be called the Bidikmisi Scholarship which was later renamed
in 2020. From 2015 to 2019, the Directorate General of Islamic Education through the
Directorate of Islamic Religious Higher Education has provided Bidikmisi scholarships
to 37,850 students. After transforming into KIP Lecture, the quota has increased quite
significantly. If in 2019 the Bidikmisi quota was only 11,000 students, then in 2020 it will
be 17,565 people. (Romadhon, 2023). With the increase in the quota of KIP Lecture
recipients, it is necessary to take accurate steps in determining the eligibility of KIP
Lecture recipients.
The K-Nearest Neighbors (K-NN) method is one of the machine learning
algorithms that can be applied to predict or classify data based on its relationship with
existing data. (Subhan, 2021). In the context of receiving KIP scholarships, the use of the
K-NN method can help in identifying the candidates who are most eligible to receive
assistance, by comparing their profiles with data from previous scholarship recipients.
One way that can be done in the selection process for KIP Lecture Scholarship
admissions is to classify prospective scholarship recipients because the right classification
results are very important to determine the eligibility of scholarship recipients. Several
methods for classifying the eligibility of scholarship recipients have been proposed by
many researchers. Such as the research of M. Khalil who carried out the Application of
the K-Nearest Neighbor (kNN) Method in the Scholarship Recipient Selection Process.
(Kholil, 2018) A. Sumiah and N. Mirantika also compared the K-Nearest Neighbour
method to recommend determining scholarship recipients (Sumiah & Mirantika, 2020).
The objectives to be achieved through this research are as follows: 1. Implementing
more structured and organized data management to support more effective analysis and
selection. 2. Increase transparency in the selection process by providing clear information
on how decisions are made using the K-NN method. 3. Determine and use key variables
that affect the eligibility of scholarship acceptance to improve the accuracy of selection.
According to (Rachma, 2022) This study shows that there are 23 regencies/cities
that are included in the classification category of poverty level less than average and the
remaining 15 regencies/cities are included in the classification category of poverty level
more than average. The higher the per capita expenditure index, the rate of GDP, and the
average length of school-issued in an area, it shows the improvement in community
welfare and the quality of human resources in the Regency/City area. Meanwhile, the
higher the open unemployment rate in an area, it shows the decrease in the level of
community welfare in the Regency/City area. The accuracy results produced from the
Implementation of the Smart Indonesia Card Scholarship (KIP) Acceptance Using the K-NN
Method (Case Study: Politeknik Siber Cerdika Internasional)
Jurnal Indonesia Sosial Teknologi, Vol. 5, No. 8, August 2024 3225
classification using the K-Nearest Neighbor algorithm showed the highest accuracy of
76.67% with the best k parameter values of k = 1 and k = 2.
Method
The model used in this study is the standard method by applying the K-Nearest
Neighbor (K-NN) method. At this stage, the researcher will classify the indicators of
receiving KIP scholarships for students, with this model it can make it easier for
researchers to apply the model of the algorithm that aims to produce final scores in the
form of accuracy, precision, and recall.
This research is included in case study research, which is research that is carried
out intensively, in detail, and in depth on an organization, institution, and certain
symptoms. A case study is a more sophisticated strategy when the subject matter of a
study is related to "how" or "why", or if the researcher has little chance of being
investigated, when to control the events to be investigated, and when the focus of the
research is on contemporary (present) phenomena in the context of real life.
Results and Discussion
K-Nearest Neighbor Test Results
SW
Table 1 Data Set
No
Student Name
Status
Euclidean
Distance
Ranking
1
1
Muhamad
Zaenal Asikin
Fail
2.000
1
2
2
Sri Hartini
Fail
22.361
2
3
3
Iin Tarsini
Fail
22.361
3
1
4
Adam
Hernawan
Fail
22.361
4
1
5
Aurelia
Widya Astuti
Fail
22.361
5
1
6
Azka
Muharam
Fail
22.361
6
2
7
Tantra
Agun Wiguna
Fail
22.361
7
8
8
Moh
Hisyam Hauzaan
Fail
24.495
8
1
9
Arulaflah
Nurwahid
Fail
24.495
9
1
10
Egi Ahmad
Baihaqi
Accepted
31.623
10
4
11
Adila
Septiyani
Accepted
31.623
11
1
12
Fika Sabila
Accepted
31.623
12
Abi Surya Wijaya, Lena Magdalena, Chairun Nas
Jurnal Indonesia Sosial Teknologi, Vol. 5, No. 9, September 2024 3226
1
13
Rayhan
Syawal Fizriki
Accepted
31.623
13
1
14
Alif
Suryalaksana
Accepted
31.623
14
1
15
Elvira
Fitriyanti
Accepted
31.623
15
1
16
Andika
Bagus Saputra
Accepted
31.623
16
1
17
Ahmad
Syibahi
Accepted
31.623
17
1
18
Muhamma
d Khoerudin
Accepted
33.166
18
1
19
Mamduh
Rihadatul Aisy
Accepted
33.166
19
2
20
Satrio Rafif
Firmansah
Accepted
33.166
20
2
21
Siti Ainul
Kholoifah
Fail
33.166
21
Figure 2 Rapidminer Operator
Then the read excel menu is double-clicked or clicked and dragged to the main
process page as shown in the following figure 5.3:
Figure 3 Import Configuration Wizard.
Implementation of the Smart Indonesia Card Scholarship (KIP) Acceptance Using the K-NN
Method (Case Study: Politeknik Siber Cerdika Internasional)
Jurnal Indonesia Sosial Teknologi, Vol. 5, No. 8, August 2024 3227
Then the Set Role menu is double-clicked or clicked and dragged to the main
process page as shown in the following Figure 4:
Figure 4 Set Role in the Main Process
Then the K-Nearest Neighbor (K-NN) menu is double-clicked or clicked and
dragged to the main process page as shown in the following Figure 5:
Figure 5 K-NN in the Main Process
The results obtained from testing the graph shape and description rule of K-Nearest
Neighbor (K-NN) are as shown in the image below:
Abi Surya Wijaya, Lena Magdalena, Chairun Nas
Jurnal Indonesia Sosial Teknologi, Vol. 5, No. 9, September 2024 3228
Figure 6 Design of K-NN Testing with Rapidminer
Figure 7 K-NN Test Results
Figure 8 Accuracy PerformanceVector (Performance) Results
Implementation of the Smart Indonesia Card Scholarship (KIP) Acceptance Using the K-NN
Method (Case Study: Politeknik Siber Cerdika Internasional)
Jurnal Indonesia Sosial Teknologi, Vol. 5, No. 8, August 2024 3229
Based on the K-NN test data in the figure above, it is stated that 81.82% of the
accuracy level of true predictions is accepted for the use of K-NN in the process of
receiving the KIP Scholarship at the International Cyber Polytechnic.
Classification
Based on the majority classification of the number of closest K values (K-3, K-5,
K-7, K-9, and K-19) that are different, the following results are produced:
Figure 9 K-3 K-NN Test Design on Rapidminer
Figure 10 Results of K-NN Testing with K=3 Condition
Abi Surya Wijaya, Lena Magdalena, Chairun Nas
Jurnal Indonesia Sosial Teknologi, Vol. 5, No. 9, September 2024 3230
Figure 11 Results of K-NN Testing with K=5 Condition
Figure 11 Results of K-NN Testing with K=7 Condition
Figure 12 Results of K-NN Testing with K=9 Condition
Implementation of the Smart Indonesia Card Scholarship (KIP) Acceptance Using the K-NN
Method (Case Study: Politeknik Siber Cerdika Internasional)
Jurnal Indonesia Sosial Teknologi, Vol. 5, No. 8, August 2024 3231
Figure 13 Results of K-NN Testing with K=19 Condition
Results of Classification of Prospective Scholarship Recipients with K=3
The results of the K-NN test using 3 data produced 3 people who failed and the
accuracy values were as follows:
K= 3 with 3 data, with the result
Percentage of failures = 3/3 = 100%
Percentage Accepted = 0/3 = 0%
This means that the accuracy level of failing the scholarship classification test using K=3
is 100%.
Results of Classification of Prospective Scholarship Recipients with K=5
The results of the K-NN test using 5 data resulted in 5 people failing, here is the
calculation of the accuracy level:
K= 5 with 5 data with results Percentage failed = 5/5 = 100% Percentage Accepted = 0/5
= 0%. This means that the accuracy level of the scholarship classification test using K=5
results in 100% failure and the accuracy is accepted as much as 0%.
Results of Classification of Prospective Scholarship Recipients with K=7
The results of the K-NN test using 7 data resulted in 7 people failing, here is the
calculation of the accuracy level:
K= 7 with 7 data with results Percentage failed = 7/7 = 100% Percentage Accepted = 0/7
= 0%. This means that the accuracy level of the scholarship classification test using K=7
results in 100% failure and the accuracy is accepted only 0%.
Classification Results of Prospective Scholarship Recipients with K=9
The results of the K-NN test using 9 data resulted in 9 people failing, here is the
calculation of the accuracy level:
K= 9 with 9 data with results Percentage failed = 9/9 = 100% Percentage Accepted = 0/9
= 0%
This means that the accuracy level of the scholarship classification test using K=9
results in 100% failure and the accuracy of the scholarship received 0%.
Results of Classification of Prospective Scholarship Recipients with K=19
The results of the K-NN test using 19 data resulted in 9 people failing and 10 being
accepted, here is the calculation of the accuracy level:
Abi Surya Wijaya, Lena Magdalena, Chairun Nas
Jurnal Indonesia Sosial Teknologi, Vol. 5, No. 9, September 2024 3232
K= 9 dengan 9 data dengan hasil Presentase gagal = 9/19 = 47% Presentase Diterima =
10/19 = 53%. Artinya, tingkat accuracy pengujian klasifikasi Beasiswa menggunakan
K=9 menghasilkan 47% gagal dan accuracy Beasiswa diterima 53%.
Conclusion
From the research on the implementation of the Smart Indonesia Card (KIP)
scholarship at the Cerdika International Cyber Polytechnic using the K-Nearest Neighbor
(K-NN) method, several important points can be concluded as follows: 1. Effectiveness
of the K-NN Method: The K-NN method has proven to be effective in classifying
prospective KIP scholarship recipients based on various criteria such as economic
background, academic achievement, and other relevant criteria. With high accuracy, this
method aids in objective and data-driven decision-making. 2. Accuracy and Accuracy:
The implementation of K-NN in this case study shows a significant level of accuracy in
predicting scholarship recipients. This shows that the model built can be relied on to select
scholarship recipients fairly and on target. 3. System Sustainability: The K-NN-based
selection system can be applied sustainably and integrated with the existing information
system at the Cerdika International Cyber Polytechnic. Thus, the selection process can be
carried out efficiently and consistently in the future. 4. Implementation
Recommendations: With this system, the institution can more easily identify potential
worthy recipients, so that the KIP scholarship program can be more targeted and have a
positive impact on students in need.
Bibliography
Amadi, Aunur Shabur Maajid, Hasan, Salsabila, Rifanto, Nabila Akmaliya, Wildan,
Muhammad, Afifah, Nidia Qonitatul, & Nisak, Nur Maslikhatun. (2023). Upaya
Pemerintah dalam Menjamin Hak Pendidikan untuk Seluruh Masyarakat di
Indonesia: Sebuah Fakta yang Signifikan. Education, 18(1), 161–171.
Bugis, Sukriadi Amanah, Cakra, Cakra, Patombongi, Andi, & Suarna, Dedi. (2024).
Implementasi Algoritma K-Nearest Neighbor (K-Nn) Dalam Perancangan Alat
Implementation of the Smart Indonesia Card Scholarship (KIP) Acceptance Using the K-NN
Method (Case Study: Politeknik Siber Cerdika Internasional)
Jurnal Indonesia Sosial Teknologi, Vol. 5, No. 8, August 2024 3233
Pendeteksi Tingkat Kesegaran Daging. Simtek: Jurnal Sistem Informasi Dan
Teknik Komputer, 9(1), 55–61.
Hisyam, Ciek Julyati, Khotimah, Husnul, Dewi, Kartika, & Virdi, Santika. (2024).
Analisis Fenomena Hedonisme di Kalangan Mahasiswa Penerima Beasiswa KIP
Kuliah: Perspektif Sosio-Ekonomi Baru. Populer: Jurnal Penelitian Mahasiswa,
3(2), 16–30.
Khalil, Moenawar. (2018). Penerapan Metode K Nearest Neighbord Dalam Proses
Seleksi Penerima Beasiswa. Proceeding Seminar Nasional Sistem Informasi Dan
Teknologi Informasi, 1(1), 13–18.
Maryaningsih, Maryaningsih, Siswanto, Siswanto, & Mesterjon, Mesterjon. (2013).
Metode Logika Fuzzy Tsukamoto Dalam Sistem Pengambilan Keputusan
Penerimaan Beasiswa. Jurnal Media Infotama, 9(1).
Maulida, Yusni, & Sari, Lapeti. (2015). Analisis kualitas sumber daya manusia dan
pengaruhnya terhadap pertumbuhan ekonomi di kabupaten Pelalawan. Riau
University.
Negara, Prodi Ilmu Administrasi. (n.d.). Evaluasi Program Beasiswa Kartu Indonesia
Pintar-Kuliah (KIP-K) Di Universitas Islam Negeri Ar-Raniry Banda Aceh.
Rachma, Clariza Adelina. (2022). Implementasi Algoritma K-Neraest Neighbor dalam
penentuan Klasifikasi Tingkat Kedalaman Kemiskinan Provinsi Jawa Timur.
Universitas Islam Negeri Maulana Malik Ibrahim.
Romadhon, Manzilur Rahman. (2023). Perbandingan Algoritma K-Nearest Neighbor (K-
NN) dan Naive Bayes pada data penerima beasiswa Kartu Indonesia Pintar (KIP)
Kuliah. Universitas Islam Negeri Maulana Malik Ibrahim.
Subhan, Subhan. (2021). Klasifikasi Konten Web Radikal Di Indonesia menggunakan
Web Content Mining Dan Algoritma K-Nearest Neighbor. Jurnal Informasi, Sains
Dan Teknologi, 4(2), 70–77.
Sumiah, Aah, & Mirantika, Nita. (2020). Perbandingan Metode K-Nearest Neighbor dan
Naive Bayes untuk Rekomendasi Penentuan Mahasiswa Penerima Beasiswa pada
Universitas Kuningan. Buffer Informatika, 6(1), 1–14.
Zainal, Rahmi, Joesyiana, Kiki, Zainal, Haznil, Wahyuni, Sri, & Adriyani, Annesa.
(2023). Manajemen Pengelolaan Keuangan bagi Mahasiswa Penerima Beasiswa
KIP Kuliah pada Perguruan Tinggi di Lingkungan Yayasan Pendidikan Persada
Bunda (STIE–STISIP–STBA–STIH). JIPM: Jurnal Inovasi Pengabdian
Masyarakat, 1(1), 1–5.
