CC BY
54
K-Means Algorithm Implementation for Clustering of Foreign Tourists Visiting
Gita Muditha Kario and Endang Amalia
Abstract—The tourism sector plays an active role in economic growth for a country. Indonesia, which is one of the ASEAN states, shows that the role of tourism is one of the important sectors in the economy in Indonesia. However, the influence of the tourism sector has not been satisfactory for the government. The role of foreign tourist visits affects the economy in Indonesia by increasing foreign exchange for the country. In 2018, foreign exchange from the tourism sector continued to increase by 15.4 percent on an annual basis. However, it is unfortunate that Indonesia is still relatively small compared to other countries in the number of foreign tourist visits. The purpose of this study is to analyze the application of data mining in classifying the number of foreign tourist visits by Indonesia in ASEAN. The grouping is done by applying the K-Means clustering algorithm method. The data are grouped into 3 clusters, namely the high visit cluster (C1), the medium visit cluster (C2), and the low visit cluster (C3). So that the results obtained from the assessment of foreign tourist visits in ASEAN, namely, C1 namely Malaysia, C2 namely Singapore and Indonesia, and C3 namely the Philippines, Thailand, Vietnam, Myanmar / Burma, Brunei Darussalam, Cambodia, and Laos. The results of this study can be seen that Indonesia is in the medium visit grouping (C2). With this data, it can be a reference for the government to improve the tourism sector in visiting foreign tourists in Indonesia.
Keywords—Data mining, clustering, k-means, algorithm, foreign tourists, tourism
I. Introduction
Data mining, which is commonly referred to as knowledge discovery in databases (KDD), is an activity of collecting and using data that finds regularities and pattern relationships in large data sets. Output data mining is for future use in decision making. Clustering is one of the techniques in data mining. Clustering is a multivariate technique whose main objective is to classify objects based on their characteristics. Many methods can be applied for clustering involve the K-Means method, the LVQ (Learning Vector Quantization) method, FCM (Fuzzy C-Means), and so on.
The tourism sector has received significant attention from several countries in the world recently. The tourism sector influences economic growth of the country concerned. One of them is the countries in ASEAN with eastern customs, culture, societal friendliness, and natural beauty which are distinct advantages that can attract tourists to visit these countries. The application of data mining in the tourism sector can be a solution by analyzing large amounts of data. In the results of research launched by Twitter, Indonesia is included in the 10 countries most frequently visited by
tourists using Twitter from the ASIA Pacific [1]. What about ASEAN, whether Indonesia is one of the countries with high foreign tourist visits. The application of data mining is intended to provide real solutions to the government to find out the highest amount of foreign tourist visits in ASEAN.
Indonesia, which is one of the ASEAN states, shows that in economy the role of tourism is one of the important sectors in Indonesia. However, the influence of the tourism sector has not been satisfactory for the government. According to the 2018 Ministry of Tourism's Performance report, the contribution of the tourism sector to the economy is still single digits. In 2018 the portion of tourism to Gross Domestic Product (GDP) was only 5.25 percent. In the same year, the realization of investment in the tourism sector reached US $ 1.6 billion or 80.43 percent of the target set by the government at that time, US $ 2 billion [2]. Nevertheless, the opportunity for the tourism sector to develop in Indonesia is still large, one of which is the influence of foreign tourist visits. The role of foreign tourist visits can affect the economy in Indonesia by increasing foreign exchange for the country. In 2018, foreign exchange from the tourism sector continued to increase, reaching Rp.229.5 trillion or an annual increase of 15.4 percent. However, it is unfortunate that Indonesia is still relatively small compared to other countries in the total of foreign tourist. In 2019, the Central Statistics Agency recorded the total of foreign tourist visits to Indonesia at 16.1 million, an increase of only 1.88 percent compared to 2018 [2]. As a result, the role of foreign tourists is very important so that the government continues to improve the achievement of the tourism sector. The application of data mining is intended to determine the position of Indonesian grouping in foreign tourist visits in ASEAN so that the government can also compare foreign tourist visits from various countries in ASEAN as one of the effective information to create an appropriate strategy to develop the tourism sector in Indonesia.
With the object of foreign tourist visits in ASEAN to find out Indonesia's position, you can use the grouping method with the K-Means algorithm. This study have purpose is to hope that the results of grouping data using the K-Means algorithm can help the government to know that Indonesia is at high or low visit grouping. So that the government can increase the tourism sector in Indonesia to increase the number of foreign tourist visits which have an impact on the introduction of tourist objects in Indonesia, increase foreign exchange and the country's economy. With the above background, the authors wrote this research entitled Implementation of Data Mining on Foreign Tourist Visits Using the K-Means Clustering Algorithm.
II. Theoretical framework
A. Data Mining
Data mining can be defined as a process of finding new patterns with statistical methods, machine learning, database systems, and artificial intelligence. Data mining usually called knowledge discovery, which is taking data patterns to be processed, and then the output is very important information [3]. The purpose of data mining is to produce useful information from large data sources [4]. Data mining is a process of obtaining useful information from large database warehouses, using pattern recognition technology, these patterns are recognized by certain tools that can provide data analysis, as well as mathematical and statistical techniques [5]. Data mining can also be interpreted as extracting new information taken from large chunks of data that helps carry out analysis from reviewing data sets to find unknown relationships and compress data in new ways or methods that can be implemented and data that are useful for making decisions [6].
B. K-Means Clustering Algorithm
K-Means Clustering algorithm, K-means is included in partitioning clustering, where each data must be included in a certain cluster. K-means is a method derived from the simple idea of minimizing the double errors found in grouping problems [7]. K-Means functions to separate data into k separate subdivisions, where k is a positive integer number. So that the data obtained are grouped in the same cluster because they have the same character. The K-means algorithm classifies the data closest to the central cluster (centroid) [8]. The K-means algorithm is well known for its ability and convenience to cluster big data and outliers very quickly. The steps for the K-Means Clustering method are as follows [9], [10]:
1. Determine the value of c for the number of clusters to be formed
2. Choose the center of the initial cluster (centroid) of c.
3. Calculate the closest distance of each data to the centroid using the Euclidean distance formula. The following is the Euclidean Distance formula
= ^(x1-S1)2 + (y1 - t1)2 (Formula 1)
n is the number of data that is incorporated in a data center (centroid).
6. Repeating steps 3 to 5, until the components of each cluster have not changed or the result is the same as the previous one.
C. RapidMiner
RapidMiner is data processing software that is open (open source) using algorithmic principles and data mining, RapidMiner can use on all operating systems cause written by the java language [11]. RapidMiner can be used to integrate data with data mining, analyze text mining, and predictive analytics including operators for data preprocessing, input, output, and data visualization, so that they can make decisions for their users [12].
D. Research Methodology and Techniques Methodology and Techniques Research are the processes of searching for knowledge in a systematic study to find solutions to a problem. Research systematically in the form of formulating problems that occur, then hypotheses, then collecting data or facts, then analyzing it so that it becomes a conclusion in the form of a solution to the problem. The purpose of the research methodology is to find answers and truths to existing problems with scientific application. It not only covers research methods but also considers logic in the context of the research being made and can explain the reasons for using the method or technique.
In conducting a study, a research process is needed which consists of the steps required when carrying out research to be effective. In the research process, the following sequence of steps provides a useful procedural guide [13]:
(x,y): the coordinates of the object (s,t): the coordinates of the object centroid d(x,y): Euclidean distance is the distance between point x and points y using mathematical calculations.
4. Grouping each data based on its closeness to the centroid (smallest distance).
5. Update value of the centroid. The new value of the centroid is obtained from the average total of clusters obtained by using the formula x1+x2+...xn, y1+y2+...yn - (Formula 2)
n
x1 + x2 + ... xn is the x coordinate point of the data incorporated in a centroid.
y1 + y2 + ... yn is the y coordinate point of the data incorporated in a centroid.
Fig. 1. Flowchart Research Process [13]
(1) Define Research Problem
Define research problem is the initial stage that will be carried out in research, namely an overview of the problem that becomes a reference in research to be resolved.
(2) Review The Literature
Research must be supported by a theory that underlies the research. A literature review can be done. Literature sources can be from anywhere such as journals, books, bibliographies, etc., depending on the problem being studied.
(3) Formulate Hypotheses
Formulating a hypothesis is an initial assumption to be developed in research which is the main focus so that research can be more focused.
(4) Design Research
The design research is a structured concept plan that is carried out in research.
(5) Collect Data
Appropriate data is required at the data collection stage because several problems were found to be inadequate in data. For data collection, there are several ways including, primary data collection can be done through experiments or surveys.
(6) Analyse Data
Analyse data can be performed with several operations on existing data such as applying categories, coding, and drawing conclusions.
(7) Interpret And Report
If the researcher does not have a hypothesis, it will explain with findings based on several theories. That is known as interpretation. Finally, after the research has been carried out, the researcher makes a report about the research [13].
small compared to other countries, while foreign tourist visits have an important role in the economy of the country. The government needs to know the grouping of foreign tourist visits to be able to compare foreign tourist visits from various countries in ASEAN so that the government can create an appropriate strategy to increase the tourism sector in Indonesia in increasing the number of foreign tourist visits which have the impact to increase foreign exchange and the country's economy.
• The next stage to be carried out is a literature survey. At this stage, conducting a survey of literature related to research. The literature search is useful to avoid duplication of research implementation. At this stage, the writer uses 2 previous studies to become a comparison material to determine the advantages and disadvantages as well as to know and compare the methods used in previous studies.
III. Research methods
In this study, using the research method C.R. Kothari. The flowchart of the research stages carried out exist in the following figure
Table 1. Literature survey
No
Author
Journal Contents
Fig. 2. Flowchart Research Stages
Based on Figure 2, the following is an explanation of the stages of the research
• The first stage in this research is to formulate a problem of research. The formulation of the problem needs to be done so that the research process is more focused so that it can provide instructions for data collection and search for the right method for solving the problem. The problem that occurs in this study is that foreign tourist visits in Indonesia are still relatively
Agus
Perdana
Windart
Andy Sapta dan Fitri Larasati Sibuea
In his research, he conducted a grouping of rice import data by using the method of K-Means clustering so that it could assess the rice imports of the leading countries [14].
In his research, students grouped data applying the K-means clustering method to determine a pattern assessment from the criteria determined by the school to produce data on outstanding students [15].
The next stage is to develop a hypothesis, that is, in the grouping of data on foreign tourist visits there are significant differences between one grouping to another.
The next stage is the clustering stage applying the K-Means algorithm, then data analysis is carried out with the Rapid Miner application. In the first stage, namely decide the total of cluster from secondary data obtained from the document of foreign tourist visits per month according to nationality in 2020 produced by the Ministry of Law and Human Rights (Directorate General of Immigration) through the website. https://www.bps.go.id. The data will then be processed by grouping foreign tourist visits according to nationality in ASEAN into 3 clusters, namely the high visit cluster, medium visit cluster, and low visit cluster. After determining the cluster then selecting the initial cluster center (centroid) from the initial data. Then calculate Euclidean distance from each data to each centroid with the formula of Euclidean Distance. Next, grouping each data by a centroid. Then update the value of the centroid of the corresponding cluster mean. The value of the new centroid is needed for further data grouping, where the results of grouping members of each cluster do not change or the results are the same as the previous one. If the calculation results are not the same then perform a loop of the steps to calculate the euclidean distance.
1
o
2
However, if the results are the same, the calculation is complete.
• In the final stage, the results of data processing carried out through the testing process will be used in the concluding stage of the results as the conclusion of the study.
IV. Results and discussion
This section, clustering will be formed, the data obtained will be calculated first using the K-Means algorithm. Sample data as much as 10 records from the document of Foreign Tourist Visits per month by Nationality in 2020 in ASEAN. The data is then accumulated. The sum exist in table 2.
Table 2. Data of Foreign Tourist Visits per month (ASEAN) in 2020
Nationality
Data of Foreign Tourist Visits per month (ASEAN) in 2020
Jan Feb Mar Apr May June July Aug Sep Oct Nov Dec
Malaysia 206532 164372 113848 61527 66385 62013 58053 57863 53373 45325 43285 46264
Brunei Darussalam 1219 923 444 8 17 17 4 8 13 26 9 9
Philippines 17174 13487 6041 1427 1819 1535 806 676 714 1003 1275 1342
Singapore 138625 84669 39751 2075 1335 1132 1169 1328 1405 1614 1930 2432
Thailand 7349 7463 2679 325 498 359 315 290 318 404 376 380
Vietnam 9152 6691 1499 265 304 238 181 190 162 175 169 180
Laos 307 396 36 - - - 2 2 - 1 - 2
Kamboja 1136 788 193 2 5 5 13 17 14 15 5 6
Myanmar/Bur ma 3992 1910 1309 783 736 664 502 485 442 502 490 539
Indonesia 55075 46769 26399 1588 2801 2527 2715 2625 3140 2309 1542 1309
Table 3. Accumulated Data Foreign Tourist Visits
No
Nationality
Amount
1 Malaysia 978.840
2 Brunei Darussalam 2.697
3 Philippine 47.299
4 Singapore 277.465
5 Thailand 20.756
6 Vietnam 19.206
7 Laos 746
8 Kamboja 2.199
9 Myanmar/Burma 12.354
10 Indonesia 148.799
After summarizing the data, the value of all foreign tourist visits according to nationality (ASEAN) in 2020 will be obtained. Then the data will be made into clusters, so first it will be determined as follows:
1. 2.
3.
Decide the total of cluster, there are 3 clusters Decide the random cluster center. The cluster center value 1 (C1) = 978,840, the cluster center value 2 (C2) = 47,299, and the cluster center value 3 (C3) = 746.
Decide the cluster value of each data. At this stage, the K-Means algorithm is used to classify the data
into 3 cluster, namely the high visit cluster (C1), the medium visit cluster (C2), and the low visit cluster (C3). In the application of the K-means algorithm, the midpoint or centroid is obtained. Euclidean Distance Space is used to count the distance between data and centoid. one of the equations that can be used is Euclidean Distance Space. The clustering process of each processed data is taken from the closest distance. For example, to calculate the distance of the first instance from the center of the first cluster is:
d(X,y) = V(xi - 5i)2 + (yi - fi)2
d1,1 = 7(978.840 - 978.840)2 = 0 di,2 = 7(47.299 di
978.840)2 = 931.541
*1j3 = V(746 - 978.840)2 = 978.094 The calculation continues on each data. From the results of the above calculations, the value of cluster 1, cluster 2, and cluster 3 in iteration 1 has been grouped. The calculation results of alliteration 1 calculations exist in Table 4.
Table 4. Calculations Iteration 1
Initials
Nationality
Amount
Cluster
C1
C2
C3
Distance Result
1 Malaysia 978.840 0 931.541 978.094 0
2 Singapore 277.465 701.375 230.166 276.719 230.166
3 Indonesia 148.799 830.041 101.500 148.053 101.500
4 Philippines 47.299 931.541 0 46.553 0
5 Thailand 20.756 958.084 26.543 20.010 20.010
6 Vietnam 19.206 959.634 28.093 18.460 18.460
7 Myanmar/Burma 12.354 966.486 34.945 11.608 11.608
8 Brunei Darussalam 2.697 976.143 44.602 1.951 1.951
9 Kamboja 2.199 976.641 45.100 1.453 1.453
C1 C2 C2 C2 C3 C3 C3 C3 C3
10
Laos
746
978.094
46.553
C3
In table 3, in this process, K-means will continue iterating so that the last iteration data is the same as the previous iteration data grouping. The graph of iteration 1 grouping exist in the following figure:
lcluii«r_0 •chifUr.i • ciusitr_3
ClUSler • CluüUCjO • clust«r_l * clu:ter_i 240,000 230,000 220,000 210,000 300,000 190,000 100,000 170,000
ieo,ooo
150,000 140,000 130,000 g 120,000 В 110,000 a 100,000 90,000 50,000 70,000 «0,000 50,000 40,000 30,000 20,000 10,000 о
-10,000
IciiimOf—
luittr_l]
Fig. 3. Clustering Iteration Data 1
Fig. 4. Graph Iteration Data 1
In iteration 1, the input value data cluster is obtained which exist in table 4. In iteration 2, the centroid value search will be carried out for iteration 2 with the results of grouping in iteration 1. The results exist in Table 5:
Table 5. Centroid Iteration Data 2
Attribute
Cluster 1
(C1)
Cluster 2 (C2)
Cluster 3 (C3)
Amount of visit
978.840
157.854
9.660
0
0
After the centroid has been obtained, then the next process is to find the closest distance to each data. Process the data that grouped, the search for the closet or shortest distance in iteration 2, and the calculation of the results of the clustering data exist in the following table:
Table 6. Calculations Iteration 2
Cluster
Initials Nationality Amount C1 C2 C3 Distance Result
1 Malaysia 978.840 0 820.986 969.180 0 C1
2 Singapore 277.465 701.375 119.611 267.805 119.611 C2
3 Indonesia 148.799 830.041 9.055 139.139 9.055 C2
4 Philippines 47.299 931.541 110.555 37.639 37.639 C3
5 Thailand 20.756 958.084 137.098 11.096 11.096 C3
6 Vietnam 19.206 959.634 138.648 9.546 9.546 C3
7 Myanmar/Burma 12.354 966.486 145.500 2.694 2.694 C3
8 Brunei Darussalam 2.697 976.143 155.157 6.963 6.963 C3
9 Kamboja 2.199 976.641 155.655 7.461 7.461 C3
10 Laos 746 978.094 157.108 8.914 8.914 C3
Table 6 shows the results of the grouping iteration 2 has different results from iteration 1. The process will continue in the next iteration. The resulting graph of iteration 2 exist in the following figure:
cluster * tiust«r_0 « tt*Jiiw_l * clusier_2
125,000 -
120,000 « 115,000 119,01» 105.000 100,0 M 95,000 90.0 M •5.0 00 60.000 75,000 70,000 J ¿5,000 I 60,000 I 5S,000 50,000 45,000 +0,0 00 35,000 30,0 W ■ 25,000 20,000 15,000
10,000 £ 5,000
0 *
-5,000 _._____
Fig. 5. Clustering Iteration Data 2
ft tlusiff_0 • dusier_l ■ CluS!ïr_2
—|c1lM»«r_l]
From Figure 5, the data on foreign tourist visits by nationality (ASEAN), iteration groupings 2 obtained 3 cluster. The high visit cluster (C1) is Malaysia, the medium visit cluster (C2) is Singapore and Indonesia and the low visit cluster (C3) is the Philippines, Thailand, Vietnam, Myanmar / Burma, Brunei Darussalam, Cambodia, and Laos. The results of iteration 2 show dissimilar results from the results of iteration 1. Then the process will continue to iteration 3 in the same way, determining the new centroid from iteration 2 and finding the closest distance from the value of the centroid in iteration 3.
Table 7. Centroid Iteration Data 3
Attribute Cluster 1 Cluster 2 Cluster
(C1) (C2) 3(C3)
Amount of visit 978.840 213.132 15.037
After the centroid has been obtained, then the next process is to find the closest distance to each data. Process the data that grouped, the search for the closet or shortest distance in iteration 2, and the calculation of the results of the clustering data exist in the following table:
Fig. 6. Graph Iteration Data 2
Table 8 Calculations Iteration 3
Initials Nationality Amount Cluster
C1 C2 C3 Distance Result
1 Malaysia 978.840 0 765.708 963.803 0 C1
2 Singapore 277.465 701.375 64.333 262.428 64.333 C2
3 Indonesia 148.799 830.041 64.333 133.762 64.333 C2
4 Philippines 47.299 931.541 165.833 32.262 32.262 C3
5 Thailand 20.756 958.084 192.376 5.719 5.719 C3
6 Vietnam 19.206 959.634 193.926 4.169 4.169 C3
7 Myanmar/Burma 12.354 966.486 200.778 2.683 2.683 C3
8 Brunei Darussalam 2.697 976.143 210.435 12.340 12.340 C3
9 Kamboja 2.199 976.641 210.933 12.838 12.838 C3
10 Laos 746 978.094 212.386 14.291 14.291 C3
S&.flpp ■ t.ùiinù
syion-
50.0PÖ ■ +&JPD0
| ■ << : g 3irflPP-iiijnD ■ ZO^H 15.ÜOD ■ lGUUÜ i.ÛÙD
i s * t 6 ? t î ii
Nmomfey
Fig. 7. Clustering Iteration Data 3
Fig. 8. Graph Iteration Data 3
In the 3rd iteration, the data grouping was carried out in 3 clusters with iteration 2, the same results were obtained. From 10 data on foreign tourist visits by nationality (ASEAN), 1 cluster of high visits (C1), namely Malaysia, 2 medium visit clusters (C2) namely Singapore and Indonesia, and 7 clusters of low visits (C3) the Philippines, Thailand, Vietnam, Myanmar / Burma, Brunei Darussalam, Cambodia, and Laos.
V. Conclusion
To assess the grouping of the number of Foreign Tourist Visits in ASEAN, it can be applying the K-Means clustering algorithm. The data will be processed to obtain the value of foreign tourist visits in Indonesia. The data is processed using RapidMiner by conducting K-Means which are grouped into 3 clusters, namely the high visit cluster (C1), the medium visit cluster (C2), and the low visit cluster (C3). Data centroid C1 = 978,840, data centroid C2 = 213.132, data centroid C3 = 15.037 So that results obtained from an assessment of foreign tourist visits in ASEAN, namely, 1 high visit cluster (C1), namely Malaysia, 2 medium visit clusters (C2) namely Singapore and Indonesia, and 7 low visit clusters (C3) namely Philippines, Thailand, Vietnam, Myanmar / Burma, Brunei Darussalam, Cambodia, and Laos. The results of this study can be seen that Indonesia is in the medium visit grouping (C2). With this data, it can be a source of information for the government as a reference
for improving the tourism sector so that it can maintain and even increase the number of foreign tourist visits. The government can also compare with other countries so that it can be equal to countries in high clusters, even more than other countries in foreign tourist visits which have an impact on the introduction of tourist objects in Indonesia and increase in foreign exchange. In the future for similar research using the K-Means clustering method, it can be compared with other clustering methods to get maximum modeling results. In processing the data, the clustering can give weight to the criteria in the grouping process to get more accurate results. Future research can focus on Indonesia so that the government can find out which areas need to be improved in foreign tourist visits.
Acknowledgment
The authors like to thank all those who have played an important role in the success of this research. This research is still not perfect, but it is hoped that it will be useful for researchers and for readers.
REFERENCES
[1] A. M. M. P. Senja, "Alasan Utama Turis Asing Berwisata ke Indonesia," 26 March 2019. [Online]. Available: https://travel.kompas.com/read/2019/03/26/171100327/alasan-utama-turis-asing-berwisata-ke-indonesia.. [Accessed March 2021].
[2] I. CNN, "Menghitung Kontribusi Sektor Pariwisata Bagi Ekonomi RI," 26 February 2020. [Online]. Available: https://www.cnnindonesia.com/ekonomi/20200226121314-532-478265/menghitung-kontribusi-sektor-pariwisata-bagi-ekonomi-ri. [Accessed March 2021].
[3] Suyanto, Data Mining Untuk Klasifikasi dan Klasterisasi Data, Bandung: Penerbit Informatika, 2017, p. 10.
[4] R. Purohit and D. Bhargava, "An Illustration to Secured Way of Data Mining Using Privacy Preserving Data Mining," Journal of Statistics and Management Systems, p. 637, 2017. doi:https://doi.org/10.1080/09720510.2017.1395183.
[5] E. Sikumbang, "Implementasi Data Mining Untuk Memprediksi Masa Studi Mahasiswa Menggunakan Algoritma C4.5 (Studi Kasus: Universitas Dehasen Bengkulu)," Jurnal Teknik Komputer, p. 156, 2018.
[6] E. Prasetyo, Data Mining: Mengolah Data Menjadi Informasi Menggunakan Matlab, Bandung: Andi Offset, 2014, p. 15.
[7] S. Nagari and L. Inayati, "Implementation of Clustering Using K-Means Method to Determine Nutritional Status," Jurnal Biometrika danKependudukan, p. 63, 2020. doi:10.20473/jbk.v9i1.2020.
[8] Asroni and R. Adrian, "Penerapan Metode K-Means Untuk Clustering Mahasiswa Berdasarkan Nilai Akademik Dengan Weka Interface Studi Kasus Pada Jurusan Teknik Informatika UMM Magelang," JurnalIlmiah Semesta Teknika, p. 78, 2015.
[9] W. Azis and D. Atmajaya, "WS Azis and D Atmajaya. Pengelompokan Minat Baca Mahasiswa Menggunakan Metode K-Means," ILKOM Jurnal Ilmiah, pp. 89-90, 2016.
[10] S. Handoko, Fauziah and E. Handayani, "Implementasi Data Mining Untuk Menentukan Tingkat Penjualan Paket Data Telkomsel menggunakan Metode K-Means Clustering," Jurnal Ilmiah Teknologi dan Rekayasa, pp. 80-81, 2020. doi:https://doi.org/10.35760/tr.2020.v25i1.2677.
[11] S. Haryati, A. Sudarsono and E. Suryana, "Implementasi Data Mining Untuk Memprediksi Masa Studi Mahasiswa Menggunakan Algoritma C4.5 (Studi Kasus: Universitas Dehasen Bengkulu),"
Jurnal Media Infotama, p. 133, 2015.
[12] B. Rahmat, A. Gafar, N. Fajriani, U. Ramdani, F. Uyun, P. Yuwanda and N. Ransi, "Implementasi K-Means Clustering Pada RapidMiner
Untuk Analisis Daerah Rawan Kecelakaan," Seminar Nasional Riset Kuantitatif Terapan, p. 60, 2017.
[13] C. Kothari, Garg and Gaurav, Research Methodology: Methods and Techniques [Fourth multi color edition], New Age International, 2019, pp. 1-11.
[14] A. Windarto, "Implementation of Data Mining on Rice Imports by Major Country of Origin Using Algorithm Using K-Means Clustering Method," International Journal Of Artificial Intelligence Research., 2017. doi:10.29099/ijair.v1i2.17.
[15] F. Sibuea and A. Sapta, "Pemetaan Siswa Berprestasi Menggunakan Metode K-Means Clustering," JURTEKSI (Jurnal Teknologi dan Sistem Informasi), p. 88, 2017.
Biography Of Authors
Gita Muditha Kario Information System, Faculty of Technique, Widyatama University, Cikutra Street No.204A Sukapada, Bandung, 40125, Indonesia
gita.muditha@widyatama.ac.id
Endang Amalia Information System, Faculty of Technique, Widyatama University, Cikutra Street No.204A Sukapada, Bandung, 40125, Indonesia endang.amalia@widyatama.ac.id
