Hendra Supendar
Research Subjects
Hendra Supendar has a total of 11 co-authors. Here are the top co-authors by number of shared publications:
- Nining Suharyanti
Universitas Bina Sarana Informatika,Indonesia
4 shared publications - Rusdiansyah Rusdiansyah
Universitas Bina Sarana Informatika,Indonesia
4 shared publications - Tuslaela Tuslaela
Universitas Nusa Mandiri, Indonesia
4 shared publications - Ahmad Setiadi
Universitas Bina Sarana Informatika
2 shared publications - Miwan Kurniawan Hidayat
3)Teknik Industri, Universitas Bina Sarana Informatika, Jakarta, Indonesia
2 shared publications - Henny Leidiyana
Universitas Bina Sarana Informatika
1 shared publication
Latest Publications
- Open Access
- 2 October 2024
- Published by Politeknik Ganesha in sinkron
- Vol. 8 (4) , 2439-2446
Cluster Analysis of Food Social Assistance in DKI Jakarta: K-Means Approach to Identify Expenditure Patterns and Beneficiaries
This study aims to evaluate the effectiveness of the K-Means algorithm in grouping social assistance recipients in DKI Jakarta based on various demographic and economic factors, such as income, number of family members, and living conditions. The main objective of this study is to optimize resource allocation in social assistance programs by identifying different recipient clusters, so that aid distribution becomes more targeted. In this study, the K-Means algorithm was used with an optimal number of clusters of 3, and produced an accuracy rate of 85%, indicating that this algorithm is effective in grouping large-scale and complex data. However, there are challenges related to the sensitivity of K-Means to outliers and data imbalances that affect the results of the analysis. The results also show that areas such as Central Jakarta and South Jakarta receive more social assistance compared to other areas such as North Jakarta and East Jakarta, reflecting differences in needs in various regions. These findings emphasize the importance of selecting the right variables, such as access to health facilities and economic conditions, in producing more accurate groupings. Overall, this study provides valuable insights into efforts to optimize the distribution of social assistance in DKI Jakarta and recommends further research to address the limitations that exist in the use of the K-Means algorithm, especially in the context of data that is imbalanced or has large variations.
- 2 October 2024
- Published by Politeknik Ganesha in sinkron
- Vol. 8 (4) , 2439-2446
Open AccessCluster Analysis of Food Social Assistance in DKI Jakarta: K-Means Approach to Identify Expenditure Patterns and Beneficiaries
This study aims to evaluate the effectiveness of the K-Means algorithm in grouping social assistance recipients in DKI Jakarta based on various demographic and economic factors, such as income, number of family members, and living conditions. The main objective of this study is to optimize resource allocation in social assistance programs by identifying different recipient clusters, so that aid distribution becomes more targeted. In this study, the K-Means algorithm was used with an optimal number of clusters of 3, and produced an accuracy rate of 85%, indicating that this algorithm is effective in grouping large-scale and complex data. However, there are challenges related to the sensitivity of K-Means to outliers and data imbalances that affect the results of the analysis. The results also show that areas such as Central Jakarta and South Jakarta receive more social assistance compared to other areas such as North Jakarta and East Jakarta, reflecting differences in needs in various regions. These findings emphasize the importance of selecting the right variables, such as access to health facilities and economic conditions, in producing more accurate groupings. Overall, this study provides valuable insights into efforts to optimize the distribution of social assistance in DKI Jakarta and recommends further research to address the limitations that exist in the use of the K-Means algorithm, especially in the context of data that is imbalanced or has large variations.
- Open Access
- 1 January 2024
- Published by Politeknik Ganesha in sinkron
- Vol. 9 (1) , 54-60
Implementation of K-Means Clustering in Food Security by Regency in East Java Province in 2022
Food is the main need that society must fulfill. If food security is disrupted, it will have a negative impact on the nation's life. The agricultural sector has an important role in West Java Province. This province has a large area of agricultural land, so it has high potential to produce abundant agricultural production. However, knowing the adequate number of farmers is very important. Therefore, the implementation of K-Means Clustering can make a significant contribution to the East Java Provincial Agriculture Service in grouping farmers by district. To achieve optimal results, determining the best K value needs to be considered carefully. K-Means Cluster Analysis is a method of non-hierarchical Cluster Analysis that groups data into one or more groups. Data with the same characteristics is grouped into one cluster and data with different characteristics is grouped into another cluster. The data used in this research are land area and rice production in the Regency of East Java Province in 2022. Based on the results of research with the object of Food Security, it can be concluded that, the results of the analysis of the application of manual data mining calculations in Excel Software using the K-Means Clustering method, resulted in two types of clustering in the form of C0, namely the Highest Land Area and Production group with 4 districts: Jember Regency, Ngawi Regency, Bojonegoro Regency and Lamongan Regency, for C1 clustering, namely the Lowest Land Area and Production group with 25 districts in East Java Province
- 1 January 2024
- Published by Politeknik Ganesha in sinkron
- Vol. 9 (1) , 54-60
Open AccessImplementation of K-Means Clustering in Food Security by Regency in East Java Province in 2022
Food is the main need that society must fulfill. If food security is disrupted, it will have a negative impact on the nation's life. The agricultural sector has an important role in West Java Province. This province has a large area of agricultural land, so it has high potential to produce abundant agricultural production. However, knowing the adequate number of farmers is very important. Therefore, the implementation of K-Means Clustering can make a significant contribution to the East Java Provincial Agriculture Service in grouping farmers by district. To achieve optimal results, determining the best K value needs to be considered carefully. K-Means Cluster Analysis is a method of non-hierarchical Cluster Analysis that groups data into one or more groups. Data with the same characteristics is grouped into one cluster and data with different characteristics is grouped into another cluster. The data used in this research are land area and rice production in the Regency of East Java Province in 2022. Based on the results of research with the object of Food Security, it can be concluded that, the results of the analysis of the application of manual data mining calculations in Excel Software using the K-Means Clustering method, resulted in two types of clustering in the form of C0, namely the Highest Land Area and Production group with 4 districts: Jember Regency, Ngawi Regency, Bojonegoro Regency and Lamongan Regency, for C1 clustering, namely the Lowest Land Area and Production group with 25 districts in East Java Province
- Open Access
- 4 April 2023
- Published by Politeknik Ganesha in sinkron
- Vol. 8 (2) , 873-879
Application of the Naïve Bayes Algorithm in Determining Sales Of The Month
One important factor for creating a healthy and growing company is the existence of sales rewards for employees to achieve sales targets every month. Assessing employees is not an easy thing when there are so many employees. This will make the assessment team have to look at the criteria carefully and carefully. Data manipulation can occur because it is difficult to make decisions with such large criteria and data without automated data mining. As a result, the company will not get competitive human resources. Sales targets are one of the keys to sales success because with sales targets, the sales prediction value can be used as a guide as a reference in determining product sales. One way to make better sales predictions is by utilizing data mining processing using the Naive Bayes algorithm. The Naive Bayes algorithm calculates the probability value of each of the attributes examined including attendance, sales targets and sales returns. Research with employee absence criteria, monthly sales and monthly sales invoice returns. From the results of the research that has been done, it can be concluded that the application of the Naive Bayes classifier method to the target data set for sales of goods achieves an optimization level of 95.78%, with attendance criteria greatly affecting employee performance so that product sales targets each month can be achieved
- 4 April 2023
- Published by Politeknik Ganesha in sinkron
- Vol. 8 (2) , 873-879
Open AccessApplication of the Naïve Bayes Algorithm in Determining Sales Of The Month
One important factor for creating a healthy and growing company is the existence of sales rewards for employees to achieve sales targets every month. Assessing employees is not an easy thing when there are so many employees. This will make the assessment team have to look at the criteria carefully and carefully. Data manipulation can occur because it is difficult to make decisions with such large criteria and data without automated data mining. As a result, the company will not get competitive human resources. Sales targets are one of the keys to sales success because with sales targets, the sales prediction value can be used as a guide as a reference in determining product sales. One way to make better sales predictions is by utilizing data mining processing using the Naive Bayes algorithm. The Naive Bayes algorithm calculates the probability value of each of the attributes examined including attendance, sales targets and sales returns. Research with employee absence criteria, monthly sales and monthly sales invoice returns. From the results of the research that has been done, it can be concluded that the application of the Naive Bayes classifier method to the target data set for sales of goods achieves an optimization level of 95.78%, with attendance criteria greatly affecting employee performance so that product sales targets each month can be achieved
- Open Access
- 1 January 2023
- Published by AIP Publishing in AIP Conference Proceedings
- Vol. 2714 (1) , 020010
Classification of potato leaf diseases based on texture, shape and color features using the random forest algorithm
Early identification of leaf diseases is a good step to help food security in agriculture. Late blight and early blight are common diseases that attack potato l
- 1 January 2023
- Published by AIP Publishing in AIP Conference Proceedings
- Vol. 2714 (1) , 020010
Open AccessClassification of potato leaf diseases based on texture, shape and color features using the random forest algorithm
Early identification of leaf diseases is a good step to help food security in agriculture. Late blight and early blight are common diseases that attack potato l
- Open Access
- 1 January 2023
- Published by AIP Publishing in AIP Conference Proceedings
- Vol. 2714 (1)
Comparison of machine learning algorithm and feature selection particle swarm optimization on software effort estimation
Software Effort Estimation is an important process in any software project. Several studies have been carried out in estimating software using various methods, both machine learning (ML) and non-machine learning (non-ML) methods. In this study, using datasets that can be used in general (public), then an experiment was carried out by comparing the two k-Nearst Neighbor (k-NN) machine learning algorithms and Linear Regression besides that it was also combined with the Particle Swarm Optimization attribute selection on the project parameters. The dataset of effort estimation software used in this study is Desharnais, Maxwell, Kitchenham CSC, and Kamrer. The results of this study show that the comparison of all datasets used by the k-NN model is better than the Linear Regression model. In addition, the results of this study also show that the Particle Swarm Optimization (PSO) feature selection can reduce the RMSE value using either the k-NN model or the Linear Regression model. However, of the four datasets used, there is one with the opposite value, namely an increase in the RMSE value on the Maxwell dataset using the Linear Regression model. Where the RMSE value without the selection feature is 2594,036, while after adding the selection feature it becomes 37.05,870. If you compare the 2 models used in this research, the k-NN model shows that the RMSE value is better than the Linear Regression model.
- 1 January 2023
- Published by AIP Publishing in AIP Conference Proceedings
- Vol. 2714 (1)
Open AccessComparison of machine learning algorithm and feature selection particle swarm optimization on software effort estimation
Software Effort Estimation is an important process in any software project. Several studies have been carried out in estimating software using various methods, both machine learning (ML) and non-machine learning (non-ML) methods. In this study, using datasets that can be used in general (public), then an experiment was carried out by comparing the two k-Nearst Neighbor (k-NN) machine learning algorithms and Linear Regression besides that it was also combined with the Particle Swarm Optimization attribute selection on the project parameters. The dataset of effort estimation software used in this study is Desharnais, Maxwell, Kitchenham CSC, and Kamrer. The results of this study show that the comparison of all datasets used by the k-NN model is better than the Linear Regression model. In addition, the results of this study also show that the Particle Swarm Optimization (PSO) feature selection can reduce the RMSE value using either the k-NN model or the Linear Regression model. However, of the four datasets used, there is one with the opposite value, namely an increase in the RMSE value on the Maxwell dataset using the Linear Regression model. Where the RMSE value without the selection feature is 2594,036, while after adding the selection feature it becomes 37.05,870. If you compare the 2 models used in this research, the k-NN model shows that the RMSE value is better than the Linear Regression model.
- Open Access
- 1 October 2022
- Published by Politeknik Ganesha in sinkron
- Vol. 7 (4) , 2208-2213
Data Mining implementation on SMUN Scholarship recipient candidates using the C4.5 algorithm
This research is motivated by SMUN previously only providing scholarships for underprivileged students, while for students who excel, the method is not yet, therefore, research by implementing data mining with the C4.5 algorithm method so that in the future, if the scholarship program is for outstanding students already exists, then SMUN can immediately apply it. The sample data are 33 scholarship recipients with subjects that have been determined based on the report card scores of prospective scholarship recipients. The selection results for receiving scholarships to outstanding students so far have only been calculated in general terms, the possibility of incorrect results will cause losses for students who have good achievements. As a result, there are students who should get the scholarship rights and do not receive it. The problem of this research is how to make it easier to analyze students who need and are entitled to receive scholarships, how to make it easier to select and determine scholarship recipients. The research uses the C4.5 algorithm to overcome these problems. With the C4.5 algorithm, the percentage accuracy value can be calculated which is used to determine whether the student is entitled to receive a scholarship or not. The data collection technique in this study is to use student report cards. classification value in determining students who get scholarships with the highest accuracy of 81.81%. The accuracy value was obtained by experimenting with the testing process with the number of student data as many as 33 students. It is hoped that it will make it easier to receive scholarships for students who excel academically in school.
- 1 October 2022
- Published by Politeknik Ganesha in sinkron
- Vol. 7 (4) , 2208-2213
Open AccessData Mining implementation on SMUN Scholarship recipient candidates using the C4.5 algorithm
This research is motivated by SMUN previously only providing scholarships for underprivileged students, while for students who excel, the method is not yet, therefore, research by implementing data mining with the C4.5 algorithm method so that in the future, if the scholarship program is for outstanding students already exists, then SMUN can immediately apply it. The sample data are 33 scholarship recipients with subjects that have been determined based on the report card scores of prospective scholarship recipients. The selection results for receiving scholarships to outstanding students so far have only been calculated in general terms, the possibility of incorrect results will cause losses for students who have good achievements. As a result, there are students who should get the scholarship rights and do not receive it. The problem of this research is how to make it easier to analyze students who need and are entitled to receive scholarships, how to make it easier to select and determine scholarship recipients. The research uses the C4.5 algorithm to overcome these problems. With the C4.5 algorithm, the percentage accuracy value can be calculated which is used to determine whether the student is entitled to receive a scholarship or not. The data collection technique in this study is to use student report cards. classification value in determining students who get scholarships with the highest accuracy of 81.81%. The accuracy value was obtained by experimenting with the testing process with the number of student data as many as 33 students. It is hoped that it will make it easier to receive scholarships for students who excel academically in school.
- Open Access
- 23 September 2019
- Published by Universitas Islam 45 in PIKSEL : Penelitian Ilmu Komputer Sistem Embedded and Logic
- Vol. 7 (2) , 123-132
Kualitas Pelayanan Dalam Voice Over Internet Protokol Berbasis Shorewall
Communication is very important and a success factor of the system at the company. Communication can be done using telephone media and internet media. PT. Interdev Prakarsa has several branches and communication between branches is still done using telephone media, and it is very costly for the company especially since the company is already using internet media in the network. The solution to this problem in this study was designing an internet-based technology as communication between branches, the technology is called Voice over Internet Protocol (VoIP) with an open source operating system and an open source firewall that was Shorewall. Result showed that after installation and testing, this firewall reliable enough to overcome the problem of attack problems from outside. The results of VoIP tightening on service quality found that the application of VoIP did not consume a lot of CPU Benchmarks, i.e. only 0.80 to 1.35 per cent. The bandwidth used is also very small between 86 to 86.8 kbps for incoming calls and 83.4 up to 84.3 kbps for outgoing calls. Communication built between VoIP peripherals has also been tested to run well because the value of delay, jitter and packet loss is included in the good category. Keywords: VoIP, Shorewall, Comunication Abstrak Komunikasi pada sebuah perusahaan sangatlah penting, dimana komunikasi menjadi alat ukur keberhasilan sistem di perusahaan tersebut. Komunikasi yang dilakukan dapat menggunakan media telepon dan media internet. PT. Interdev Prakarsa memiliki beberapa cabang dan komunikasi antar cabang masih dilakukan dengan menggunakan media telepon, dan itu sangatlah menghabiskan cost perusahaan apalagi perusahaan tersebut sudah menggunakan media internet dalam jaringan. Solusi dari permasalahan tersebut adalah dengan mendesain sebuah teknologi yang berbasis jaringan internet sebagai komunikasi antar cabang, teknologi tersebut bernama Voice over Internet Protocol (VoIP) dengan sebuah sistem operasi open source dan sebuah firewall open source yaitu Shorewall dimana setelah dilakukan penginstallan dan pengetesan, firewall ini cukup handal untuk mengatasi masalah masalah serangan dari luar. Hasil dari pengetasan VoIP terhadap kualitas pelayanan didapatkan bahwa penerapan VoIP tidak banyak menghabiskan Benchmark CPU yaitu hanya 0,80 sampai degan 1.35 persen per call. Untuk bandwidth yang di pakai juga sangatlah kecil berada di antara 86 sampai dengan 86.8 kbps untuk panggilan masuk dan 83.4 sampai dengan 84.3 kbps untuk panggilan keluar. Komunikasi yang dibangun antar peripheral VoIP pun teruji berjalan dengan baik karena nilai dari delay, jitter dan packet loss termasuk dalam katagori baik. Kata kunci: VoIP, Shorewall, Komunikasi
- 23 September 2019
- Published by Universitas Islam 45 in PIKSEL : Penelitian Ilmu Komputer Sistem Embedded and Logic
- Vol. 7 (2) , 123-132
Open AccessKualitas Pelayanan Dalam Voice Over Internet Protokol Berbasis Shorewall
Communication is very important and a success factor of the system at the company. Communication can be done using telephone media and internet media. PT. Interdev Prakarsa has several branches and communication between branches is still done using telephone media, and it is very costly for the company especially since the company is already using internet media in the network. The solution to this problem in this study was designing an internet-based technology as communication between branches, the technology is called Voice over Internet Protocol (VoIP) with an open source operating system and an open source firewall that was Shorewall. Result showed that after installation and testing, this firewall reliable enough to overcome the problem of attack problems from outside. The results of VoIP tightening on service quality found that the application of VoIP did not consume a lot of CPU Benchmarks, i.e. only 0.80 to 1.35 per cent. The bandwidth used is also very small between 86 to 86.8 kbps for incoming calls and 83.4 up to 84.3 kbps for outgoing calls. Communication built between VoIP peripherals has also been tested to run well because the value of delay, jitter and packet loss is included in the good category. Keywords: VoIP, Shorewall, Comunication Abstrak Komunikasi pada sebuah perusahaan sangatlah penting, dimana komunikasi menjadi alat ukur keberhasilan sistem di perusahaan tersebut. Komunikasi yang dilakukan dapat menggunakan media telepon dan media internet. PT. Interdev Prakarsa memiliki beberapa cabang dan komunikasi antar cabang masih dilakukan dengan menggunakan media telepon, dan itu sangatlah menghabiskan cost perusahaan apalagi perusahaan tersebut sudah menggunakan media internet dalam jaringan. Solusi dari permasalahan tersebut adalah dengan mendesain sebuah teknologi yang berbasis jaringan internet sebagai komunikasi antar cabang, teknologi tersebut bernama Voice over Internet Protocol (VoIP) dengan sebuah sistem operasi open source dan sebuah firewall open source yaitu Shorewall dimana setelah dilakukan penginstallan dan pengetesan, firewall ini cukup handal untuk mengatasi masalah masalah serangan dari luar. Hasil dari pengetasan VoIP terhadap kualitas pelayanan didapatkan bahwa penerapan VoIP tidak banyak menghabiskan Benchmark CPU yaitu hanya 0,80 sampai degan 1.35 persen per call. Untuk bandwidth yang di pakai juga sangatlah kecil berada di antara 86 sampai dengan 86.8 kbps untuk panggilan masuk dan 83.4 sampai dengan 84.3 kbps untuk panggilan keluar. Komunikasi yang dibangun antar peripheral VoIP pun teruji berjalan dengan baik karena nilai dari delay, jitter dan packet loss termasuk dalam katagori baik. Kata kunci: VoIP, Shorewall, Komunikasi
- Open Access
- 2 February 2015
- Published by STMIK Antar Bangsa in Jurnal Sistem Informasi
- Vol. 4 (1) , 55-63
PENERAPAN METODE SVM UNTUK KLASIFIKASI RESIKO KREDIT KEPEMILIKAN KENDARAAN (MOTOR)
—Given Credit Risk in a lease is always there , So credit analysis should be done carefully . if Consumer troubled in paying installment payments , so would be detrimental to the lease . Therefore , quantitative and qualitative Credit analyzing will provide clarity for decision makers . To achieve this goal , credit preparations should be done by collecting information and data. The quality of the result of the analysis depends on the quality of human resources , the data, and analysis techniques .In this research discussed regarding the application using support vector machine ( SVM ) method for determination of the Credit Risk of Motor Vehicle ownership . The resulting model is then evaluated its performance by calculating accuracy model prediction using Confusion Matrix and ROC curve, obtained 79.78 % accuracy value and AUC value is 0.894. The classification is Good Enough because have AUC values between from 0.8 to 0.9.Intisari— Mengingat resiko tidak kembalinya kredit pada sebuah leasing, selalu ada, maka analisis kredit dengan cermat perlu dilakukan. Jika konsumen bermasalah dalam pembayaran angsuran, maka akan merugikan pihak leasing. Oleh karena itu, penilaian kuantitatif dan kualitatif dalam menganalisa kredit akan memberikan kejelasan bagi pembuat keputusan. Untuk mewujudkan hal tersebut, perlu dilakukan persiapan kredit, yaitu dengan mengumpulkan informasi dan data untuk bahan analisis. Kualitas hasil analisis tergantung pada kualitas SDM, data yang diperoleh, dan teknik analisis.Dalam penelitian ini dibahas mengenai penerapan metode Support vector machines (SVM) untuk penentuan resiko kredit kepemilikan kendaraan bermotor. Model yang dihasilkan kemudian dievaluasi performansinya dengan cara menghitung akurasi model dalam memprediksi menggunakan metode Confusion Matrix dan Kurva ROC, didapat nilai akurasi 79.78% dan termasuk klasifikasi cukup baik karena memiliki nilai AUC antara 0.8-0.9, yaitu sebesar 0.894.Kata kunci: Support Vector Machines, Confusion Matrix, Kurva ROC
- 2 February 2015
- Published by STMIK Antar Bangsa in Jurnal Sistem Informasi
- Vol. 4 (1) , 55-63
Open AccessPENERAPAN METODE SVM UNTUK KLASIFIKASI RESIKO KREDIT KEPEMILIKAN KENDARAAN (MOTOR)
—Given Credit Risk in a lease is always there , So credit analysis should be done carefully . if Consumer troubled in paying installment payments , so would be detrimental to the lease . Therefore , quantitative and qualitative Credit analyzing will provide clarity for decision makers . To achieve this goal , credit preparations should be done by collecting information and data. The quality of the result of the analysis depends on the quality of human resources , the data, and analysis techniques .In this research discussed regarding the application using support vector machine ( SVM ) method for determination of the Credit Risk of Motor Vehicle ownership . The resulting model is then evaluated its performance by calculating accuracy model prediction using Confusion Matrix and ROC curve, obtained 79.78 % accuracy value and AUC value is 0.894. The classification is Good Enough because have AUC values between from 0.8 to 0.9.Intisari— Mengingat resiko tidak kembalinya kredit pada sebuah leasing, selalu ada, maka analisis kredit dengan cermat perlu dilakukan. Jika konsumen bermasalah dalam pembayaran angsuran, maka akan merugikan pihak leasing. Oleh karena itu, penilaian kuantitatif dan kualitatif dalam menganalisa kredit akan memberikan kejelasan bagi pembuat keputusan. Untuk mewujudkan hal tersebut, perlu dilakukan persiapan kredit, yaitu dengan mengumpulkan informasi dan data untuk bahan analisis. Kualitas hasil analisis tergantung pada kualitas SDM, data yang diperoleh, dan teknik analisis.Dalam penelitian ini dibahas mengenai penerapan metode Support vector machines (SVM) untuk penentuan resiko kredit kepemilikan kendaraan bermotor. Model yang dihasilkan kemudian dievaluasi performansinya dengan cara menghitung akurasi model dalam memprediksi menggunakan metode Confusion Matrix dan Kurva ROC, didapat nilai akurasi 79.78% dan termasuk klasifikasi cukup baik karena memiliki nilai AUC antara 0.8-0.9, yaitu sebesar 0.894.Kata kunci: Support Vector Machines, Confusion Matrix, Kurva ROC