Muhammad Noor Hasan Siregar
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Muhammad Noor Hasan Siregar has a total of 34 co-authors. Here are the top co-authors by number of shared publications:
- Dahlan Abdullah
Universitas Malikussaleh
2 shared publications - Ade Ismail Abdul Kodir
Sultan Agung Islamic University
1 shared publication - Adi Supriyatna
Universitas Bina Sarana Informatika
1 shared publication - Agus Perdana Windarto
STIKOM Tunas Bangsa, Pematangsiantar, Indonesia
1 shared publication - Akim Manaor Hara Pardede
STMIK Kaputama, Jl. Veteran No. 4A-9A, Binjai-Sumatera Utara, Indonesia
1 shared publication Andysah Putera Utama Siahaan
School of Computer Engineering and Communication, Universiti Malaysia Perlis, Malaysia
1 shared publication
Latest Publications
- Open Access
- 1 June 2026
- Published by Institute of Advanced Engineering and Science in Bulletin of Electrical Engineering and Informatics
- Vol. 15 (3) , 2275-2286
Hybrid XAI and deep learning architecture for trustworthy dental diagnostics
Dental periodontal disease is a persistent an inflammatory disorder affecting tooth supporting tissues and stays a main motive of tooth loss. Although dental radiographs are essential for early diagnosis, their interpretation is often subjective and inconsistent due to reliance on clinician expertise. This study proposes an automated and interpretable diagnostic framework using a convolutional neural network (CNN) integrated with gradient-weighted class activation mapping (Grad-CAM). The CNN performs binary classification of periapical radiographs into periodontal and normal categories, while Grad-CAM provides visual explanations of the model’s decision-making process. Experimental results show that the proposed model achieves a classification accuracy of 94.17%, indicating reliable diagnostic performance. The generated heatmaps consistently highlight clinically relevant regions, particularly alveolar bone loss in periodontal cases, whereas normal images exhibit no pathological activation. These findings demonstrate that the proposed CNN–Grad-CAM framework enhances both diagnostic accuracy and interpretability. The study contributes a transparent and trustworthy artificial intelligence solution to support objective periodontal disease diagnosis in dental radiology.
- 1 June 2026
- Published by Institute of Advanced Engineering and Science in Bulletin of Electrical Engineering and Informatics
- Vol. 15 (3) , 2275-2286
Open AccessHybrid XAI and deep learning architecture for trustworthy dental diagnostics
Dental periodontal disease is a persistent an inflammatory disorder affecting tooth supporting tissues and stays a main motive of tooth loss. Although dental radiographs are essential for early diagnosis, their interpretation is often subjective and inconsistent due to reliance on clinician expertise. This study proposes an automated and interpretable diagnostic framework using a convolutional neural network (CNN) integrated with gradient-weighted class activation mapping (Grad-CAM). The CNN performs binary classification of periapical radiographs into periodontal and normal categories, while Grad-CAM provides visual explanations of the model’s decision-making process. Experimental results show that the proposed model achieves a classification accuracy of 94.17%, indicating reliable diagnostic performance. The generated heatmaps consistently highlight clinically relevant regions, particularly alveolar bone loss in periodontal cases, whereas normal images exhibit no pathological activation. These findings demonstrate that the proposed CNN–Grad-CAM framework enhances both diagnostic accuracy and interpretability. The study contributes a transparent and trustworthy artificial intelligence solution to support objective periodontal disease diagnosis in dental radiology.
- Open Access
- 17 February 2021
- Published by STIKOM Tunas Bangsa Pematangsiantar in Jurasik (Jurnal Riset Sistem Informasi dan Teknik Informatika)
- Vol. 6 (1) , 21
Analisis Kombinasi itemset pada Bisnis Online dengan Teknik Asosiasi Data mining
Online business is one of the industries that thrives on social media. With business competition starting to grow a lot these days, businesses are setting up online businesses to boost sales. One suggestion is to reduce the price on combination of items that are commonly purchased at the same time. Using the transaction data obtained through purchase, an association rule may be used to discover the rules for combinations of items. The association process uses an a priori algorithm to access sales transaction data. The positive results of this study can be used to produce the strategies in the development of online businesses.
- 17 February 2021
- Published by STIKOM Tunas Bangsa Pematangsiantar in Jurasik (Jurnal Riset Sistem Informasi dan Teknik Informatika)
- Vol. 6 (1) , 21
Open AccessAnalisis Kombinasi itemset pada Bisnis Online dengan Teknik Asosiasi Data mining
Online business is one of the industries that thrives on social media. With business competition starting to grow a lot these days, businesses are setting up online businesses to boost sales. One suggestion is to reduce the price on combination of items that are commonly purchased at the same time. Using the transaction data obtained through purchase, an association rule may be used to discover the rules for combinations of items. The association process uses an a priori algorithm to access sales transaction data. The positive results of this study can be used to produce the strategies in the development of online businesses.
- Open Access
- 1 November 2019
- Published by IOP Publishing in Journal of Physics: Conference Series
- Vol. 1363 (1) , 012067
Application of Data Mining Prediction of Electricity Deviation Flow Using Metode Backpropogation at PLN Binjai Area
PLN is the only state-owned electricity company that provides and regulates electricity to the public. However, in the provision of electricity to every community, there are many frauds or irregularities in utilization of electricity by irresponsible parties was found by PLN. Electricity deviation caused by irresponsible parties, like the theft of electricity can have a tremendous impact due to loss of electrical energy. Electricity Deviation, using Electricity Deviation data for 3 years, from 2012 to 2014. Sales data processed by the network, in the form of amount of Kwh deviation per month. The data needed in the process of analysis of this Neural Network is pure database results by taking the input variable in the form of Amount of Kwh Deviation and Amount of Kwh Deviation for the next Month for the following year as output. Electricity for 3 years is processed into Matlab using GUI facilities, to produce information on the predicted results of the Kwh Deviation in November 2014 to March, May, June, August, November, December 2015 experiencing has increase in Kwh, while in April July, September, and Okrober 2015 decreased.
- 1 November 2019
- Published by IOP Publishing in Journal of Physics: Conference Series
- Vol. 1363 (1) , 012067
Open AccessApplication of Data Mining Prediction of Electricity Deviation Flow Using Metode Backpropogation at PLN Binjai Area
PLN is the only state-owned electricity company that provides and regulates electricity to the public. However, in the provision of electricity to every community, there are many frauds or irregularities in utilization of electricity by irresponsible parties was found by PLN. Electricity deviation caused by irresponsible parties, like the theft of electricity can have a tremendous impact due to loss of electrical energy. Electricity Deviation, using Electricity Deviation data for 3 years, from 2012 to 2014. Sales data processed by the network, in the form of amount of Kwh deviation per month. The data needed in the process of analysis of this Neural Network is pure database results by taking the input variable in the form of Amount of Kwh Deviation and Amount of Kwh Deviation for the next Month for the following year as output. Electricity for 3 years is processed into Matlab using GUI facilities, to produce information on the predicted results of the Kwh Deviation in November 2014 to March, May, June, August, November, December 2015 experiencing has increase in Kwh, while in April July, September, and Okrober 2015 decreased.
- Open Access
- 1 August 2019
- Published by IOP Publishing in Journal of Physics: Conference Series
- Vol. 1255 (1) , 012001
Analysis of the K-Means Algorithm on Clean Water Customers Based on the Province
One of the important needs of environmental health is clean water. Clean water is the most important necessity of living beings in supporting survival. The study aimed to cluster the number of cleaned water customers by province (1995-2015). The method used is data mining clustering using k-means. The sample data used 34 provinces with attribute assessment of the number of cleaned water customers by province. The clustering process is done with 3 clusters, namely (C1) Cluster High, (C2) Cluster Normal and (C3) Cluster Low, for the number of cleaned water customers who are low on the need of clean water. The results showed, C1: 6 provinces, C2: 4 provinces and C3: 24 provinces. The end centroid values used are: C1 (296587.22), C2 (995898.56) and C3 (70832.29). The results obtained on the Davies-Bouldin index for “the number of cleaned water consumers” are -0.470. based on performance results, it can be concluded that k-means algorithm is best because it has the smallest Davies-Bouldin index value. Based on research results, 70% of Indonesian people are still low awareness of the need for clean water.
- 1 August 2019
- Published by IOP Publishing in Journal of Physics: Conference Series
- Vol. 1255 (1) , 012001
Open AccessAnalysis of the K-Means Algorithm on Clean Water Customers Based on the Province
One of the important needs of environmental health is clean water. Clean water is the most important necessity of living beings in supporting survival. The study aimed to cluster the number of cleaned water customers by province (1995-2015). The method used is data mining clustering using k-means. The sample data used 34 provinces with attribute assessment of the number of cleaned water customers by province. The clustering process is done with 3 clusters, namely (C1) Cluster High, (C2) Cluster Normal and (C3) Cluster Low, for the number of cleaned water customers who are low on the need of clean water. The results showed, C1: 6 provinces, C2: 4 provinces and C3: 24 provinces. The end centroid values used are: C1 (296587.22), C2 (995898.56) and C3 (70832.29). The results obtained on the Davies-Bouldin index for “the number of cleaned water consumers” are -0.470. based on performance results, it can be concluded that k-means algorithm is best because it has the smallest Davies-Bouldin index value. Based on research results, 70% of Indonesian people are still low awareness of the need for clean water.
- 23 May 2019
- Published by STIKOM Tunas Bangsa Pematangsiantar in IJISTECH (International Journal of Information System and Technology)
- Vol. 2 (2) , 59-66
Model Combination of Activation Functions in Neural Network Algorithms (Case: Retail State Sukuk by Group)
This study aims to maximize the activation function used in backpropogation networks in finding the best architectural model. The case study used is the sale of state retail sukuk based on professional groups. The combination of activation functions used for training and testing is tansig-tansig, tansig-purelin and tansig logsig. The architectural model used is the architectural model 6-2-1 and 6-5-1. The evaluation parameters used are epoch, MSE training, MSE testing and accuracy level of truth. Data processing is assisted by using Matlab software. The results showed that the tansig-logsig activation function had more stable results than tansig-tansig and tansig-purelin.
- 23 May 2019
- Published by STIKOM Tunas Bangsa Pematangsiantar in IJISTECH (International Journal of Information System and Technology)
- Vol. 2 (2) , 59-66
Model Combination of Activation Functions in Neural Network Algorithms (Case: Retail State Sukuk by Group)
This study aims to maximize the activation function used in backpropogation networks in finding the best architectural model. The case study used is the sale of state retail sukuk based on professional groups. The combination of activation functions used for training and testing is tansig-tansig, tansig-purelin and tansig logsig. The architectural model used is the architectural model 6-2-1 and 6-5-1. The evaluation parameters used are epoch, MSE training, MSE testing and accuracy level of truth. Data processing is assisted by using Matlab software. The results showed that the tansig-logsig activation function had more stable results than tansig-tansig and tansig-purelin.
- Open Access
- 9 September 2018
- Published by Universitas Islam Sumatera Utara in InfoTekJar (Jurnal Nasional Informatika dan Teknologi Jaringan)
- Vol. 3 (1) , 1-5
Model Arsitektur Artificial Neural Network pada Pelanggan Listrik Negara (PLN)
Perusahaan Listrik Negara (PLN) merupakan sebuah BUMN yang mengurusi semua aspek kelistrikan yang ada di Indonesia. Meningkatnya kebutuhan masyarakat akan listrik seiring tumbuhnya populasi membuat pelanggan listrik terus bertambah. Penelitian ini bertujuan untuk membuat sebuah model prediksi dengan memanfaatkan kecerdasan buatan yakni Jaringan saraf Tiruan dengan menggunakan algoritma Backpropogation. Data penelitian bersumber dan diolah oleh Badan Pusat Statistik Indonesia (https://www.bps.go.id). Data masukan adalah kelompok pelanggan PLN yang dibagi kedalan 5 kategori yakni Sosial, Rumah Tangga, Bisnis, Industri dan Publik dengan data jumlah pelanggan (2006-2015). Hasil penelitian menunjukkan bahwa 4 pengujian model arsitektur yaitu 5-10-1, 5-25-1, 5-10-25-1 dan 5-25-10-1 diperoleh model 5-25-1 adalah model arsitektur terbaik dengan parameter MSE Pelatihan 0,0009994101, MSE Pengujian 0,0011603685, Epoch 520 dan Akurasi 80%. Diharapkan penelitian ini dapat memberikan prediksi kepada pihak PLN kedepanya tentang jumlah peningkatan pelanggan PLN mengingat listrik adalah salah satu kebutuhan masyarakat.
- 9 September 2018
- Published by Universitas Islam Sumatera Utara in InfoTekJar (Jurnal Nasional Informatika dan Teknologi Jaringan)
- Vol. 3 (1) , 1-5
Open AccessModel Arsitektur Artificial Neural Network pada Pelanggan Listrik Negara (PLN)
Perusahaan Listrik Negara (PLN) merupakan sebuah BUMN yang mengurusi semua aspek kelistrikan yang ada di Indonesia. Meningkatnya kebutuhan masyarakat akan listrik seiring tumbuhnya populasi membuat pelanggan listrik terus bertambah. Penelitian ini bertujuan untuk membuat sebuah model prediksi dengan memanfaatkan kecerdasan buatan yakni Jaringan saraf Tiruan dengan menggunakan algoritma Backpropogation. Data penelitian bersumber dan diolah oleh Badan Pusat Statistik Indonesia (https://www.bps.go.id). Data masukan adalah kelompok pelanggan PLN yang dibagi kedalan 5 kategori yakni Sosial, Rumah Tangga, Bisnis, Industri dan Publik dengan data jumlah pelanggan (2006-2015). Hasil penelitian menunjukkan bahwa 4 pengujian model arsitektur yaitu 5-10-1, 5-25-1, 5-10-25-1 dan 5-25-10-1 diperoleh model 5-25-1 adalah model arsitektur terbaik dengan parameter MSE Pelatihan 0,0009994101, MSE Pengujian 0,0011603685, Epoch 520 dan Akurasi 80%. Diharapkan penelitian ini dapat memberikan prediksi kepada pihak PLN kedepanya tentang jumlah peningkatan pelanggan PLN mengingat listrik adalah salah satu kebutuhan masyarakat.
- 25 May 2018
- Published by STIKOM Tunas Bangsa Pematangsiantar in IJISTECH (International Journal of Information System and Technology)
- Vol. 1 (2) , 30-40
ANN: Model of Back-Propagation Architecture on the Logs Production by Type of Wood
Indonesia is rich in forest products. The production forest is a forest area that can be utilized for the community, such as logs, rattan, and some forest plants that have high economic value. This research aims to make the best architectural model by using artificial neural network. The method used is backpropagation algorithm. With this model it will continue to predict the output of logs. Data are sourced from BPS-Statistics Indonesia. Based on the results of research results of logs production using backpropogation method, obtained the result of 3 model architecture (18-18-1, 18-25-1 and 18-18-25- 1) that model of architecture 18- 25-1 is the best model with 72% accuracy, MSE: 0.0221670942 and epochs: 660.
- 25 May 2018
- Published by STIKOM Tunas Bangsa Pematangsiantar in IJISTECH (International Journal of Information System and Technology)
- Vol. 1 (2) , 30-40
ANN: Model of Back-Propagation Architecture on the Logs Production by Type of Wood
Indonesia is rich in forest products. The production forest is a forest area that can be utilized for the community, such as logs, rattan, and some forest plants that have high economic value. This research aims to make the best architectural model by using artificial neural network. The method used is backpropagation algorithm. With this model it will continue to predict the output of logs. Data are sourced from BPS-Statistics Indonesia. Based on the results of research results of logs production using backpropogation method, obtained the result of 3 model architecture (18-18-1, 18-25-1 and 18-18-25- 1) that model of architecture 18- 25-1 is the best model with 72% accuracy, MSE: 0.0221670942 and epochs: 660.
- Open Access
- 30 April 2018
- Published by IOP Publishing in Journal of Physics: Conference Series
- Vol. 1007 (1) , 012004
Searching Process with Raita Algorithm and its Application
Searching is a common process performed by many computer users, Raita algorithm is one algorithm that can be used to match and find information in accordance with the patterns entered. Raita algorithm applied to the file search application using java programming language and the results obtained from the testing process of the file search quickly and with accurate results and support many data types.
- 30 April 2018
- Published by IOP Publishing in Journal of Physics: Conference Series
- Vol. 1007 (1) , 012004
Open AccessSearching Process with Raita Algorithm and its Application
Searching is a common process performed by many computer users, Raita algorithm is one algorithm that can be used to match and find information in accordance with the patterns entered. Raita algorithm applied to the file search application using java programming language and the results obtained from the testing process of the file search quickly and with accurate results and support many data types.
- Open Access
- 13 November 2017
- Published by STIKOM Tunas Bangsa Pematangsiantar in IJISTECH (International Journal of Information System and Technology)
- Vol. 1 (1) , 22-33
Neural Network Analysis With Backpropogation In Predicting Human Development Index (HDI) Component by Regency/City In North Sumatera
Human Development Index (HDI) measures human development outcomes based on a number of basic components of quality of life. As a measure of the quality of life, HDI is built through a basic three-dimensional approach. Data obtained from the Central Bureau of Statistics 2015 for Human Development Index (HDI) by Regency / City in North Sumatera Province consisting of 32 alternatives and with 4 parameters ie life expectancy (year), expectation, school length (%), the average length of school (year) and per capita real expenditure (Rp). By using backpropagation obtained result of 6 testing of architecture pattern that is: 4-5-1, 4-10-1, 4-5-10-1, 4-10-5-1, 4-10-20-1 and 4- 15-20-1 obtained best architectural pattern is 4-10-20-1 with epoch 2126, error 0.0011757393, execution time 00:16 and accuracy 100%.
- 13 November 2017
- Published by STIKOM Tunas Bangsa Pematangsiantar in IJISTECH (International Journal of Information System and Technology)
- Vol. 1 (1) , 22-33
Open AccessNeural Network Analysis With Backpropogation In Predicting Human Development Index (HDI) Component by Regency/City In North Sumatera
Human Development Index (HDI) measures human development outcomes based on a number of basic components of quality of life. As a measure of the quality of life, HDI is built through a basic three-dimensional approach. Data obtained from the Central Bureau of Statistics 2015 for Human Development Index (HDI) by Regency / City in North Sumatera Province consisting of 32 alternatives and with 4 parameters ie life expectancy (year), expectation, school length (%), the average length of school (year) and per capita real expenditure (Rp). By using backpropagation obtained result of 6 testing of architecture pattern that is: 4-5-1, 4-10-1, 4-5-10-1, 4-10-5-1, 4-10-20-1 and 4- 15-20-1 obtained best architectural pattern is 4-10-20-1 with epoch 2126, error 0.0011757393, execution time 00:16 and accuracy 100%.
- Open Access
- 28 February 2017
- Published by Center for Journal Management and Publication, Lambung Mangkurat University in KLIK - KUMPULAN JURNAL ILMU KOMPUTER
- Vol. 4 (1) , 59-69
IMPLEMENTASI WEIGHT PRODUCT MODEL (WPM) DALAM MENENTUKAN PEMILIHAN SEPEDA MOTOR SPORT BERBASIS SPK
Perkembangan dunia otomtif khususnya sepeda motor menjadi salah satu trend dikalangan masyrakat saat ini. Banyak jenis produk khususnya sepeda motor sport yang ditawarkan kepada masyrakat. Produk produk ini memiliki keunggulan masing masing. Penelitian ini bertujuan untuk membangun sebuah sistem pendukung keputusan dengan menggunakan metode Weight Product Model (WPM) untuk menentukan pemilihan sepeda motor sport yang paling diminati. Penelitian ini dilaksanakan dengan observasi dan interview untuk melakukan pengumpulan data sepeda motor. Kesimpulan hasil penelitian ini adalah Pemberian kriteria-kriteria dalam pemilihan sepeda motor sport dapat membantu dalam mengambil keputusan untuk menentukan sepeda motor sport yang bagus dan sesuai dengan keinginan konsumen. Dengan Menerapkan metode Weight Product Model (WPM) proses pemilihan sepeda sepeda motor sport lebih efisien dan praktis.Kata Kunci: Pemilihan, SPK, Sepeda motor sport, Metode WPM, Pematangsiantar
- 28 February 2017
- Published by Center for Journal Management and Publication, Lambung Mangkurat University in KLIK - KUMPULAN JURNAL ILMU KOMPUTER
- Vol. 4 (1) , 59-69
Open AccessIMPLEMENTASI WEIGHT PRODUCT MODEL (WPM) DALAM MENENTUKAN PEMILIHAN SEPEDA MOTOR SPORT BERBASIS SPK
Perkembangan dunia otomtif khususnya sepeda motor menjadi salah satu trend dikalangan masyrakat saat ini. Banyak jenis produk khususnya sepeda motor sport yang ditawarkan kepada masyrakat. Produk produk ini memiliki keunggulan masing masing. Penelitian ini bertujuan untuk membangun sebuah sistem pendukung keputusan dengan menggunakan metode Weight Product Model (WPM) untuk menentukan pemilihan sepeda motor sport yang paling diminati. Penelitian ini dilaksanakan dengan observasi dan interview untuk melakukan pengumpulan data sepeda motor. Kesimpulan hasil penelitian ini adalah Pemberian kriteria-kriteria dalam pemilihan sepeda motor sport dapat membantu dalam mengambil keputusan untuk menentukan sepeda motor sport yang bagus dan sesuai dengan keinginan konsumen. Dengan Menerapkan metode Weight Product Model (WPM) proses pemilihan sepeda sepeda motor sport lebih efisien dan praktis.Kata Kunci: Pemilihan, SPK, Sepeda motor sport, Metode WPM, Pematangsiantar