Wirarama Wedashwara
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Wirarama Wedashwara has a total of 54 co-authors. Here are the top co-authors by number of shared publications:
- Andy Hidayat Jatmika
Program Studi Teknik Informatika, Universitas Mataram
4 shared publications - Ariyan Zubaidi
University of Mataram, Indonesia
4 shared publications Ariyan Zubaidi
Program Studi Teknik Informatika, Fakultas Teknik, Universitas Mataram
3 shared publications- Candra Ahmadi
National Taiwan University of Science and Technology
3 shared publications - Heri Wijayanto
Universitas Mataram
3 shared publications - I Wayan Agus Arimbawa
Mataram University,Dept. Information System,Mataram,Indonesia
3 shared publications
Latest Publications
- Open Access
- 30 May 2026
- Published by Institute of Multidisciplinary Research and Community Service in Research in Education, Technology, and Multiculture
- Vol. 5 (1) , 1-14
CLV-Enhanced RFM Framework for Customer Segmentation in Indonesian SMEs Using K-Means Clustering
Micro, Small, and Medium Enterprises (MSMEs) contribute more than 61% of Indonesia's Gross Domestic Product, yet most of them still face limitations in leveraging transactional data for customer retention strategies. Prior studies have extensively combined Recency, Frequency, and Monetary (RFM) analysis with the K-Means algorithm for customer segmentation, but the majority treat Customer Lifetime Value (CLV) only as a post-hoc label assigned to clusters after the clustering process is finalized, rather than as a feature that shapes the segment structure from the beginning. This study addresses three research questions: how CLV can be effectively integrated as a clustering input, what segmentation structure emerges from this approach, and what concrete retention strategies can be derived for MSMEs with limited analytical capabilities. The proposed framework incorporates CLV, calculated as the discounted historical net sales using a 10% annual discount rate, as the fourth feature in the K-Means feature space with K-Means++ initialization, alongside RFM variables standardized using Z-scores. The framework is applied to a real dataset from a coffee shop MSME in Indonesia, comprising 19,126 transactions, of which 4,310 are member transactions from 472 unique registered customers, recorded throughout January–December 2023. The optimal number of clusters is determined through the convergence of the Elbow Method and the Silhouette Coefficient, both indicating four clusters as the best solution with a silhouette score of 0.5105. The segmentation divided customers into four tiers: Platinum, Gold, Silver, and Bronze. A key finding was a highly concentrated value distribution, with just 1.48% of customers (n = 7) contributing 21.66% of total revenue and CLV. This pattern is significantly more skewed than the traditional 80/20 Pareto rule. This concentration is interpreted through the lenses of habit formation, small base amplification, and comparative empirical evidence. The four-tier framework translates into differentiated retention strategies: VIP retention, value uplift, repeat-purchase incentives, and win-back campaigns, with monetary thresholds calibrated to each segment's median value. Keywords: K-Means, Clustering, RFM, Customer Lifetime Value, Customer Segmentation, MSMEs.
- 30 May 2026
- Published by Institute of Multidisciplinary Research and Community Service in Research in Education, Technology, and Multiculture
- Vol. 5 (1) , 1-14
Open AccessCLV-Enhanced RFM Framework for Customer Segmentation in Indonesian SMEs Using K-Means Clustering
Micro, Small, and Medium Enterprises (MSMEs) contribute more than 61% of Indonesia's Gross Domestic Product, yet most of them still face limitations in leveraging transactional data for customer retention strategies. Prior studies have extensively combined Recency, Frequency, and Monetary (RFM) analysis with the K-Means algorithm for customer segmentation, but the majority treat Customer Lifetime Value (CLV) only as a post-hoc label assigned to clusters after the clustering process is finalized, rather than as a feature that shapes the segment structure from the beginning. This study addresses three research questions: how CLV can be effectively integrated as a clustering input, what segmentation structure emerges from this approach, and what concrete retention strategies can be derived for MSMEs with limited analytical capabilities. The proposed framework incorporates CLV, calculated as the discounted historical net sales using a 10% annual discount rate, as the fourth feature in the K-Means feature space with K-Means++ initialization, alongside RFM variables standardized using Z-scores. The framework is applied to a real dataset from a coffee shop MSME in Indonesia, comprising 19,126 transactions, of which 4,310 are member transactions from 472 unique registered customers, recorded throughout January–December 2023. The optimal number of clusters is determined through the convergence of the Elbow Method and the Silhouette Coefficient, both indicating four clusters as the best solution with a silhouette score of 0.5105. The segmentation divided customers into four tiers: Platinum, Gold, Silver, and Bronze. A key finding was a highly concentrated value distribution, with just 1.48% of customers (n = 7) contributing 21.66% of total revenue and CLV. This pattern is significantly more skewed than the traditional 80/20 Pareto rule. This concentration is interpreted through the lenses of habit formation, small base amplification, and comparative empirical evidence. The four-tier framework translates into differentiated retention strategies: VIP retention, value uplift, repeat-purchase incentives, and win-back campaigns, with monetary thresholds calibrated to each segment's median value. Keywords: K-Means, Clustering, RFM, Customer Lifetime Value, Customer Segmentation, MSMEs.
- Open Access
- 12 June 2025
- Published by Universitas Pendidikan Nasional in TIERS Information Technology Journal
- Vol. 6 (1) , 112-128
Optimizing Neural Network Weights and Biases Using Particle Swarm Optimization for Classification Task
The current digital era greatly demands reliable automatic classification systems, especially to handle large and increasingly complex data volumes. One attractive alternative is the Particle Swarm Optimization (PSO) algorithm, which is recognized for its effective global search. Nevertheless, the performance of PSO for training artificial neural networks with complex or large-scale data is still uncertain. The primary purpose of this research is to create and assess a classification engine based on MLP, which uses a PSO algorithm to generate weights and biases. The assessment was made on three different types of data - dummy data, Iris data, and Sasak script images. For the dummy and Iris datasets, the model successfully achieved 100% accuracy, demonstrating the effectiveness of the PSO-MLP approach on simpler data. However, the results differed significantly for the more extensive and complex image dataset, where the model experienced a drastic decline in performance. In the image classification test with 6 classes, the model with one hidden layer achieved 71% accuracy, while the model with two hidden layers only reached 56%. For 12-class classification, accuracy dropped to 35% and 25%, respectively, and for 18 classes, the model achieved only 27% and 7%. These results indicate that while PSO is effective in optimizing perceptron weights and biases for smaller and simpler datasets, its ability to handle large-scale image classification with increasing complexity remains limited. Therefore, there is a need for optimization strategies to enhance the accuracy of optimization for more complex data.
- 12 June 2025
- Published by Universitas Pendidikan Nasional in TIERS Information Technology Journal
- Vol. 6 (1) , 112-128
Open AccessOptimizing Neural Network Weights and Biases Using Particle Swarm Optimization for Classification Task
The current digital era greatly demands reliable automatic classification systems, especially to handle large and increasingly complex data volumes. One attractive alternative is the Particle Swarm Optimization (PSO) algorithm, which is recognized for its effective global search. Nevertheless, the performance of PSO for training artificial neural networks with complex or large-scale data is still uncertain. The primary purpose of this research is to create and assess a classification engine based on MLP, which uses a PSO algorithm to generate weights and biases. The assessment was made on three different types of data - dummy data, Iris data, and Sasak script images. For the dummy and Iris datasets, the model successfully achieved 100% accuracy, demonstrating the effectiveness of the PSO-MLP approach on simpler data. However, the results differed significantly for the more extensive and complex image dataset, where the model experienced a drastic decline in performance. In the image classification test with 6 classes, the model with one hidden layer achieved 71% accuracy, while the model with two hidden layers only reached 56%. For 12-class classification, accuracy dropped to 35% and 25%, respectively, and for 18 classes, the model achieved only 27% and 7%. These results indicate that while PSO is effective in optimizing perceptron weights and biases for smaller and simpler datasets, its ability to handle large-scale image classification with increasing complexity remains limited. Therefore, there is a need for optimization strategies to enhance the accuracy of optimization for more complex data.
- Open Access
- 31 March 2025
- Published by Universitas Mataram in Jurnal Begawe Teknologi Informasi (JBegaTI)
- Vol. 6 (1)
Front End Design of Inventory Website at West Nusa Tenggara Statistics Agency using Bootstrap Framework
Pengelolaan inventaris barang yang efisien dan transparan sangat penting bagi Badan Pusat Statistik (BPS) Nusa Tenggara Barat (NTB), namun masih banyak dilakukan secara manual, sehingga rentan terhadap kesalahan dan ketidakefisienan. Untuk mengatasi hal ini, dikembangkan website inventori berbasis framework Bootstrap dengan fokus pada perancangan antarmuka (front-end) guna mempermudah akses dan pengelolaan data tanpa mengubah sistem yang sudah berjalan. Metode Scrum digunakan untuk memastikan pengembangan yang iteratif dan responsif terhadap kebutuhan pengguna, sehingga menghasilkan antarmuka yang lebih efektif dan mudah digunakan. Pengujian menggunakan System Usability Scale (SUS) menunjukkan skor 88, yang tergolong dalam kategori "excellent", menandakan bahwa sistem ini mampu meningkatkan efisiensi dan transparansi dalam pengelolaan inventaris di BPS NTB.
- 31 March 2025
- Published by Universitas Mataram in Jurnal Begawe Teknologi Informasi (JBegaTI)
- Vol. 6 (1)
Open AccessFront End Design of Inventory Website at West Nusa Tenggara Statistics Agency using Bootstrap Framework
Pengelolaan inventaris barang yang efisien dan transparan sangat penting bagi Badan Pusat Statistik (BPS) Nusa Tenggara Barat (NTB), namun masih banyak dilakukan secara manual, sehingga rentan terhadap kesalahan dan ketidakefisienan. Untuk mengatasi hal ini, dikembangkan website inventori berbasis framework Bootstrap dengan fokus pada perancangan antarmuka (front-end) guna mempermudah akses dan pengelolaan data tanpa mengubah sistem yang sudah berjalan. Metode Scrum digunakan untuk memastikan pengembangan yang iteratif dan responsif terhadap kebutuhan pengguna, sehingga menghasilkan antarmuka yang lebih efektif dan mudah digunakan. Pengujian menggunakan System Usability Scale (SUS) menunjukkan skor 88, yang tergolong dalam kategori "excellent", menandakan bahwa sistem ini mampu meningkatkan efisiensi dan transparansi dalam pengelolaan inventaris di BPS NTB.
- Open Access
- 30 January 2025
- Published by Universitas Mataram in Prosiding SAINTEK
- Vol. 7, 174-180
FUZZY ASSOCIATION RULE MINING BERBASIS PIVOT TABLE UNTUK PENGENALAN POLA DATA INTERNET OF THING
Perangkat Internet of Things (IoT) memiliki data yang bervariasi berdasarkan sensor yang digunakan. Data IoT juga memiliki variasi berdasarkan kondisi lingkungan dan waktu pengambilan data. Fuzzy Association Rule Mining (FARM) adalah implementasi basis data fuzzy pada association rule mining (ARM). Setiap term pada FARM terdiri dari fuzzy membership function bukan diskrit. Sehingga dapat mendiskripsikan kondisi data dengan lebih dinamis. Penelitian mengusulkan implementasi fungsi pivot table pada library pandas Bahasa pemrograman python untuk pembuatan fuzzy rule pada FARM. Pivot table memungkinkan ekstraksi pola fitur data yang unik dan menampilkan jumlahnya secara diskrit. Penelitian dilakukan menggunakan data IoT untuk irigasi tetes bertenaga surya. Fitur dari data IoT mencakup suhu dan kelembaban udara, kelembaban tanah, suhu tanah, intensitas Cahaya, arus dan tegangan pada aki dan panel surya. Evaluasi penelitian menghasilkan 156 rule dari 16000 data. Support tertinggi adalah 0.357 dan support rata-rata adalah 1.256. Sedangkan nilai confidence tertinggi adalah 0.883 dan rata-rata 0.673. Inferensi derajat keanggotan rule tertinggi adalah 0.673 dan rata-rata 0.102. Penggunaan pivot table menyederhanakan proses ekstraksi rule.
- 30 January 2025
- Published by Universitas Mataram in Prosiding SAINTEK
- Vol. 7, 174-180
Open AccessFUZZY ASSOCIATION RULE MINING BERBASIS PIVOT TABLE UNTUK PENGENALAN POLA DATA INTERNET OF THING
Perangkat Internet of Things (IoT) memiliki data yang bervariasi berdasarkan sensor yang digunakan. Data IoT juga memiliki variasi berdasarkan kondisi lingkungan dan waktu pengambilan data. Fuzzy Association Rule Mining (FARM) adalah implementasi basis data fuzzy pada association rule mining (ARM). Setiap term pada FARM terdiri dari fuzzy membership function bukan diskrit. Sehingga dapat mendiskripsikan kondisi data dengan lebih dinamis. Penelitian mengusulkan implementasi fungsi pivot table pada library pandas Bahasa pemrograman python untuk pembuatan fuzzy rule pada FARM. Pivot table memungkinkan ekstraksi pola fitur data yang unik dan menampilkan jumlahnya secara diskrit. Penelitian dilakukan menggunakan data IoT untuk irigasi tetes bertenaga surya. Fitur dari data IoT mencakup suhu dan kelembaban udara, kelembaban tanah, suhu tanah, intensitas Cahaya, arus dan tegangan pada aki dan panel surya. Evaluasi penelitian menghasilkan 156 rule dari 16000 data. Support tertinggi adalah 0.357 dan support rata-rata adalah 1.256. Sedangkan nilai confidence tertinggi adalah 0.883 dan rata-rata 0.673. Inferensi derajat keanggotan rule tertinggi adalah 0.673 dan rata-rata 0.102. Penggunaan pivot table menyederhanakan proses ekstraksi rule.
- Open Access
- 31 March 2024
- Published by Universitas Mataram in Jurnal Begawe Teknologi Informasi (JBegaTI)
- Vol. 5 (1) , 115-125
Implementasi Sistem Monitoring Jaringan Menggunakan Zabbix Berbasis SNMP pada UPT Pusat Teknologi Informasi dan Komputer (PUSTIK) Universitas Mataram
UPT Pusat Teknologi Informasi dan Komunikasi Universitas Mataram merupakan salah satu tempat/pusat untuk melakukan pemantauan terhadap semua jaringan yang berada di lingkungan Universitas Mataram. Administrator jaringan memerlukan pemantauan jaringan dalam mengelola jaringan. Kemajuan teknologi memungkinkan pemantauan jaringan menggunakan protokol SNMP yang digabungkan dengan sistem pemantauan menggunakan Zabbix. Tujuan penelitian ini adalah agar administrator jaringan dapat lebih cepat memecahkan masalah dengan memantau jaringan yang ada. Sistem Monitoring ini dapat menampilkan informasi dari perangkat jaringan seperti Data Lalu Lintas, Penggunaan Bandwidth, Temperature, dan Response Time. Hasil dari percobaan yang telah dilakukan telah terbukti mampu memantau jaringan menggunakan Zabbix berbasis SNMP dengan menampilkan hasil yang diharapkan.
- 31 March 2024
- Published by Universitas Mataram in Jurnal Begawe Teknologi Informasi (JBegaTI)
- Vol. 5 (1) , 115-125
Open AccessImplementasi Sistem Monitoring Jaringan Menggunakan Zabbix Berbasis SNMP pada UPT Pusat Teknologi Informasi dan Komputer (PUSTIK) Universitas Mataram
UPT Pusat Teknologi Informasi dan Komunikasi Universitas Mataram merupakan salah satu tempat/pusat untuk melakukan pemantauan terhadap semua jaringan yang berada di lingkungan Universitas Mataram. Administrator jaringan memerlukan pemantauan jaringan dalam mengelola jaringan. Kemajuan teknologi memungkinkan pemantauan jaringan menggunakan protokol SNMP yang digabungkan dengan sistem pemantauan menggunakan Zabbix. Tujuan penelitian ini adalah agar administrator jaringan dapat lebih cepat memecahkan masalah dengan memantau jaringan yang ada. Sistem Monitoring ini dapat menampilkan informasi dari perangkat jaringan seperti Data Lalu Lintas, Penggunaan Bandwidth, Temperature, dan Response Time. Hasil dari percobaan yang telah dilakukan telah terbukti mampu memantau jaringan menggunakan Zabbix berbasis SNMP dengan menampilkan hasil yang diharapkan.
- 1 January 2024
- Published by AIP Publishing in AIP Conference Proceedings
- Vol. 3026 (1)
3D mapping of fire hotspot in East Rinjani forest area using GIS and remote sensing
Wildfires in Indonesia reached their peak during the drought of 2015, with estimated burnt forest and land of up to 2.6 million hectares. During the dry season
- 1 January 2024
- Published by AIP Publishing in AIP Conference Proceedings
- Vol. 3026 (1)
3D mapping of fire hotspot in East Rinjani forest area using GIS and remote sensing
Wildfires in Indonesia reached their peak during the drought of 2015, with estimated burnt forest and land of up to 2.6 million hectares. During the dry season
- 1 January 2024
- Published by AIP Publishing in AIP Conference Proceedings
- Vol. 3026 (1) , 050027
Data storage and modeling system for GPS, Gyro and camera data using apache flume and hadoop map reduce
To collect and store large amounts of data from IoT devices requires an agent-based system such as Apache flume. Through Apache flume's agent-based data collect
- 1 January 2024
- Published by AIP Publishing in AIP Conference Proceedings
- Vol. 3026 (1) , 050027
Data storage and modeling system for GPS, Gyro and camera data using apache flume and hadoop map reduce
To collect and store large amounts of data from IoT devices requires an agent-based system such as Apache flume. Through Apache flume's agent-based data collect
- Open Access
- 31 December 2023
- Published by Universitas Mataram in Journal of Computer Science and Informatics Engineering (J-Cosine)
- Vol. 7 (2)
Sistem Monitoring Kualitas Udara di Ruangan Tertutup Berbasis IoT Menggunakan Sensor MQ-135 dan GP2Y1014AU0F
Air quality is now decreasing, mainly in urban areas and industrial centers. Decreasing air quality can occur indoors, especially in a closed room such as a kitchen room. Conditions like this will certainly be very detrimental because it can interfere with health, disturbing comfort and can damage property. In this study, a system was made to monitor air quality, especially in closed rooms by utilizing IoT technology using MQ-135 and GP2Y1014AU0F sensors. Based on tests conducted in the kitchen room, when cooking the gas status was in the medium category and the dust status was in the good category, whereas after cooking the gas and dust status were in the good category. Then in the warehouse room, the gas status is in the medium category and the dust status is in the unhealthy category. Based on the test results, the system built has been running well and is possible to be implemented.
- 31 December 2023
- Published by Universitas Mataram in Journal of Computer Science and Informatics Engineering (J-Cosine)
- Vol. 7 (2)
Open AccessSistem Monitoring Kualitas Udara di Ruangan Tertutup Berbasis IoT Menggunakan Sensor MQ-135 dan GP2Y1014AU0F
Air quality is now decreasing, mainly in urban areas and industrial centers. Decreasing air quality can occur indoors, especially in a closed room such as a kitchen room. Conditions like this will certainly be very detrimental because it can interfere with health, disturbing comfort and can damage property. In this study, a system was made to monitor air quality, especially in closed rooms by utilizing IoT technology using MQ-135 and GP2Y1014AU0F sensors. Based on tests conducted in the kitchen room, when cooking the gas status was in the medium category and the dust status was in the good category, whereas after cooking the gas and dust status were in the good category. Then in the warehouse room, the gas status is in the medium category and the dust status is in the unhealthy category. Based on the test results, the system built has been running well and is possible to be implemented.
- 2 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 1, 1-6
Hybrid Data Collection System for IoT Based Solar Powered Drip Irrigation system using LoRa WAN
Outdoor data communication using the internet in Internet of Things (IoT) applications has a risk of interference caused by environmental conditions. Providing an access point for outdoor applications requires more electricity. The proposed research aimed to optimize data communication between local storage, internet and wireless sensor network (WSN). A hybrid data collection system for an IoT-based solar drip irrigation system using LoRa WAN. The Hybrid system in this study consists of a WAN for local control and data collection using LoRa and uploading to the internet. Apart from storing data in indoor and internet local storage, the system allows local users such as farmers to access a web-based user interface without internet. The WSN system in research uses nRF from previous studies. The blackbox testing shows valid results for nRF, LoRa and HTTP GET data communication series to local and online web server interfaces. The results of a data communication loop programming produces a total average time of 2469ms and a delay of 3000ms on outdoor devices. The indoor device uses 24272ms and a delay of 5000ms for one loop. The Quality of Service (QoS) evaluation showed that the lowest PDR archived by nRF with 97.285%. In other side the longest delay is archived by LoRa with 936ms. Through solar power testing, the overall QoS decreases when the light intensity is below 101 lux, and the solar panel voltage is below 0.121v. But the decrease doesn't seem significant, especially in LoRa with the lowest value of 98.378% and 934ms delay. Meanwhile, the lowest delay was achieved by nRF with 247.6ms. The lowest PDR was archieved by HTTP GET with 96.7638%. Meanwhile, the longest delay was archieved by LoRa 876.6ms.
- 2 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 1, 1-6
Hybrid Data Collection System for IoT Based Solar Powered Drip Irrigation system using LoRa WAN
Outdoor data communication using the internet in Internet of Things (IoT) applications has a risk of interference caused by environmental conditions. Providing an access point for outdoor applications requires more electricity. The proposed research aimed to optimize data communication between local storage, internet and wireless sensor network (WSN). A hybrid data collection system for an IoT-based solar drip irrigation system using LoRa WAN. The Hybrid system in this study consists of a WAN for local control and data collection using LoRa and uploading to the internet. Apart from storing data in indoor and internet local storage, the system allows local users such as farmers to access a web-based user interface without internet. The WSN system in research uses nRF from previous studies. The blackbox testing shows valid results for nRF, LoRa and HTTP GET data communication series to local and online web server interfaces. The results of a data communication loop programming produces a total average time of 2469ms and a delay of 3000ms on outdoor devices. The indoor device uses 24272ms and a delay of 5000ms for one loop. The Quality of Service (QoS) evaluation showed that the lowest PDR archived by nRF with 97.285%. In other side the longest delay is archived by LoRa with 936ms. Through solar power testing, the overall QoS decreases when the light intensity is below 101 lux, and the solar panel voltage is below 0.121v. But the decrease doesn't seem significant, especially in LoRa with the lowest value of 98.378% and 934ms delay. Meanwhile, the lowest delay was achieved by nRF with 247.6ms. The lowest PDR was archieved by HTTP GET with 96.7638%. Meanwhile, the longest delay was archieved by LoRa 876.6ms.
- Open Access
- 24 December 2022
- Published by Atlantis Press SARL
- p. 142-155
IoT Based Water Turbinity Classification Using Color Sensor TCS3200
- 24 December 2022
- Published by Atlantis Press SARL
- p. 142-155
Open AccessIoT Based Water Turbinity Classification Using Color Sensor TCS3200