Garuda - Garba Rujukan Digital

PENGEMBANGAN SISTEM PERHITUNGAN JUMLAH KENDARAAN BERDASARKAN JENIS KENDARAAN MENGGUNAKAN ALGORITMA YOLO SECARA REALTIME Kusuma, Adri Surya; Pradana, Afu Ichsan; Pamekas, Bondan Wahyu
SKANIKA: Sistem Komputer dan Teknik Informatika Vol 7 No 2 (2024): Jurnal SKANIKA Juli 2024
Publisher : Universitas Budi Luhur

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36080/skanika.v7i2.3201

This study developed a vehicle counting system based on vehicle types using the YOLO algorithm in real-time. With the advancement of artificial intelligence technology, particularly in computer vision, outdoor media has also advanced through the computation of CCTV image results installed on such media. The significance of this research is found in its ability to provide accurate and real-time traffic data that can be utilized by local governments, traffic management companies, property developers, and advertising companies for traffic planning, security, and marketing analysis. The research method involved training the YOLOv9 model and centroid-based tracking on a dataset that included four vehicle classes: car, motorcycle, bus, and truck. The results showed that the developed system could detect and track vehicles with high accuracy, achieving a mAP of 0.9, precision of 0.909, recall of 0.828, and an F1-Score of 0.867. However, the system's performance heavily depends on the hardware specifications used, and the detection for motorcycles has lower evaluation scores compared to other vehicle types. This study indicates that improving hardware specifications and optimizing the model can enhance system performance. These findings are important as they reveal the considerable potential of using the YOLO algorithm in real-time traffic monitoring applications.

Komparasi Metode Decission Tree, Logistic Regression, SVM, dan ANN Dalam Klasifikasi Kualitas Air Hartanti, Dwi; Pradana, Afu Ichsan
SMARTICS Journal Vol 9 No 1 (2023): SMARTICS Journal (April 2023)
Publisher : Universitas PGRI Kanjuruhan Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21067/smartics.v9i1.8113

Water is the most important part of human life because it is the source of human life and about 71% of the earth's area is water. Every human being has the human right to clean water which is a basis for the realization of a decent and dignified life for humans. Classification is one of the techniques in data mining. This study uses water quality data using four algorithmic methods, namely Decission Tree, Logistic Regression, SVM, and ANN. The aim of this research is to compare which method has the maximum accuracy value for water quality classification. The accuracy results obtained are the Decission Tree method of 60.19%, the Logistic Regression method of 62.80%, the SVM method of 68.59%, and the ANN method of 69.54%.

Teknik K-Fold Cross Validation untuk Mengevaluasi Kinerja Mahasiswa Wijiyanto, Wijiyanto; Pradana, Afu Ichsan; Sopingi, Sopingi; Atina, Vihi
Jurnal Algoritma Vol 21 No 1 (2024): Jurnal Algoritma
Publisher : Institut Teknologi Garut

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33364/algoritma/v.21-1.1618

A student's ability to complete courses is influenced by various factors, including academic and non-academic aspects. Understanding the factors that influence it is very important to know in order to anticipate and prevent the possibility of failure in the study. It turns out that non-academic factors also have a big influence on student success, especially family factors, such as the level of education obtained by parents, the employment status of parents and the income of both parents. To be able to understand these factors, studies are needed to predict student performance based on family background factors using machine learning models, support vector machine algorithms, naÃ¯ve Bayes, neural networks and decision trees. The data used was 365 records and 11 attributes, separated by 70% for train data and 30% for test data, which was then used by kfold cross validation to evaluate the results using the parameters n_split=10 and random_state=42. In the confusion matrix parameters, the average (mean) accuracy value for the support vector machine model was 87.68%, naÃ¯ve Bayes was 90.97%, neural network was 87.95% and decision tree was 85.75%. Meanwhile, the best fold result for SVM is located at the 10th fold with an accuracy of 94.44%, for NB it is located at the 4th fold with an accuracy value of 97.29%, for NN it is located at the 4th fold with an accuracy value of 94.59% and for DT is located on the 5th fold with an accuracy value of 91.89%. Thus, evaluation using k-fold cross validation can be used to predict student performance based on family attributes using the 4th fold which has the highest accuracy of 97.29% in the naÃ¯ve Bayes model algorithm in order to graduate on time.

PERBANDINGAN DATA UNTUK MEMPREDIKSI KETEPATAN STUDI BERDASARKAN ATRIBUT KELUARGA MENGGUNAKAN MACHINE LEARNING Wijiyanto, Wijiyanto; Pradana, Afu Ichsan; Sopingi, Sopingi
Jurnal Informatika Vol 8, No 2 (2024): JIKA (Jurnal Informatika)
Publisher : University of Muhammadiyah Tangerang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31000/jika.v8i2.10752

Keberhasilan mahasiswa dalam menyelesaikan pendidikan tepat waktu merupakan tujuan yang penting. Berbagai faktor dapat memengaruhi keberhasilan ini, termasuk faktor non-akademik seperti data keluarga. Data yang digunakan berasal dari FIKOM-UDB dengan 365 record dan 11 atribut, di antaranya satu atribut berperan sebagai label (class). Data tersebut diproses menggunakan algoritma machine learning menggunakan pemodelan naïve bayes dan neural network. Sebelumnya, data dibagi menjadi data latih dan data uji dengan perbandingan prosentase yang berbeda, yaitu 90:10, 80:20, 70:30, 60:40, dan 50:50, untuk mencari kinerja terbaik berdasarkan nilai akurasi. Evaluasi menggunakan confusion matrix menghasilkan performa terbaik untuk naïve bayes dengan perbandingan 80:20, mencapai nilai akurasi sebesar 92%, precision 0.93, recall 0.98, dan F1-score 0.96. Sementara untuk neural network, performa terbaik terdapat pada perbandingan 50:50 dengan nilai akurasi sebesar 91%, precision 0.93, recall 0.97, dan F1-score 0.95. Hasil menunjukkan bahwa performa terendah untuk naïve bayes terjadi pada perbandingan 90:10, sementara untuk neural network terjadi pada perbandingan 80:20. Dengan demikian, algoritma naïve bayes menunjukkan performa yang lebih baik dibandingkan neural network sehingga, Fakultas dapat menerapkan model naïve bayes dalam memprediksi mahasiswa dalam rangka untuk mengantisipasi dan mengatasi permasalahan yang timbul terkait kelulusan mahasiswa dengan tepat waktu.

PENGEMBANGAN SISTEM KEAMANAN KOTAK AMAL DI MASJID RAYA ASSALAM MASARAN BERBASIS IOT (INTERNET OF THINGS) Aji, Sindhu Purnomo; Pradana, Afu Ichsan; Maulindar, Joni
Infotech: Journal of Technology Information Vol 10, No 1 (2024): JUNI
Publisher : ISTEK WIDURI

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37365/jti.v10i1.261

A mosque is a building used by Muslims to carry out worship and religious activities. Masjid Raya Assalam Masaran is one of the mosques located in Karangtengah Hamlet, Masaran District, Sragen Regency, Central Java. However, a few months ago there was a case of theft of charity boxes in this mosque, so the security of charity boxes is an important concern for mosque administrators. Therefore, the author developed a charity box security system to detect and anticipate theft. The system developed using the Prototype method is able to detect indications of theft, provide notifications via the Telegram application, track the whereabouts of the charity box using GPS technology, and lock the charity box using a solenoid door lock with a fingerprint sensor. The test results show that this system has excellent accuracy, response speed, and consistency, so it is expected to provide a sense of security and trust to worshipers in donating at the mosque charity box.ABSTRAKMasjid merupakan bangunan yang digunakan umat Muslim untuk melaksanakan ibadah dan kegiatan keagamaan. Masjid Raya Assalam Masaran merupakan salah satu masjid yang terletak di Dusun Karangtengah, Kec. Masaran, Kabupaten Sragen, Jawa Tengah. Namun, beberapa bulan yang lalu terdapat kasus pencurian kotak amal di masjid ini, sehingga keamanan kotak amal menjadi perhatian penting bagi pengurus masjid. Oleh karena itu, penulis mengembangkan sebuah sistem keamanan kotak amal untuk mendeteksi dan mengantisipasi adanya tindakan pencurian. Sistem yang dikembangkan menggunakan metode Prototype mampu mendeteksi indikasi pencurian, memberikan notifikasi melalui aplikasi Telegram, melacak keberadaan kotak amal menggunakan teknologi GPS, serta mengunci kotak amal menggunakan solenoid door lock dengan sensor sidik jari. Hasil pengujian menunjukkan sistem ini memiliki ketepatan, kecepatan respons, dan konsistensi yang sangat baik, sehingga diharapkan dapat memberikan rasa aman dan kepercayaan kepada jamaah dalam menyumbang di kotak amal masjid.

PENGEMBANGAN APLIKASI MOBILE UNTUK MONITORING DETAK JANTUNG, SATURASI OKSIGEN DARAH, DAN SUHU TUBUH DENGAN INTEGRASI IOT MENGGUNAKAN ESP32 Pratama, Fandi Aziz; Pradana, Afu Ichsan; Hartanti, Dwi
Infotech: Journal of Technology Information Vol 10, No 1 (2024): JUNI
Publisher : STMIK WIDURI

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37365/jti.v10i1.244

Healthy living is the desire of every human being. Maintaining health involves disease prevention as well as ongoing physical and mental care. Health parameters such as heart rate, blood oxygen saturation, and body temperature are important indicators of a person's physical and mental condition. A normal heart rate and oxygen levels are essential for efficient heart function, while body temperature can indicate a variety of health conditions. Currently, health checks are usually carried out by visiting health facilities such as hospitals, health centers, and so on, which are considered less effective because they take a lot of time. Although health checking tools such as pulse oximeters and digital thermometers are currently available, these tools can only display health information when used and the measurement data cannot be saved to a database for long-term monitoring. So, a health monitoring tool is needed that can be used personally to monitor the user's health periodically. In this research the author developed a mobile health monitoring application using the Dart programming language and the Flutter framework. This application is integrated with the MAX30102 oximeter sensor and MLX90614 temperature sensor via the Esp32 microcontroller with Bluetooth Low Energy (BLE) communication protocol. This system allows users to carry out personal health checks without being disturbed by daily routines, measurement data is displayed in real-time on the application and stored on the server for long-term monitoring. This research succeeded in developing a mobile IoT application with Esp32 to monitor heart rate, blood oxygen saturation and body temperature. Black box testing showed 100% success with measurement accuracy values of 95.55% for heart rate, 99.01% for oxygen saturation, and 99.269% for body temperature.ABSTRAKHidup sehat merupakan keinginan dari setiap manusia. Menjaga kesehatan melibatkan pencegahan penyakit serta perawatan fisik dan mental yang berkesinambungan. Parameter kesehatan seperti detak jantung, saturasi oksigen darah, dan suhu tubuh merupakan indikator penting kondisi fisik dan mental seseorang. Detak jantung dan kadar oksigen yang normal sangat penting untuk fungsi jantung yang efisien, sementara suhu tubuh dapat mengindikasikan berbagai kondisi kesehatan. Saat ini, pemeriksaan kesehatan biasanya dilakukan dengan mengunjungi fasilitas kesehatan seperti rumah sakit, puskesmas, dan sebagainya yang dinilai kurang efektif karena memakan banyak waktu. Meskipun saat ini tersedia alat pengecekan kesehatan seperti pulse oksimeter dan termometer digital, namun alat tersebut hanya dapat menampilkan informasi kesehatan ketika digunakan dan data hasil pengukuran tidak dapat disimpan ke database untuk dilakukan pemantauan jangka panjang. Maka, dibutuhkanlah alat pemantauan kesehatan yang dapat digunakan secara pribadi untuk memantau kesehatan pengguna secara berkala. Pada penelitian ini penulis mengembangkan aplikasi mobile monitoring kesehatan menggunakan bahasa pemrograman Dart dan framework Flutter. Aplikasi ini terintegrasi dengan sensor oksimeter MAX30102 dan sensor suhu MLX90614 melalui mikrokontroler Esp32 dengan protokol komunikasi Bluetooth Low Energy (BLE). Sistem ini memungkinkan pengguna melakukan pemeriksaan kesehatan secara pribadi tanpa terganggu rutinitas harian, data hasil pengukuran ditampilkan secara real-time pada aplikasi serta disimpan pada server untuk pemantauan jangka panjang. Penelitian ini berhasil mengembangkan aplikasi mobile IoT dengan Esp32 untuk memantau detak jantung, saturasi oksigen darah, dan suhu tubuh. Pada pengujian blackbox menunjukkan keberhasilan 100% dengan nilai akurasi pengukuran 95,55% untuk detak jantung, 99,01% untuk saturasi oksigen, dan 99,269% untuk suhu tubuh.

SISTEM MONITORING PENYIRAMAN TANAMAN TOMAT OTOMATIS BERBASIS IOT PADA PERKEBUNAN DI DESA SROYO MENGGUNAKAN APLIKASI BLYNK Putra, Enggar Wijaya; Pradana, Afu Ichsan; Hartanti, Dwi
Infotech: Journal of Technology Information Vol 10, No 1 (2024): JUNI
Publisher : ISTEK WIDURI

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37365/jti.v10i1.256

Humans can make various devices as tools in various kinds of work and production, as tools that facilitate daily human activities, such as tools that can be used to water plants automatically. Therefore, it is necessary to develop an automatic watering device to overcome the problem of manual watering. The tool is expected to work automatically as needed. Therefore, the use of this tool is expected to facilitate the process of planting plants and improve the overall quality of plant care. This research intends to build an automatic tomato plant watering system using the prototype method in this research because it is a method that describes a structured system and must go through several processes in its creation. In this research the author makes a tomato plant monitoring application using Blynk, this application is connected to the soil moisture sensor and soil pH sensor through an ESP32 microcontroller connected to the internet. This monitoring tool allows users to observe remotely and the information provided is displayed in real time on the Blynk application. The result of this research is intended to find a more effective method of remote plant survey.ABSTRAKManusia dapat menjadikan berbagai perangkat sebagai alat dalam berbagai macam pekerjaan dan produksi, sebagai alat yang memudahkan aktivitas manusia sehari-hari, seperti alat yang dapat digunakan untuk menyiram tanaman secara otomatis. Oleh karena itu, perlu dikembangkan suatu alat penyiraman otomatis untuk mengatasi permasalahan penyiraman manual. Alat tersebut diharapkan dapat bekerja secara otomatis sesuai kebutuhan. Oleh karena itu, penggunaan alat ini diharapkan dapat mempermudah proses penanaman tanaman dan meningkatkan kualitas perawatan tanaman secara keseluruhan. Penelitian ini bermaksud untuk membangun sistem penyiraman tanaman tomat otomatis dengan menggunakan metode prototype pada penelitian ini karena merupakan metode yang menggambarkan suatu sistem yang terstruktur dan harus melalui beberapa proses dalam pembuatannya. Pada penelitian ini penulis membuat sebuah aplikasi monitoring tanaman tomat menggunakan Blynk, aplikasi ini terhubung dengan sensor kelembaban tanah dan sensor pH tanah melalui mikrokontroler ESP32 yang terhubung dengan internet. Alat pemantauan ini memungkinkan pengguna untuk mengamati dari jarak jauh dan informasi yang diberikan ditampilkan secara real time di aplikasi Blynk. Untuk hasil pengujian menunjukkan rata-rata ketelitian pengukuran kelembaban tanah sebesar 97,28% dan rata-rata ketelitian pengukuran pH tanah sebesar 97,59%. Hasil dari penelitian ini dimaksudkan untuk menemukan metode survei tanaman jarak jauh yang lebih efektif.

IMPLEMENTASI SYSTEM TRACKING BARANG DENGAN GLOBAL POSITIONING SYSTEM (GPS) BERBASIS IOT MENGGUNAKAN ESP32 DI PUSKESMAS KLEGO II Rakadipa, Dhimas Arya; Pradana, Afu Ichsan; Widyaningsih, Pipin
Infotech: Journal of Technology Information Vol 10, No 1 (2024): JUNI
Publisher : ISTEK WIDURI

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37365/jti.v10i1.247

Puskesmas is to help provide quality health services, but at affordable costs for the community, especially the lower middle class economy. One of the puskesmas that was the research site was Klego II Health Center. Here they provide services in the form of borrowing equipment such as wheelchairs, oxygen, stretchers and so on, but here there is no security system in the lending process such as filling out a form or leaving an identity card, therefore the author created a tool in the form of a mini GPS tracking which functions as the tool used. to find out the tools lent by the community, so as to reduce the level of loss of goods due to borrowing goods. The system development method used by the author is a prototype method and from the results of testing the tool it can be concluded that this tool can work very well, where this tool can display latitude and longitude obtained from GPS neo 6m and can work indoors or outdoors, its use is It is quite easy and fast to get a signal, but this tool is also very sensitive to movement, where a lot of movement will affect the quality of the signal received. ABSTRAKPuskesmas adalah membantu pelayanan kesehatan bermutu, namun biaya terjangkau untuk Masyarakat, terutama ekonomi menengah kebawah salah satu puskesmas yang menjadi tempat penelitian adalah Puskesmas Klego II. Disini memberikan pelayanan berupa peminjaman alat seperti kursi roda, oksigen, tandu dan lain-lain, namun disini tidak adanya sistem keamanan pada proses peminjaman seperti mengisi form atau meninggalkan kartu identitas oleh sebab itu penulis membuat alat berupa gps tracker mini yang berfungsi sebagai alat yang digunakan untuk mengetahui alat yang dipinjam oleh Masyarakat,sehingga dapat mengurangi Tingkat kehilangan barang akibat peminjaman barang. Metode pengembangan sistem yang digunakan penulis adalah metode prototipe dan dari hasil pengujian alat dapat disimpulkan alat ini dapat bekerja sangat baik yang Dimana alat ini dapat menampilkan latitude dan longitude yang didapat dari GPS neo 6m serta dapat bekerja didalam atau diluar ruangan, penggunaan nya yang cukup mudah dan cepat mendapat sinyal akan tetapi alat ini juga sangat sensitive terhadap Gerakan yang Dimana banyak nya Gerakan akan mempengaruhi kualitas sinyal yang diterima.

Perancangan Sistem Monitoring Penyiraman Dan Pemupukan Pada Tanaman Hias Sri Rezeki Setiawan, Gilang; Maulindar, Joni; Pradana, Afu Ichsan
INTECOMS: Journal of Information Technology and Computer Science Vol 7 No 4 (2024): INTECOMS: Journal of Information Technology and Computer Science
Publisher : Institut Penelitian Matematika, Komputer, Keperawatan, Pendidikan dan Ekonomi (IPM2KPE)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31539/intecoms.v7i4.10622

Perubahan iklim yang terjadi akhir-akhir ini menyebabkan sulitnya memprediksi cuaca, sehingga budidaya tanaman hias sri rejeki menjadi kurang efektif ketika hujan tiba-tiba datang ataupun ketika panas datang. Untuk mengatasi masalah tersebut, penggunaan teknologi Internet of things (IoT) dapat memberikan kemudahan dalam mengelola tanaman hias sri rejeki. Beberapa penelitian telah dilakukan untuk mengembangkan system yang dapat melakukan control dan monitoring terhadap tanaman hias sri rejeki berbasis IoT dengan memanfaatkan sensor yang akan mengirimkan sinyal ke perangkat yang terhubung dengan jaringan internet untuk memberikan informasi tentang kondisi kelembaban tanah dan tingkat keasaman tanah secara tepat, akurat serta dapat melakukan informasi secara realtime. Penelitian ini menggunakan metode pengembangan untuk merancang sistem perancangan sistem monitoring penyiraman dan pemupukan pada tanaman hias sri rezeki dengan kontrol dan monitoring yang terhubung ke perangkat seluler. Sistem ini dilengkapi dengan sensor kelembaban tanah, dan sensor pH tanah yang akan memberikan informasi secara real-time pada aplikasi Blynk. Dengan sistem ini, pengguna dapat mengatur tanaman hias sri rejeki dari jarak jauh dan menghindari tanamaman hias yang mengakibatkan layu, sehingga menghemat waktu dan tenaga. Hasil perancangan yang dilakukan menunjukan sistem dapat dirancang dengan baik.

Teknik K-Fold Cross Validation untuk Mengevaluasi Kinerja Mahasiswa Wijiyanto, Wijiyanto; Pradana, Afu Ichsan; Sopingi, Sopingi; Atina, Vihi
Jurnal Algoritma Vol 21 No 1 (2024): Jurnal Algoritma
Publisher : Institut Teknologi Garut

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33364/algoritma/v.21-1.1618

A student's ability to complete courses is influenced by various factors, including academic and non-academic aspects. Understanding the factors that influence it is very important to know in order to anticipate and prevent the possibility of failure in the study. It turns out that non-academic factors also have a big influence on student success, especially family factors, such as the level of education obtained by parents, the employment status of parents and the income of both parents. To be able to understand these factors, studies are needed to predict student performance based on family background factors using machine learning models, support vector machine algorithms, naÃ¯ve Bayes, neural networks and decision trees. The data used was 365 records and 11 attributes, separated by 70% for train data and 30% for test data, which was then used by kfold cross validation to evaluate the results using the parameters n_split=10 and random_state=42. In the confusion matrix parameters, the average (mean) accuracy value for the support vector machine model was 87.68%, naÃ¯ve Bayes was 90.97%, neural network was 87.95% and decision tree was 85.75%. Meanwhile, the best fold result for SVM is located at the 10th fold with an accuracy of 94.44%, for NB it is located at the 4th fold with an accuracy value of 97.29%, for NN it is located at the 4th fold with an accuracy value of 94.59% and for DT is located on the 5th fold with an accuracy value of 91.89%. Thus, evaluation using k-fold cross validation can be used to predict student performance based on family attributes using the 4th fold which has the highest accuracy of 97.29% in the naÃ¯ve Bayes model algorithm in order to graduate on time.

Title

Found 24 Documents
Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title Search

Found 24 Documents Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title

Found 24 Documents
Search