Claim Missing Document
Check
Articles

Found 5 Documents
Search

Low-Code Platform for Health Protocols Implementation in Sabilussalam Mosque During The COVID-19 Pandemic Sofia Umaroh; Kurnia Ramadhan Putra; nur Fitrianti; Mira Musrini Barmawi
REKA ELKOMIKA: Jurnal Pengabdian kepada Masyarakat Vol 3, No 2 (2022): REKA ELKOMIKA
Publisher : Institut Teknologi Nasional, Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26760/rekaelkomika.v3i2.96-105

Abstract

The COVID-19 pandemic condition has changed various aspects of life, especially Muslims’ prayer activities. COVID-19 pandemic has especially affected mosques as a place for Muslims to pray five times a day and to have Friday prayer congregation. To prevent the spread of COVID-19, the Sabilussalam Mosque located at Jl. Dr. Hatta Bandung has restricted its capacity and ensured visitors’ body temperature below 37,4 C manually. However, physical contact still occured, and its capacity still exceeded 50%. Therefore, the adoption of self-check temperature and automatic capacity counter as a method of mitigating the COVID-19 pandemic in mosques was needed. This community service aims at implementing a system for controlling mosque capacity and for avoiding physical contact during praying. The Low Code-based app counts temperature data broadcasted by K3 Pro and stored in NowDB cloud-service. As a result, the system manages to control the mosque's capacity to a maximum of 50% without any physical contact because it relies on internet connection.
PERBANDINGAN METODE PERHITUNGAN JARAK EUCLIDEAN, HAVERSINE, DAN MANHATTAN DALAM PENENTUAN POSISI KARYAWAN Yusup Miftahuddin; Sofia Umaroh; Fahmi Rabiul Karim
Jurnal Tekno Insentif Vol 14 No 2 (2020): Jurnal Tekno Insentif
Publisher : Lembaga Layanan Pendidikan Tinggi Wilayah IV

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36787/jti.v14i2.270

Abstract

Abstrak - Kinerja karyawan merupakan hal yang diperhatikan di dalam instansi. Institut Teknologi Nasional Bandung merupakan salah satu instansi dengan jumlah karyawan yang banyak, sehingga sulit dilakukan pemantauan keberadaan seluruh karyawan. Salah satu alternatif dalam mengatasi masalah tersebut adalah pembuatan sistem untuk memantau lokasi keberadaan karyawan dengan memanfaatkan smartphone untuk pengambilan titik koordinat. Pada era modern ini smartphone merupakan barang yang hampir tidak pernah ditinggalkan. Dengan memanfaatkan titik koordinat, perhitungan jarak dapat dihitung dengan menggunakan 3 metode yaitu euclidean, manhattan, dan haversine. Dari pengujian yang telah dilakukan, rata-rata waktu yang diperlukan untuk proses pengiriman koordinat dari smartphone ke database sistem adalah 0,9 detik. Selain itu, penelitian ini bertujuan untuk membandingkan ketiga metode berdasarkan keakurasian dan waktu. Perbandingan tingkat keakurasian dilakukan dengan membandingkan persentase error hasil perhitungan jarak dengan pengukuran secara manual menggunakan pita ukur. Hasil akhir dari pengujian tiga metode tersebut diperoleh bahwa metode perhitungan Manhattan membutuhkan waktu pengolahan data yang paling cepat dalam pengujian 100 data yaitu 0,00034045 detik. Metode perhitungan Haversine menghasilkan akurasi perhitungan jarak teringgi yaitu 98,66%. Dan metode perhitungan Haversine menghasilkan akurasi keputusan tertinggi dalam menentukan keputusan lokasi keberadaan karyawan yaitu 90%. Hasil penelitian ini dapat digunakan sebagai pertimbangan pemilihan metode perhitungan jarak bagi para peneliti. Abstract - Employee performance is a matter of concern within the agency. The Bandung National Institute of Technology is one institution with a large number of employees, making it difficult to monitor the whereabouts of all employees. One alternative in overcoming the problem is the creation of a system to monitor the location of employees by utilizing smartphones to capture coordinates. In this modern era smartphone is an item that is almost never left. By utilizing coordinate points, distance calculation can be calculated using 3 methods namely euclidean, manhattan, and haversine. From the tests that have been done, the average time required for the sending of coordinates from the smartphone to the system database is 0.9 seconds. In addition, this study aims to compare the three methods based on accuracy and time. Comparison of the level of accuracy is done by comparing the percentage of error calculation results with the distance measurement manually using a measuring tape. The final results of the three methods test was obtained that the Manhattan calculation method requires the fastest data processing time in testing 100 data that is 0,00034045 seconds. The Haversine calculation method produces the highest distance calculation accuracy which is 98.66%. And the Haversine calculation method produces the highest decision accuracy in determining the location of the employee's decision that is 90%. The results of this study can be used as consideration for the selection of distance calculation methods for researchers.
Identifikasi Jenis Font Menggunakan Metode Genetic Modified K-Nearest Neighbor Yusup Miftahuddin; Sofia Umaroh; Agistya Anugrah Dwiutama
Rekayasa Hijau : Jurnal Teknologi Ramah Lingkungan Vol 4, No 3 (2020)
Publisher : Institut Teknologi Nasional, Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26760/jrh.v4i3.157-166

Abstract

ABSTRAKFont adalah kumpulan karakter lengkap yang memiliki ukuran dan gaya. Saat melihat desain atau aplikasi, desainer grafis dan pengembang front-end terinspirasi untuk menggunakan font yang sama. Namun font telah menjadi gambar atau aplikasi dan sulit untuk mengetahui jenis font yang digunakan. Genetic Modified K-Nearest Neighbor (GMKNN) merupakan metode gabungan dari Modified K-Nearest Neighbor (MKNN) dan Genetic Algorithm (GA) untuk menentukan k Optimal. Dalam penelitian ini, sebuah sistem akan dirancang untuk mengenali jenis font sans-serif menggunakan metode Genetic Modified K-Nearest Neighbor (GMKNN) untuk mengukur akurasi dan waktu komputasi. Kinerja sistem dalam proses mengidentifikasi jenis font mendapat nilai presisi rata-rata 74,6%. Nilai akurasi adalah 72,2% dan nilai recall 72%. Dari hasil presisi dan recall yang diperoleh nilai f-measure sebesar 72,2% dan rata-rata waktu komputasi untuk satu karakter diperoleh adalah 945,04190395673 detikKata kunci: Pengolahan Citra Digital, Identifikasi Font, Pengenalan Pola dan GMKNNABSTRACTFonts is a complete collection of characters that have size and style. When looking at a design or an application, graphic designers and front-end developers are inspired to use the same font. Unfortunately, the font has become an image or an application so it is difficult to identify the font types. Genetic Modified K-Nearest Neighbor (GMKNN) is a hybrid method of Modified K-Nearest Neighbor (MKNN) and Genetic Algorithm (GA) to determine optimal k, it also reduces the complexity of MKNN and improves the classification accuracy. In this research, a system is designed to recognize font sans-serif types using GMKNN method to measure accuracy and time computation. The performance of the system in the process of identifying font types gets an average precision value of 74.6%. The recall and accuracy values are 72% and 72,2%, respectively. Based on the results of precision and recall obtained, the system gets an f-measured value of 72.2% and time obtained for one character is 945,04190395673 seconds.Keywords: Image Proccesing, Font Identify, Pattern Recognition, and GMKNN
RESULTANT: Data Preparation Techniques to Improve XGBoost Algorithm Performance KURNIA RAMADHAN PUTRA; SOFIA UMAROH; NUR FITRIANTI; SATRIA NUGRAHA
MIND (Multimedia Artificial Intelligent Networking Database) Journal Vol 8, No 1 (2023): MIND Journal
Publisher : Institut Teknologi Nasional, Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26760/mindjournal.v8i1.42-51

Abstract

ABSTRAKPrediksi credit scoring saat ini banyak digunakan dalam layanan peer-to-peer lending oleh perusahaan teknologi finansial. Salah satu teknologi yang digunakan untuk credit scoring adalah data mining menggunakan algoritma machine learning XGBoost yang memiliki tingkat akurasi yang tinggi. RESULTANT diusulkan sebagai teknik yang digunakan untuk memaksimalkan hasil dari salah satu tahapan data mining yaitu preparasi data. Dataset yang digunakan adalah data Lending Club dengan total 2.260.701 record dan 151 variabel. Tahapan yang dilakukan pada RESULTANT adalah seleksi fitur, penanganan missing value, penanganan data outlier dan penanganan data ketidakseimbangan. Dari tahap RESULTANT, dihasilkan 44 variabel akhir yang siap digunakan untuk membangun model dengan menggunakan algoritma XGBoost. Hasil menunjukkan bahwa RESULTANT mampu meningkatkan performa algoritma XGBoost dengan akurasi 99,17%, presisi 99,28%, recall 99,05%, spesifisitas 99,29%, ROC/AUC 99,94%, dan skor f1 99,17%.Kata kunci: XGBoost, Preparasi Data, Seleksi Fitur, Missing Value, OutlierABSTRACTCredit scoring predictions are currently widely used in peer-to-peer lending services by financial technology companies. One of the technologies used for credit scoring is data mining using the XGBoost machine learning algorithm which has a high degree of accuracy. We present RESULTANT as a technique used to maximize the results of one of the stages of data mining, namely data preparation. The dataset used is Lending Club data with a total of 2,260,701 records and 151 variables. The stages carried out in RESULTANT are feature selection, handling missing values, handling outlier data and handling imbalance data. From the RESULTANT stage, 44 final variables are produced which are ready to be used to build models using the XGBoost algorithm. The results showed that RESULTANT was able to improve the performance of the XGBoost algorithm with accuracy 99,17%, precision 99,28%, recall 99,05%, specificity 99,29%, ROC/AUC 99.94%, and f1-score 99,17%.Keywords: XGBoost, Data Preparation, Feature Selection, Missing Value, Outlier
Pengaruh Kualitas Sistem Aplikasi Ovo Terhadap Kepuasan Pelanggan Sofia Umaroh; Rini Rindiyani; Muhammad Ridwan Prasetyo
Rekayasa Hijau : Jurnal Teknologi Ramah Lingkungan Vol 7, No 1 (2023)
Publisher : Institut Teknologi Nasional, Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26760/jrh.v7i1.49-60

Abstract

ABSTRAKRekayasa Hijau: Teknologii sudah sangat berkembang pesat untuk mendorong persaingan dalam dunia bisnis. Salah satunya adalah bisnis berbasis teknologi digital dalam bidang keuangan atau Financial Technology (Fintech). OVO merupakan Fintech yang bergerak di sektor dompet digital (E-wallet). Dengan banyaknya pengguna OVO, maka kualitas sistem dan kepuasan pengguna harus sangat diperhatikan oleh OVO sendiri agar pengguna dapat merasakan kepuasan setiap menggunakan layanan dari OVO. Data yang digunakan dalam penelitian ini adalah data primer yang diperoleh dari kuesioner yang dibagikan kepada mahasiswa perguruan tinggi teknologi pengguna OVO dengan minimal 58 responden dari 52 responden berdasarkan referensi Statistical Power dari Cohen dengan jumlah 2 arah panah, significance level 5% dan minimum R2 0,25. Skala pengukuran yang digunakan adalah skala linier dengan skala 4. Artinya, 4 = sangat setuju, 3 = setuju, 2 = tidak setuju, dan 1 = sangat tidak setuju. Data responden dianalisis menggunakan metode SEM-PLS. Kemudian dilakukan tahap pengujian validitas dan reliabilitas dengan indicator reliability, internal consistency reliability, convergen validity, dan discriminan validity. Pada tahap pengujian validitas dan reliabilitas terdapat indikator yang tidak valid atau reliabel yaitu pada indikator X1.6. Lalu dilakukan analisis multivariate antar konstruk dengan menggunakan pengujian coefficients of determinaton (R2), pengujian ukuran efek (f2), dan uji hipotesis. Pada analisis ini dilakukan dua kali iterasi dimana pada iterasi pertama pada tahap uji hipotesis, hanya hipotesis 2 yang diterima sedangkan hipotesis 1 ditolak. Namun, pada iterasi kedua yang dilakukan dengan menghapus indikator X1.6, hipotesis 1 dan hipotesis 2 diterima.Kata kunci: E-Wallet, Kualitas Sistem, Kepuasan Pelanggan, OVO, SEM-PLSABSTRACTTechnology has developed rapidly to encourage competition in the business world. One of them is a digital technology-based business in the financial sector or Financial Technology (Fintech). OVO is a Fintech engaged in the digital wallet (E-wallet) sector. With so many OVO users, the quality of the system and user satisfaction must be paid close attention to by OVO itself so that users can feel satisfaction every time they use services from OVO. The data used in this study are primary data obtained from questionnaires distributed to college students using OVO technology with a minimum of 58 respondents out of 52 respondents based on Cohen's Statistical Power reference with a total of 2 arrow directions, a significance level of 5% and a minimum R2 of 0, 25. The measurement scale used is a linear scale with a scale of 4. That is, 4 = strongly agree, 3 = agree, 2 = disagree, and 1 = strongly disagree. Respondent data was analyzed using the SEM-PLS method. Then carried out the validity and reliability testing phase with reliability indicators, internal consistency reliability, convergent validity, and discriminant validity. In the validity and reliability testing stage, there are indicators that are not valid or reliable, namely indicator X1.6. Then a multivariate analysis between constructs was carried out by using the coefficients of determinants (R2) test, effect size test (f2), and hypothesis testing. In this analysis two iterations were carried out where in the first iteration at the hypothesis testing stage, only hypothesis 2 was accepted while hypothesis 1 was rejected. However, in the second iteration which was carried out by removing the X1.6 indicator, hypothesis 1 and hypothesis 2 were accepted.Keywords: E-Wallet, System Quality, Customer Satisfaction, OVO, SEM-PLS