Claim Missing Document
Check
Articles

Found 2 Documents
Search
Journal : JOURNAL OF SCIENCE AND SOCIAL RESEARCH

IDENTIFIKASI TINGKAT KEMATANGAN BUAH MANGGA MENGGUNAKAN METODE K-MEANS CLUESTERING DAN MEDIAN FILTER Yanti, Rahma; Yasmin, Nabilla; Putra, Kharisma Utama; Irawan, Hendri; Sovia, Rini
JOURNAL OF SCIENCE AND SOCIAL RESEARCH Vol 8, No 2 (2025): May 2025
Publisher : Smart Education

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.54314/jssr.v8i2.2893

Abstract

Abstract: This study aims to develop an automatic system for identifying the ripeness level of mangoes using the K-Means Clustering and Median Filter methods. The background of this research is based on the agricultural industry's need for an objective ripeness assessment, as manual methods are often subjective and inefficient. The K-Means Clustering method is used to categorize mango ripeness based on skin color characteristics, while the Median Filter is applied to enhance image quality by reducing noise before clustering. This study utilizes a dataset of 120 mango images, consisting of 47 images for training and 73 images for testing. The results indicate that the combination of these two methods achieves a classification accuracy of 98%. These findings contribute to the development of digital image processing technology for applications in the agricultural and food industries. Keyword: Ripeness identification, K-Means Clustering, Median Filter, Image Processing, Mango. Abstrak: Penelitian ini bertujuan untuk mengembangkan sistem identifikasi tingkat kematangan buah mangga secara otomatis menggunakan metode K-Means Clustering dan Median Filter. Latar belakang penelitian ini didasarkan pada kebutuhan industri pertanian dalam menentukan tingkat kematangan mangga secara objektif, mengingat metode manual sering kali subjektif dan kurang efisien. Metode K-Means Clustering digunakan untuk mengelompokkan tingkat kematangan mangga berdasarkan karakteristik warna kulit, sedangkan Median Filter diterapkan untuk meningkatkan kualitas citra dengan mengurangi noise sebelum dilakukan proses klasterisasi. Penelitian ini menggunakan dataset sebanyak 120 citra mangga, yang terdiri dari 47 citra untuk pelatihan dan 73 citra untuk pengujian. Hasil penelitian menunjukkan bahwa kombinasi kedua metode ini mampu mengklasifikasikan tingkat kematangan mangga dengan akurasi sebesar 98%. Temuan ini memberikan kontribusi dalam pengembangan teknologi pemrosesan citra digital untuk aplikasi dalam industri pertanian dan pangan. Kata kunci: Identifikasi kematangan, K-Means Clustering, Median Filter, Pengolahan Citra, Mangga.
IMPLEMENTASI ALGORITMA FUZZY UNTUK PENILAIAN KEPUASAN NASABAH PNM MEKAR DI PASAMAN Yanti, Rahma; Ramadani, Sela; Selvia, Dina; Sovia, Rini
JOURNAL OF SCIENCE AND SOCIAL RESEARCH Vol 8, No 4 (2025): November 2025
Publisher : Smart Education

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.54314/jssr.v8i4.4849

Abstract

Customer satisfaction assessment is an essential component in improving the service quality of PNM Mekar, a microfinance institution focused on empowering women through ultra-micro financing. Conventional evaluations rely heavily on subjective perceptions, creating a need for a more structured and objective method. This study applies the Fuzzy Logic algorithm to measure customer satisfaction by transforming numerical data into linguistic variables through fuzzification. Annual operational data, including the number of customers and returning customers, were processed using membership functions and fuzzy rules, followed by defuzzification to obtain a crisp satisfaction value. The results indicate that all satisfaction levels fall into the low category, suggesting the need for service improvement. The fuzzy-based model proves effective in providing adaptive, consistent, and realistic satisfaction evaluation.