Limbong, Josua Josen Alexander
Unknown Affiliation

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

PENERAPAN METODE DEMPSTER SHAFER UNTUK DIAGNOSA KERUSAKAN BANGUNAN AKIBAT GEMPA DI KOTA JAYAPURA Limbong, Alfa Natasya; Limbong, Josua Josen Alexander; Nurhayati, Siti; Tonggiroh, Mursalim
JATISI Vol 11 No 4 (2024): JATISI (Jurnal Teknik Informatika dan Sistem Informasi)
Publisher : Universitas Multi Data Palembang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35957/jatisi.v11i4.7619

Abstract

One of the impacts of the earthquake that hit Jayapura City in early 2023 was damage to buildings (houses and buildings). The ineffectiveness of the process of inventorying and identifying building damage is one of the obstacles faced by the Jayapura City BPBD, due to the limited number of personnel, the wide coverage of the painting area, and the inventory and discovery processes that are still being carried out manually. In addition, community knowledge about building damage and post-earthquake handling solutions is still limited. For this reason, an expert system is needed as a tool to carry out an inventory and help the level of damage to buildings along with recommendations for solving them quickly and precisely. This study aims to apply the Dempster Shafer method to diagnose earthquake damage to buildings. This method is a case approach technique used to measure a possibility that occurs based on a cause. For problem assistance using the Fishbone analysis method, system design using the Unifield Modeling Language (UML), and for testing using Black-Box Testing. The results of research using the Dempster Shafer method recommend three categories of building damage levels, namely light damage, moderate damage, and heavy damage. This study also recommends solutions for handling according to the type and level of damage to buildings experienced, so that the results of the inventory and damage assistance can be faster and more precise, and the community can be more vigilant in anticipating the arrival of aftershocks.
Analisis Klasifikasi Sentimen Ulasan pada E-Commerce Shopee Berbasis Word Cloud dengan Metode Naive Bayes dan K-Nearest Neighbor Limbong, Josua Josen Alexander; Sembiring, Irwan; hartomo, kristoko dwi
Jurnal Teknologi Informasi dan Ilmu Komputer Vol 9 No 2: April 2022
Publisher : Fakultas Ilmu Komputer, Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.25126/jtiik.2022924960

Abstract

Saat ini internet memungkinkan pengguna untuk membuat ulasan secara online diberbagai jenis platform. Salah satunya aplikasi e-commerce Shopee pada website google play store dimana kelas sentimen positif dan negatif yang terdapat pada ulasan online jelas mencerminkan persepsi pengguna tentang berbagai jenis layanan dan produk yang ada. Selain itu, pelanggan berpotensial yang membaca ulasan online dapat secara signifikan terpengaruh oleh sentimen dari ulasan yang tertera pada kolom ulasan. Hal ini menandakan ulasan yang bersentimen positif ataupun negatif yang ditinggalkan oleh pengguna sangat mempengaruhi pengguna lainnya dalam memilih layanan maupun produk yang dicari. Oleh karena itu perlunya analisis sentimen untuk mengklasifikasi dataset yang begitu banyak sehingga dapat dengan mudah mengetahui apa saja sentimen pelanggan. penelitian ini menggunakan data ulasan sebanyak 500 ulasan . Kemudian ulasan tersebut diklasifikasi menggunakan aplikasi orange dengan metode Naïve Bayes dan K-Nearest Neighbor (KNN). Kemudian selanjutnya menggunakan metode word cloud untuk mengetahui topik-topik yang sering diulas oleh pelanggan. Hasilnya setelah menggunakan metode Naive Bayes memperoleh hasil nilai accuracy 0,914, precision 0,915, recall 0,914 dan F1 score 0,916. Sedangkan metode KNN memperoleh nilai accuracy 0,928,  precision 0,929,  recall 0,928, dan F1 score 0,926. Hal ini membuktikan bahwa dalam penelitian ini kinerja metode KNN lebih baik. Kemudian berdasarkan hasil word cloud yang diperoleh didapatkan informasi kata dengan sentimen positif yang paling sering diulas oleh pelanggan diantaranya terkait kata: gratis, bagus, suka, murah, mudah, dan cepat. Sedangkan informasi sentimen negatif yang diperoleh seperti kata : kecewa, jelek, mahal, bohong, ribet, dan perbaiki. AbstractToday the internet allows users to create online reviews on various types of platforms. One of them is the Shopee e-commerce application on the google play store website, where the positive and negative sentiment classes contained in online reviews reflect user perceptions about the various types of services and products available. Also besides, potential customers who read online reviews can be significantly affected by the sentiment of the reviews listed in the review column. This indicates that positive or negative reviews left by users greatly influence other users in choosing the services or products they are looking for. Therefore the need for sentiment analysis to classify such a large dataset so that you can easily find out what customer sentiments are. This study uses a dataset of 500 reviews. Then the reviews are classified using the orange application with the Naïve Bayes and K-Nearest Neighbor (KNN) methods. Then use the word cloud method to find out topics that are frequently reviewed by customers. The results, after using the Naïve Bayes method, get the accuracy value of 0.914, precision 0.915, recall 0.914, and F1 score 0.916. Meanwhile, the KNN method obtained an accuracy value of 0.928, precision 0.929, recall 0.928, and F1 score 0.926. This proves that in this study the performance of the KNN method is better. Then based on the word cloud results obtained word information with positive sentiments that are most often shared by customers related to words: free, good, like, cheap, easy, and fast. Meanwhile, the negative sentiment information obtained includes the words: disappointed, ugly, expensive, lying, complicated, and fix.