Analisis Kepuasan Penggunaan Aplikasi Shopee Menggunakan Algoritma Naïve Bayes

Abstract

The Covid-19 pandemic that attacks the world's health conditions, including Indonesia, greatly affects daily life. The impacts include the need to maintain distance and avoid crowds, which forces people to reduce interactions between people. Therefore, people need other ways to be able to meet their daily needs. One alternative that can be done is to use e-commerce applications to conduct online transactions or also known as online shopping. It was reported on September 30, 2021 that shopee occupies the most superior e-commerce in Indonesia with the number of monthly visitors which currently reaches 93 million subscribers, followed by Tokopedia with 86 million subscribers and in third place with 35 million customers occupied by Bukalapak. Therefore, an analysis of satisfaction with the use of e-commerce applications with the Shopee case study will be carried out to determine the response from users to the applications that are declared the most desirable. The application will be made based on a website to make it easier to use and use the Naïve Bayes algorithm as the analysis method. The Naïve Bayes algorithm is one method that can classify data which in this case are comments from application users into three types of comments, namely positive, negative and neutral. The process that is run also includes a preprocessing task as raw data processing into ready-to-use data and a confusion matrix as a calculation of the accuracy of the resulting application. Testing of the application is carried out in two ways, namely using a confusion matrix using a system with 80% results and a user acceptance test with 94.8% results, namely very good predicate. From the analysis, it can be concluded that Shopee users are satisfied with the application used and the classification of user reviews using the Naïve Bayes Algorithm produces a fairly good accuracy.

Implementasi Algoritma Naïve Bayes dalam Menganalisis Sentimen Review Pengguna Tokopedia pada Produk Kesehatan

“…Hal tersebut menunjukkan bahwa kemampuan faktor-faktor independen dalam penelitian ini mempengaruhi variabel dependen sebesar 49,7%, sedangkan sisanya 50,3% dijelaskan oleh variabel lain. [6] 2. METODOLOGI PENELITIAN Identifikasi masalah dilakukan untuk menjelaskan dan mendefenisikan masalah-masalah yang dihadapi.…”

Implementasi Algoritma Naïve Bayes Dalam Menganalisis Jumlah Live Stream VTuber Skem

“…[13]-[14]-[15]-[16]-[17] class yang belum diketahui H : Hipotesis data merupakan suatu class spesifik P(H|X) :Probabilitas hipotesis H berdasar kondisi X (posteriori probabilitas) P(H) : Probabilitas hipotesis H (prior probabilitas) P(X|H) :Probabilitas X berdasarkan kondisi pada hipotesis H P(X) : Probabilitas X Untuk menjelaskan metode Naive Bayes, perlu diketahui bahwa proses klasifikasi memerlukan sejumlah petunjuk untuk menentukan kelas apa yang cocok bagi sampel yang dianalisis tersebut. Karena itu, metode Naive Bayes di atas disesuaikan sebagai berikut: Di mana Variabel C merepresentasikan kelas, sementara variabel F1 ... Fn merepresentasikan karakteristik petunjuk yang dibutuhkan untuk melakukan klasifikasi.…”