Voteteknika (Vocational Teknik Elektronika dan Informatika)
Vol 12, No 3 (2024): Vol. 12, No 3, September 2024

Analisis Sentimen Aplikasi Lion Parcel Menggunakan Lexicon Based dan Support Vector Machine

Bobby Girsang (Universitas Kristen Immanuel)
Sunneng Sandino Berutu (Universitas Kristen Immanuel)
Haeni Budiati (Universitas Kristen Immanuel)
Febe Maedjaja (Universitas Kristen Immanuel)



Article Info

Publish Date
10 Sep 2024

Abstract

Dengan adanya perkembangan teknologi digital pada masa kini, masyarakat dapat melakukan pengiriman barang jarak jauh dengan lebih mudah dan efisien. Lion parcel merupakan salah satu aplikasi ekpedisi populer yang melayani pengiriman barang. Penelitian ini dilakukan untuk mengukur sentimen yang diberikan oleh para pengguna aplikasi terhadap jasa ekspedisi dari Lion Parcel. Proses analisis sentimen ini dilakukan melalui tahapan pengumpulan data, preprocessing data, labelling data menggunakan pendekatan berbasis lexicon, dan membangun model klasifikasi berbasis algoritma Support Vector Machine (SVM) melalui tahap pelatihan dan pengujian data, dan evaluasi model dengan confusion matrix. Data yang berhasil dikumpulkan melalui scrapping data dari ulasan aplikasi Lion Parcel pada google play store mencapai 1194 ulasan. Hasil yang didapatkan setelah memproses data yang diambil yaitu sentimen positif mencapai 55.51 % dengan jumlah 630 data. Sentimen netral mencapai 24.76% dengan jumlah 281 data. Sentimen negatif mencapai 19.74% dengan jumlah 224 data. Hasil pengujian SVM juga menghasilkan nilai akurasi, presisi, recall dan f1-score tertinggi berturut turut sebesar 77.78%, 85%, 94% dan 89%.Kata kunci : Lion Parcel, SVM, sentimen, Lexicon Based. With the advancement of digital technology today, people can deliver long-distance goods more easily and efficiently. Lion parcel is one of the popular expedition applications that serves the delivery of goods. The research was conducted to measure the sentiment given by the users of the application to the Lion Parcel's expedition services. This sentiment analysis process is carried out through stages of data collection, data preprocessing, data labelling using lexicon-based approaches, and building an algorithm-based classification model based on the Support Vector Machine (SVM) through training and data testing stages, and model evaluation with a confusion matrix. The data collected successfully through scrapping data from the review of the Lion Parcel app on the google play store reached 1194 reviews. The results obtained after processing the data taken are positive sentiment reached 55.51% with a total of 630 data. Neutral feeling reached 24.76% with the total of 281 data. Negative feelings reached 19.74% with 224 data. SVM test results also yielded the highest accuracy, precision, recall and f1 scores in a row of 77.78%, 85%, 94% and 89%.Keywords: Lion Parcel, SVM, Sentiment, Lexicon Based.

Copyrights © 2024






Journal Info

Abbrev

voteknika

Publisher

Subject

Computer Science & IT Control & Systems Engineering Education Electrical & Electronics Engineering

Description

Jurnal Vocational Teknik Elektronika dan Informatika (VoteTEKNIKA) is a peer-reviewed, scientifc journal published by Department of Electronics Engineering, Faculty of Engineering, Universitas Negeri Padang, Indonesia. The aim of this journal is to publish articles dedicated to all aspects of the ...