Analisis Sentimen Pengguna Aplikasi Pendanaan Modal UMKM Menggunakan Metode Naive Bayes Classifier (NBC)

Pujianto, Christian Ruben (2021) Analisis Sentimen Pengguna Aplikasi Pendanaan Modal UMKM Menggunakan Metode Naive Bayes Classifier (NBC). Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 10611710000087-Undergraduate_Thesis.pdf] Text
10611710000087-Undergraduate_Thesis.pdf - Accepted Version
Restricted to Repository staff only until 1 October 2023.

Download (777kB) | Request a copy
[thumbnail of 10611710000087-Undergraduate_Thesis.pdf] Text
10611710000087-Undergraduate_Thesis.pdf
Restricted to Repository staff only

Download (777kB) | Request a copy

Abstract

Perkembangan zaman saat ini membuat UMKM untuk cepat beradaptasi dengan lingkungan bisnis yang berbasis teknolgi digital. Salah satu tantangan UMKM yaitu kurang nya informasi mengenai pasar untuk meningkatkan modal awal. Salah satu platform pendanaan modal yang ada di Indonesia yaitu Santara dimana layanan yang disediakan oleh aplikasi Santara yaitu penyediaan modal bagi bisnis UMKM melalui crowdfunding. Platform tersebut diluncurkan pada tahun 2018 dan tergolong baru karena masih banyak masyarakat yang kurang sadar terhadap layanan crowdfunding dari aplikasi tersebut sehingga banyak respon masyarakat yang bersifat negatif maupun positif. Penelitian ini bertujuan untuk mengetahui klasifikasi ulasan aplikasi Santara menggunakan analisis sentimen dan metode naïve bayes classifier. Metode yang digunakan dalam machine learning yaitu naïve bayes classifier untuk mengukur ketepatan klasifikasi dan membandingkan performansi algoritma menggunakan proporsi data latih yang berbeda. Hasil klasifikasi menggunanakan analisis sentimen menunjukkan bahwa sebanyak 377 ulasan merupakan ulasan positif sedangkan 197 ulasan merupakan ulasan negatif. Hasil klasifikasi menggunakan metode naïve bayes classifierdidapatkan bahwa hasil pembagian data latih dan data tes sebesar 90:10 memberikan nilai yang lebih baik daripada 80:20 dan 75:25. Perbandingan tersebut dilihat dari nilai akurasi, dan AUC sebesar 89.63% dan 96.49% yang menunjukkan rasio 90:10 menghasilkan ketepatan klasifikasi yang lebih besar daripada yang lain.
========================================================================================================
Indonesia's economic growth is based on the growth rate of MSMEs because MSMEs are one of the pillars of the Indonesian economy. The current development makes MSMEs to quickly adapt to a business environment based on digital technology. One of the challenges for MSMEs is the lack of information about the market to increase initial capital. This makes MSME players digitize the problem of providing business capital through an internet-based platform. One of the capital funding platforms in Indonesia is Santara, where the services provided by the Santara application are providing capital for MSME businesses through crowdfunding. The platform was launched in 2018 and is relatively new because there are still many people who are not aware of the crowdfunding services from the application so that many community responses are negative or positive. This study aims to determine the classification of MSME capital funding application reviews, namely Santara, using sentiment analysis and the naïve Bayes classifier method to determine the accuracy of the classification of sentiments. This study uses two techniques, namely data mining and machine learning. The method used in data mining is Google Play Store data scraping. The method used in machine learning is the naïve Bayes classifier to measure classification accuracy and compare the classification performance of algorithms using different proportions of training data. The results of the classification using the nave Bayes classifier method were found that the results of the distribution of training data and test data of 90:10 gave a better value. The comparison is seen from the value of accuracy, and sensitivity of 89.63% and 96.49% which shows a ratio of 90:10 resulting in greater classification accuracy than the others.

Item Type: Thesis (Undergraduate)
Uncontrolled Keywords: Akurasi, Analisis Sentimen, Data Mining, Metode Klasifikasi, Naïve Bayes Classifier, Santara, UMKM, Accuracy, Classification Method, Data Mining, MSME, Naïve Bayes Classifier, Santara, Sentiment Analysis
Subjects: Q Science > Q Science (General) > Q325.5 Machine learning.
Divisions: Faculty of Vocational > 49501-Business Statistics
Depositing User: Christian Ruben Pujianto
Date Deposited: 18 Aug 2021 09:18
Last Modified: 18 Aug 2021 09:18
URI: http://repository.its.ac.id/id/eprint/87070

Actions (login required)

View Item View Item