ANALISIS SENTIMEN PENGGUNA TWITTER TERHADAP TOPIK CRYPTOCURRENCY MENGGUNAKAN ALGORITMA DEEP LEARNING RNN-LSTM

Syauqi Muhammad Abrar, NIM.: 19106050015 (2023) ANALISIS SENTIMEN PENGGUNA TWITTER TERHADAP TOPIK CRYPTOCURRENCY MENGGUNAKAN ALGORITMA DEEP LEARNING RNN-LSTM. Skripsi thesis, UIN SUNAN KALIJAGA YOGYAKARTA.

Preview	Text (ANALISIS SENTIMEN PENGGUNA TWITTER TERHADAP TOPIK CRYPTOCURRENCY MENGGUNAKAN ALGORITMA DEEP LEARNING RNN-LSTM) 19106050015_BAB-I_IV-atau-V_DAFTAR-PUSTAKA.pdf - Published Version Download (2MB) | Preview
	Text (ANALISIS SENTIMEN PENGGUNA TWITTER TERHADAP TOPIK CRYPTOCURRENCY MENGGUNAKAN ALGORITMA DEEP LEARNING RNN-LSTM) 19106050015_BAB-II_sampai_SEBELUM-BAB-TERAKHIR.pdf - Published Version Restricted to Registered users only Download (4MB) | Request a copy

Abstract

The rapid growth of cryptocurrency has become an increasingly important topic in the global financial world, including in Indonesia. In this context, sentiment analysis of public opinion and views on social media such as Twitter is relevant. This research focuses on two main topics, namely "Indonesian crypto exchange" and "crypto MUI fatwa," with the aim of understanding how society responds to the development of cryptocurrency in Indonesia. The problem at hand is how to analyze the sentiments contained in cryptocurrency-related tweets and the extent to which deep learning models are able to understand people's views. This research involves collecting tweet data related to the two topics, labeling data to determine positive and negative sentiments, and preprocessing to convert text data into a format suitable for deep learning models. The model used is Long Short-Term Memory (LSTM). The model is trained using training data and then tested on validation data to measure its performance. The results of this study indicate that the LSTM model can be successfully used to perform sentiment analysis on cryptocurrency-related tweets. For the topic "Indonesian crypto exchange," the model achieves an accuracy of 87.7% on training data and 89% on validation data. Meanwhile, on the topic of "crypto MUI fatwas," the model achieved an accuracy of 75.2% on training data and 74.3% on validation data. These results indicate that the LSTM model has a good ability to recognize public sentiment on cryptocurrency topics in Indonesia.

Share this knowledge with your friends :

Actions (login required)

View Item