Implementasi Analisis Sentimen dan Model Deep Learning Untuk Prediksi Harga Bitcoin

Abstract

Cryptocurrency is a digital or virtual currency that is guaranteed by cryptography. With cryptography, this digital currency has become almost impossible to counterfeit. The recording of all transactions made is stored on the blockchain. This blockchain is widespread between one computer and another and is connected in a network that is widespread so that it is not centralized in one place, otherwise known as decentralization. Sentiment analysis is used to see the reaction of the community or the public in general to an ongoing issue, whether the public reaction is positive or negative. the determination of these sentiments will be based on which label appears the most. Then the sentiment that has been obtained is determined to correlate with the increase in bitcoin prices so that sentiment can be one of the attributes that can be used to increase the accuracy of bitcoin price predictions. Bitcoin is then predicted using a deep learning model with an LSTM layer.