The design and implementation of cryptocurrencies prediction system
No Thumbnail Available
Date
2022
Journal Title
Journal ISSN
Volume Title
Publisher
Open Access Color
OpenAIRE Downloads
OpenAIRE Views
Abstract
Özetçe— Dijital para, yalnızca elektronik veya dijital biçimde mevcut olan para birimi türüdür ve yalnızca mobil veya bilgisayar aracılığıyla erişilebilir. Yüzlerce kripto para var 1.025 trilyon dolar değerinde piyasada mevcut. Bitcoin, Ethereum, tether, XRP, Binance, dogecoin, vb. birkaç kripto para birimidir. Bu tezde, çevrimiçi bir uygulama tasarlanmıştır. tarama yapan ve dokuz farklı kripto para biriminin fiyatını tahmin eden. Bu kripto para birimleri Bitcoin, Ethereum, Dogecoin, tether, XRP, Litecoin, Monero, Stellar ve NEM'dir. -de ilk olarak, veriler her kripto para birimi için çevrimiçi olarak taranır. Taranan veriler birleştirilir ve bir her kripto para biriminin fiyatlarını içeren veri seti oluşturulur. Veri seti sağlanır her birinin fiyatındaki eğilimi tahmin eden makine öğrenimi tahmin modellerine kripto para. İki makine öğrenimi algoritması, Uzun Kısa Süreli Bellek ve Conv2D değerlendirme ölçütleri aracılığıyla eğitilir, test edilir ve değerlendirilir. Öngörü sonuçları modeller, LSTM'nin conv2D modelinden daha iyi performans gösterdiğini göstermektedir
Abstract— Digital coin is type of currency available only in electronic or digital form and accessible through mobile or computer only. There are hundreds of cryptocurrencies available in the market worth $1.025 trillion. Bitcoin, Ethereum, tether, XRP, Binance, dogecoin, etc. are a few cryptocurrencies. In this thesis, an online application is designed that crawl and predicts the price of nine different cryptocurrencies. These cryptocurrencies are Bitcoin, Ethereum, Dogecoin, tether, XRP, Litecoin, Monero, Stellar, and NEM. At first, data is crawled online for each cryptocurrency. The crawled data is combined, and a data set is generated containing the prices of each cryptocurrency. The data set is provided to machine learning predictive models which predicts trend in the price of each cryptocurrency. Two machine learning algorithms Long Short-Term Memory and Conv2D are trained, tested, and evaluated through evaluation metrics. The results of predictive models depict that LSTM outperformed conv2D model.
Abstract— Digital coin is type of currency available only in electronic or digital form and accessible through mobile or computer only. There are hundreds of cryptocurrencies available in the market worth $1.025 trillion. Bitcoin, Ethereum, tether, XRP, Binance, dogecoin, etc. are a few cryptocurrencies. In this thesis, an online application is designed that crawl and predicts the price of nine different cryptocurrencies. These cryptocurrencies are Bitcoin, Ethereum, Dogecoin, tether, XRP, Litecoin, Monero, Stellar, and NEM. At first, data is crawled online for each cryptocurrency. The crawled data is combined, and a data set is generated containing the prices of each cryptocurrency. The data set is provided to machine learning predictive models which predicts trend in the price of each cryptocurrency. Two machine learning algorithms Long Short-Term Memory and Conv2D are trained, tested, and evaluated through evaluation metrics. The results of predictive models depict that LSTM outperformed conv2D model.
Description
Keywords
Bilgisayar Mühendisliği Bilimleri-Bilgisayar ve Kontrol, Computer Engineering and Computer Science and Control
Turkish CoHE Thesis Center URL
Fields of Science
Citation
WoS Q
Scopus Q
Source
Volume
Issue
Start Page
End Page
79