Treffer: Primjena umjetne inteligencije u trgovinskom pregovaranju ; The use of artificial intelligence in trade negotiation

Title:
Primjena umjetne inteligencije u trgovinskom pregovaranju ; The use of artificial intelligence in trade negotiation
Authors:
Contributors:
Stipaničev, Darko
Publisher Information:
Sveučilište u Splitu. Fakultet elektrotehnike, strojarstva i brodogradnje. Zavod za elektroniku i računarstvo.
University of Split. Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture. Department of Electronics and Computing.
Publication Year:
2021
Collection:
The University of Split Repository
Document Type:
Dissertation master thesis
File Description:
application/pdf
Language:
Croatian
Rights:
http://rightsstatements.org/vocab/InC/1.0/ ; info:eu-repo/semantics/restrictedAccess
Accession Number:
edsbas.62BDD17F
Database:
BASE

Weitere Informationen

Zadatak diplomskog rada je bio realizirati model koji primjenom metoda i algoritama umjetne inteligencije vrši trgovinsko pregovaranje, s naglaskom na kriptovalute. Na početku rada definirani su osnovni koncepti umjetne inteligencije kao i kriptovaluta. Nakon toga objašnjene su tehnologije koje su bile nužne za izradu rada te je na samom kraju prikazan detaljan proces izgradnje LSTM mreže koja se, pomoću podataka s burze, trenira, a potom na skupu novih podataka testira predviđanje i u konačnici predviđa buduću cijenu. Za izradu rada korišten je programski jezik Python te pripadne mu biblioteke NumPy, Pandas, Matplotlib, Scikit-Learn i Tensorflow. Izgrađena LSTM mreža predviđa cijene te ukoliko procjeni da je kupovina ili prodaja isplativa, izvršava stvarnu transakciju. U radu su priloženi grafovi, usporedbi predviđene i stvarne cijene, iz kojih je vidljivo da porastom količine podataka pri treniranju raste točnost predviđanja. ; The purpose of my final assignment was to create a model which uses different methods and algorithms of artificial intelligence to trade cryptocurrencies. In the beginning of the paper I have defined the basics of AI and cryptocurrencies. Later on I have explained the technologies that were necessary to create the model and at the end I have shown the full process of creating the LSTM network in detail which trains itself by using information from the stock market and after aquarring new info tests it's price prediction by comparing it to the real price and finaly predicts the future price of the crypto currency. For creating the project I have used the programming language Python with its libraries NumPy, Pandas, Matplotlib, Scikit-Learn and Tensorflow. By using it's predictions the LSTM network determes if a certain currency is worth selling or buying before executing the transaction. In my paper I have displayed some grafs in which the difference of the real price and the predicted price is shown and as we can see the difference grows smaller the more information the AI has received ...