Unit:
Κατεύθυνση Ηλεκτρονικός Αυτοματισμός (Η/Α, με πρόσθετη εξειδίκευση στην Πληροφορική και στα πληροφοριακά συστήματα)Library of the School of Science
Author:
Petropoulos Dimitrios
Supervisors info:
Διονύσιος Ι. Ρεΐσης Αναπληρωτής Καθηγητής
Original Title:
Αλγόριθμος προβλέψεων τύπου Convolutional Neural Network με εφαρμογή στην αγορά συναλλαγμάτων(FOREX)
Translated title:
Convolutional Neural Network algorithm with implementation to the currency market
Summary:
In this thesis a Convolutional Neural Network is presented with implementation to the Euro-U.S.Dollar currency market. At first, data have to be formatted because of their big volume since every single second many transactions may occur. Our goal is to reduce the data size and the also reduce the data variance.
After data formatting, the basic structure of the Neural Network is presented and the algorithm is being optimized by using data scaling which is our main optimization technique.
The final part the algorithms’ parameters such as batch size, epoch and batch size, are being optimized and the optimization results are being presented.
Main subject category:
Science
Keywords:
Convolutional Neural Network, currency market forex