NORMA eResearch @NCI Library

Prediction of Foreign Exchange Rate Using Data MiningEnsemble Method

Kumar, Vimalraj (2016) Prediction of Foreign Exchange Rate Using Data MiningEnsemble Method. Masters thesis, Dublin, National College of Ireland.

[thumbnail of Master of Science]
PDF (Master of Science)
Download (1MB) | Preview
[thumbnail of Configuration File]
PDF (Configuration File)
Download (2MB) | Preview


This project presents the implementation prediction that can accuracy predict the foreign exchange rate. how the prediction accuracy can be improved by developing an ensemble model of the deep learning algorithm, Distributed Random forest and generalised linear model using sparkling water (Spark +H20). According to the researchers of literature review from 2000-2016, there are several models that has been used for predicting the foreign exchange rate. Among which Artificial Neural Network, Linear regression, Support vector machine, Arima are best-suited models for predicting the time series data. By having the above models as a base for this project and also by considering this project with the time series data of exchange rate, an additional feature of an ensemble is done on the output of these three models. This model is also implemented on the big data for getting better accuracy and faster predictions. Each model by using sparkling water produces the accuracy
of more the 90% and the ensemble models increase the accuracy of the model by 3% with the accuracy of 93%. The evaluation and the results of this model clearly delivers the ensemble method, on the sparkling water which can efficiently improve the accuracy and performance of the model.

Item Type: Thesis (Masters)
Subjects: Q Science > QA Mathematics > Electronic computers. Computer science
T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science
H Social Sciences > HG Finance > Money > Currency
Divisions: School of Computing > Master of Science in Data Analytics
Depositing User: Caoimhe Ní Mhaicín
Date Deposited: 03 Dec 2016 14:35
Last Modified: 03 Dec 2016 14:35

Actions (login required)

View Item View Item