Analysis and design insights for an E-finance platform using parallel processing

Authors:
Darie Moldovan;Mircea Moca;Mircea Rusu
Affiliations:
Business Information Systems Department, Babeş-Bolyai University of Cluj-Napoca, România;Business Information Systems Department, Babeş-Bolyai University of Cluj-Napoca, România;IT Research Department, Company for Applied Informatics Cluj-Napoca, România
Venue:
ACA'12 Proceedings of the 11th international conference on Applications of Electrical and Computer Engineering
Year:
2012

Citing 5
Cited 0

Opinion Mining and Sentiment Analysis

Foundations and Trends in Information Retrieval
MRPGA: An Extension of MapReduce for Parallelizing Genetic Algorithms

ESCIENCE '08 Proceedings of the 2008 Fourth IEEE International Conference on eScience
Gene trajectory clustering for learning the stock market sectors

ICANNGA'09 Proceedings of the 9th international conference on Adaptive and natural computing algorithms
Towards MapReduce for Desktop Grid Computing

3PGCIC '10 Proceedings of the 2010 International Conference on P2P, Parallel, Grid, Cloud and Internet Computing
Influence of Communication Models on the Scalability of Master-Slave Platforms Running Bag-of-Tasks Applications

WSCAD-SCC '10 Proceedings of the 2010 11th Symposium on Computing Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper we provide the analysis for an e-finance platform and propose a design solution using state-of-the-art models in finance and parallel processing. The problem that we address lays in the context of financial markets that are rapidly changing and the majority of investors are facing important problems such as: the lack of integrated, advanced tools to analyze data, high-level services that automatically react to market changes and propose investment alternatives in near real-time and the lack of mechanisms to integrate clean data from multiple sources. These issues lead, in most cases, to poor investment decisions. Hence, in this paper we proposed the design of an e-finance framework to provide a set of services that tackle these issues. The proposed services rely on computing-intensive methods like: text-mining, Neural Networks and Genetic Algorithms, enhanced by applying relevant findings from the efficient-market theory study. In order to improve the execution of these services we propose their parallelization on distributed computing infrastructures. In our discussion we highlighted the importance of such a framework for the financial market investors.