volume-08-Issue 2 (2017)
Latest Articles
Complex system modeling and platform implementation for financial risk management in energy markets
JUSPN, volume-08, Issue 2 (2017) , PP 21 - 27
Published: 20 Nov 2016
DOI: 10.5383/JUSPN.08.02.004
by Oscar Manco, Oscar Botero, Santiago Medina from oscar.manco-lopez@u-pec.fr Université Paris - Est Creteil Paris, France oscar.botero@telecom-em.eu Institute Mines Telecom Paris, France smedina@unal.edu.co Universidad Nacional de Colombia, Colombia
Abstract: Energy markets are in constant evolution and transformation. The dynamics have rapidly changed and both private companies and governments require appropriate tools and systems for management and monitoring. We propose a new approach to measure and manage energy trading financial risk. The approach is twofold: first we use complex modeling to represent the context and second, we implemented an information system to provide risk management tools anytime, anywhere. More in detail, we modeled trading agent’s interactions and observe in advance scenarios that could jeopardize the system equilibrium. We then find the optimal balance for energy trading. From an academic point of view, we contrasted techniques like Nash Equilibrium and Game theory and from a practical experience; we developed a software platform in partnership with the government regulator entity in Colombia XM. read more... read less...
Keywords: Complex systems, risk management, service platform, optimization
From Relational Database to Big Data: Converting Relational to graph database, MOOC database as example
JUSPN, volume-08, Issue 2 (2017) , PP 15 - 20
Published: 20 Nov 2016
DOI: 10.5383/JUSPN.08.02.003
by Yemna Sayeb, Radhouane Ayari, Sarra Naceur, Henda Ben Ghezala from University of Manouba, RIADI, National School of computer science, Manouba campus, Manouba 2010, Tunisia
Abstract: Existing graph database management systems provide an efficient solution to data storage where graph models are widely used, the conversion of an existing application from a relational to graph data can be convenient but it is usually a hard task for database administrators. In this work, we propose a conversion tool from a relational to graph database. The approach supports the conversion of the schema and the data. read more... read less...
Keywords: NoSQL, Migration, Relational Database, Graph database.
Bankruptcy prediction and risk scoring using Hybrid Discriminant Neural Networks
JUSPN, volume-08, Issue 2 (2017) , PP 01 - 06
Published: 26 Nov 2016
DOI: 10.5383/JUSPN.08.02.001
by Fatima zahra Azayite, Said Achchab from National School for Computer Science and Systems analysis, Mohammed V University of Rabat, Morocco
Abstract: Determining the firm risk failure using financial statements has been one of the most interesting subjects for investors and decision makers. The discriminant variables that can be selected to predict firm health influence significantly the accuracy of the models especially if we have a missing data available. We developed a hybrid model of neural networks to study the risk of failure of Moroccan firms taking into account the data availability and reliability. Based on a three-step analysis, this methodology combines discriminant analysis, multilayer neural network and self-organizing-maps. This hybrid model considers the firms’ behaviour during three years to predict risk failure. It is qualified as a dynamic model because it adapts to financial environment and data availability. The model outperforms Discriminant analysis and gives a visual monitoring tool to supervise a firms’ portfolio read more... read less...
Keywords: Neural networks; Bankruptcy prediction; Self Organizing Maps; risk scoring