Database Systems Journal

ISSN 2069 - 3230

Published with the support of:
The Bucharest University of Economic Studies

Database Systems Journal, Vol. VI, Issue 2/2015
Issue Topic: NoSQL Databases

Open PDF Journal


1. Architectures for the Development of the National Interoperability Framework in Romania (p. 3-13)
Codrin-Florentin NISIOIU, University of Economic Studies, Bucharest, Romania
The authors of Digital Agenda consider that Europe do not take fully advantage of interoperability. They believe that we need effective interoperability between IT products and services to build a truly Digital Society. The Digital Agenda can only be effective if all the elements and applications are interoperable and based on open standards and platforms. In this context, I propose in this article a specific architecture for developing Romanian National Interoperability framework.
Keywords: interoperability, collaborative working environment, cloud computing.
2. Stock Market Prediction using Artificial Neural Networks. Case Study of TAL1T, Nasdaq OMX Baltic Stock (p. 14-23)
Hakob GRIGORYAN, University of Economic Studies, Bucharest, Romania
Predicting financial market changes is an important issue in time series analysis, receiving an increasing attention in last two decades. The combined prediction model, based on artificial neural networks (ANNs) with principal component analysis (PCA) for financial time series forecasting is presented in this work. In the modeling step, technical analysis has been conducted to select technical indicators. Then PCA approach was applied to extract the principal components from the variables for the training step. Finally, the ANN-based model called NARX was used to train the data and perform the time series forecast. TAL1T stock of Nasdaq OMX Baltic stock exchange was used as a case study. The mean square error (MSE) measure was used to evaluate the performances of proposed model. The experimental results lead to the conclusion that the proposed model can be successfully used as an alternative method to standard statistical techniques for financial time series forecasting.
Keywords: artificial neural networks, NARX, principal component analysis, financial time series, stock prediction.
3. Enhancing Forecasting Performance of Naive-Bayes Classifiers with Discretization Techniques (p. 24-30)
Ruxandra PETRE, University of Economic Studies, Bucharest, Romania
During recent years, the amounts of data, collected and stored by organizations on a daily basis, have been growing constantly. These large volumes of data need to be analyzed, so organizations need innovative new solutions for extracting the significant information from these data. Such solutions are provided by data mining techniques, which apply advanced data analysis methods for discovering meaningful patterns within the raw data. In order to apply these techniques, such as Naïve-Bayes classifier, data needs to be preprocessed and transformed, to increase the accuracy and efficiency of the algorithms and obtain the best results.
This paper focuses on performing a comparative analysis of the forecasting performance obtained with the Naïve-Bayes classifier on a dataset, by applying different data discretization methods opposed to running the algorithms on the initial dataset.
Keywords: Discretization, Naive-Bayes classifier, Data mining, Performance.
4. Big Data, indispensable today (p. 31-38)
Radu-Ioan ENACHE, University of Economic Studies, Bucharest, Romania
Marian Adrian ENE, University of Economic Studies, Bucharest, Romania
Big data is and will be used more in the future as a tool for everything that happens both online and offline. Of course , online is a real hobbit, Big Data is found in this medium , offering many advantages , being a real help for all consumers. In this paper we talked about Big Data as being a plus in developing new applications, by gathering useful information about the users and their behaviour.We've also presented the key aspects of real-time monitoring and the architecture principles of this technology. The most important benefit brought to this paper is presented in the cloud section.
Keywords: Big Data, Cloud Data, Financial Data, Data encryption.
5. Customer Data Analysis Model using Business Intelligence Tools in Telecommunication Companies (p. 39-47)
Monica LIA, University of Economic Studies, Bucharest, Romania
This article presents a customer data analysis model in a telecommunication company and business intelligence tools for data modelling, transforming, data visualization and dynamic reports building . For a mature market, knowing the information inside the data and making forecast for strategic decision become more important in Romanian Market. Business Intelligence tools are used in business organization as support for decision making.
Keywords: Customer Analysis, Business Intelligence, Data Warehouse, Data Mining, decisions, self-service reports, interactive visual analysis, and dynamic dashboards, Use Cases Diagram, Process Modelling, Logical Data Model, Data Mart, ETL, Star Schema, OLAP, Data Universes.
6. Business Intelligence Methods for Sustainable Development of the Railways (p. 48-55)
Aida-Maria POPA, University of Economic Studies, Bucharest, Romania
This paper aims to present a new approach of business intelligence technologies in the context of sustainable development of the railways. The concept of business intelligence is increasingly used in the developed companies and considering that the current economic market is more dynamic from year to year, business intelligence solutions plays an important role for companies to be able to develop efficient plans for both short-term and medium and long term developing. This paper will focus on two technologies: data-warehouse and data-mining and how are they use in the railway business. The subject adapts to the current development trend of European countries to direct the transport of freight and passengers to the railway for support environment.
Keywords: sustainable development, business intelligence, data warehouse, data mining.
7. Stochastic Processes and Queueing Theory used in Cloud Computer Performance Simulations (p. 56-62)
Florin-Catalin ENACHE, University of Economic Studies, Bucharest, Romania
The growing character of the cloud business has manifested exponentially in the last 5 years. The capacity managers need to concentrate on a practical way to simulate the random demands a cloud infrastructure could face, even if there are not too many mathematical tools to simulate such demands.This paper presents an introduction into the most important stochastic processes and queueing theory concepts used for modeling computer performance. Moreover, it shows the cases where such concepts are applicable and when not, using clear programming examples on how to simulate a queue, and how to use and validate a simulation, when there are no mathematical concepts to back it up.
Keywords: capacity planning, capacity management, queueing theory, statistics, metrics.