Database Systems Journal

ISSN 2069 - 3230

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

Database Systems Journal, Vol. VII, Issue 1/2016

Open PDF Journal


1. Tuning I/O Subsystem: A Key Component in RDBMS Performance Tuning (p. 3-11)
Hitesh Kumar SHARMA, University of Petroleum and Energy Studies, India
Christalin NELSON. S, University of Petroleum and Energy Studies, India
Sanjeev Kumar SINGH, Galgotia University Noida, India
In a computer system, the fastest storage component is the CPU cache, followed by the system memory. I/O to disk is thousands of times slower than an access to memory. This fact is the key for why you try to make effective use of memory whenever possible and defer I/Os whenever you can. The majority of the user response time is actually spent waiting for a disk I/O to occur. By making good use of caches in memory and reducing I/O overhead, you can optimize performance. The goal is to retrieve data from memory whenever you can and to use the CPU for other activities whenever you have to wait for I/Os. This paper examines ways to optimize the performance of the system by taking advantage of caching and effective use of the system's CPUs.
Keywords: Tuning, I/O, RDBMS.
2. A Stock Market Prediction Method Based on Support Vector Machines (SVM) and Independent Component Analysis (ICA) (p. 12-21)
Hakob GRIGORYAN, University of Economic Studies, Bucharest, Romania
The research presented in this work focuses on financial time series prediction problem. The integrated prediction model based on support vector machines (SVM) with independent component analysis (ICA) (called SVM-ICA) is proposed for stock market prediction. The presented approach first uses ICA technique to extract important features from the research data, and then applies SVM technique to perform time series prediction. The results obtained from the SVM-ICA technique are compared with the results of SVM-based model without using any pre-processing step. In order to show the effectiveness of the proposed methodology, two different research data are used as illustrative examples. In experiments, the root mean square error (RMSE) measure is used to evaluate the performance of proposed models. The comparative analysis leads to the conclusion that the proposed SVM-ICA model outperforms the simple SVM-based model in forecasting task of nonstationary time series.
Keywords: support vector machines, regression, independent component analysis, financial time series, stock prediction.
3. Forecasting Mobile Games' Retention using Weka (p. 22-27)
Roxana Ioana STIRCU, University of Economic Studies, Bucharest, Romania
In the actual market, when thousands of mobile, PC or console games are released every year, developing and publishing a successful and profitable game is a very challenging process. The gaming industry is very competitive, and all the distribution channels are full of projects competing for players. More and more companies are investing a lot of time and resources in developing an effective way to save and store all the data used and generated by their game's users. In order to develop effective and successful projects, companies adopted a lot of tools and techniques from other domains, like Statistics, Business Intelligence, or Project Management. The method most currently used is Analytics, defined as the process of discovering and communicating patterns in data, to better understand players' behavior, analyze their in-game interaction, and predicting their next in-game actions. This represents a huge step forward for the gaming industry, towards successful projects and user-tailored gaming experience. In this article the problem of users' retention is discussed, and a regression model is proposed in order to forecast players' retention, and prevent players from leaving the game.
Keywords: game analytics, metrics, user behavior, Weka, linear regression, forecast, players' retention.
4. A data mining approach for estimating patient demand for health services (p. 28-34)
Ionuţ ŢĂRANU, University of Economic Studies, Bucharest, Romania
The ability to better forecast demand for health services is a critical element to maintaining a stable quality of care. Knowing how certain events can impact requirements, health-care service supplier can better assign available resources to more effectively treat patients' needs.
The embodiment of data mining analytics can support available data to identify cyclical patterns through relevant variables, and these patterns provide actionable information to adequate decision markers at health-care structures.
The request for health-care services can be subject to change from time of year (seasonality) and economic factors. This paper exemplifies the efficacy of data mining analytics in identifying seasonality and economic factors as measured by time that affect patient demand for health-care services. It incorporates a neural network analytic method that is applied over a readily available dataset. The results indicate that day of week, month of year, and a yearly trend significantly impact the demand for patient services.
Keywords: Data mining, neuronal networks, decision support systems, healthcare IT.
5. Implementing Business Intelligence System - Case Study (p. 35-44)
Yasser AL-HADAD, University of Economic Studies, Bucharest, Romania
Razvan Daniel ZOTA, University of Economic Studies, Bucharest, Romania
Understanding and analysis data is essential for making decision within a system. Any analytical tasks can be implemented directly by the transactional system but it becomes more difficult as the transactional system grows. Analytical systems and their extension appear as a solution for complex and large datasets. We think that it's time for medium companies to get the benefit from such systems as analytical systems become more variant and in hand for every possible user. In this paper, we propose an architecture of analytical system that can adapt and integrate with existent transactional system of timber export company. The proposed analytical system should have the ability of implementing the tasks required by the decision makers of the system. Also, we try to explore the ability of SQL server of implementing our proposed architecture.
Keywords: BI(Business intelligence), DSS(Decision support systems), DW(Data warehouse), OLAP(Online Analytical Processing), ETL(Extract, transform and load), SAS(SQL server services).