DMIA 2010 Abstracts


Full Papers
Paper Nr: 3
Title:

A CASE STUDY - Classification of Stock Exchange News by Support Vector Machines

Authors:

P. Kroha, K. Kröber and R. Janetzko

Abstract: In this paper, we present a case study concerning the classification of text messages with the use of Support Vector Machines. We collected about 700.000 news and stated the hypothesis saying that when markets are going down then negative messages have a majority and when markets are going up then positive messages have a majority. This hypothesis is based on the assumption of news-driven behavior of investors. To check the hypothesis given above we needed to classify the market news. We describe the application of Support Vector Machines for this purpose including our experiments that showed interesting results. We found that the news classification has some interesting correlation with long-term market trends.

Paper Nr: 5
Title:

MODEL-DRIVEN AD HOC DATA INTEGRATION IN THE CONTEXT OF A POPULATION-BASED CANCER REGISTRY

Authors:

Yvette Teiken, Martin Rohde and Hans-Jürgen Appelrath

Abstract: The major task of a population-based Cancer Registry (CR) is the identification of risk groups and factors. This analysis makes use of data about the social background of the population. The integration of that data is not intended for the routine processes at the CR. Therefore, this process must be performed by data warehouse experts that results in high cost. This paper proposes an approach, which allows epidemiologists and physicians at the CR to realize this ad hoc data integration on their own. We use model driven software design (MDSD) with a domain specific language (DSL), which allows the epidemiologists and physicians to describe the data to be integrated in a known language. This description or rather model is used to create an extension of the existing data pool and a web service and web application for data integration. The end user can do the integration on his/her own which results in a very cost-efficient way of ad hoc data integration.

Short Papers
Paper Nr: 2
Title:

OLAP FOR FINANCIAL ANALYSIS AND PLANNING - A Proof of Concept

Authors:

Eitel J. M. Lauría and Carlos A. Greco

Abstract: We describe the design of an in-memory OLAP model for financial planning, analysis and reporting at a medium sized manufacturing company in South America. The architecture and data model design are explained within the context of the company’s requirements and constraints.

Paper Nr: 4
Title:

PURPOSE-DRIVEN APPROACH FOR FLEXIBLE STRUCTURE-INDEPENDENT DATABASE DESIGN

Authors:

Youri I. Rogozov, Alexander S. Sviridov, Sergey A. Kutcherov and Wladimir Bodrow

Abstract: This paper presents a purpose-driven approach for development of flexible databases outgoing from the relational database concept. Based on carried out analysis of both relational and entity-attribute-value database models the aspects essential for the described purpose-driven approach are defined. The necessary requirements to be satisfied by structure-independent databases are derived and discussed in detail. Several implementations of structure-independent databases using the suggested approach have been realized and presented as well. The improvement of relational database model based on proposed structure-independent database approach is formulated.