Repository Universitas Gunadarma >
E-Journal Komputer >
Please use this identifier to cite or link to this item:
|Title: ||Mining for Interesting Indicators Using Reduction Approach|
|Authors: ||Ahmad, F.|
Abu Bakar, A.
|Issue Date: ||17-Jun-2007|
|Publisher: ||Mining for Interesting Indicators Using Reduction Approach|
|Series/Report no.: ||ICEEI2007;|
|Abstract: ||The paper presents a strategy to obtain interesting indicators using a proposed reduction approach. The reduction approach presented is based on the rough set theory. The technique involves indicatorsdependencies in a dataset and then reducing the number of indicators based on rules that it discovered earlier. Several experiments have been conducted using the proposed reduction approach on a set of hardware retailing companies’ financial data. The data comprised of 30 indicators that have been commonly used by financial analyst to measure companies’ performances. A set of most interesting indicators have been discovered from the experiments. The set of indicators has been tested on another dataset for verification. The experimental result shows that the proposed reduction approach has been successful in obtaining a set of indicators with minimum number of indicators and rules with good classification accuracies.|
|Appears in Collections:||E-Journal Komputer|
Items in Repository are protected by copyright, with all rights reserved, unless otherwise indicated.