Pendekatan self-organizing maps dalam data mining untuk clustering perubahan harga saham=Seft organizing maps approach in data mining for clustering the ...

• Main Author: Perpustakaan UGM, i-lib
• Format: Article NonPeerReviewed
• Published: [Yogyakarta] : Universitas Gadjah Mada 2004
• Subjects: Jurnal i-lib UGM
• Online Access: https://repository.ugm.ac.id/17682/; http://i-lib.ugm.ac.id/jurnal/download.php?dataId=444
• Institution: Universitas Gadjah Mada

ABSTRACT Money market is hoped can collect society's moneys for building and increasing society's income. Society as the main investor on money market has to know and understand the analysis of stock infestation for knowing how much of it, so they can give the most optimal return. One of the approaches for evaluating stock infestation is technical analysis that used data or note of market, which is published. For examples stock cost, market volume, the index of consolidations stock or individual, and the other factor which have technical characters. The purpose of this research are for making a system which based to SOM for knowing what day the prices of stock are highest or lowest based on the frequent of each day appear and for knowing which algorithm is the most objective. This research use undirected data mining method that is clustering. Self-Organizing Maps (SOM) with training algorithm sequential and batch are used for clustering with output as like as clustering visualization. The result of this research show that the highest of stock prices is on Friday and the lowest is on Wednesday and Thursday. Key words : Data mining, clustering, Self-Organizing Maps.