Pranata Kevin Surya, Gunawan Alexander A S, Gaol Ford Lumban
Mathematics Department, School of Computer Science, Bina Nusantara University, Jakarta, Indonesia 11480.
Computer Science Department, School of Computer Science, Bina Nusantara University, Jakarta, Indonesia 11480.
Procedia Comput Sci. 2023;216:319-327. doi: 10.1016/j.procs.2022.12.142. Epub 2023 Jan 10.
The global pandemic covid-19 offer buying opportunity to buy business with discounted price. This phenomenon attracts new type of investor around the world. This novice investor may aware that there is indices that is followed as benchmark. This benchmark was used as guidance, however, fact shown that some of this indices constituent fails to adapt and survive during global pandemic. Research indicates that formulation on inclusion and exclusion an index may biased. This novice investor may also be aware of so called blue chips company. However, yesterday winner may become tomorrow losers. This biased classification is done by human. This experiment intentionally to counter this practice, by using cutting edge machine learning to cluster IDX company using K-Means and DBSCAN algorithm. This experiment dataset is using KOMPAS100 fundamental indicator and it's ESG attributes. Experiment result suggesting there is five cluster in terms of fundamental and ESG in KOMPAS100.
全球新冠疫情为以折扣价收购企业提供了买入机会。这一现象吸引了全球新型投资者。这些新手投资者可能意识到有一些指数被用作基准。然而，事实表明，在全球疫情期间，其中一些指数成分未能适应并存活下来。研究表明，指数纳入和排除的制定可能存在偏差。这些新手投资者可能也知道所谓的蓝筹股公司。然而，昨日的赢家可能成为明日的输家。这种有偏差的分类是人为进行的。本实验旨在通过使用前沿的机器学习，利用K均值和DBSCAN算法对印尼证券交易所（IDX）的公司进行聚类，以对抗这种做法。本实验数据集使用了印尼综合100指数（KOMPAS100）的基本面指标及其ESG属性。实验结果表明，在KOMPAS100指数的基本面和ESG方面存在五个聚类。
