KOMPARASI PERGERAKAN SAHAM DUNIA MENGGUNAKAN ALGORITMA SUPPORT VECTOR MACHINE (SVM) DAN NAÏVE BAYES PASCA PANDEMI (STUDI KASUS XIAOMI DAN SAMSUNG)

With the support of industry trends and policies, the internet has become a necessity for the world's information technology industry. Many companies hope to build an internet ecosystem to achieve rapid expansion and growth. This study selects Xiaomi and Samsung which have the largest consumer-level LOT platforms in the world. Each of these two companies has its own special points or advantages, but competes in every detail of the market. The purpose of this research is to compare stock price movements between xiaomi and samsung companies. There are two methods that we use, namely support vector machine (SVM) and naïve bayes. This study uses historical data for each company from 2018 to 2023. This data is used to study patterns that can ultimately predict stock price movements of each company. From the results of this study with the help of Orange software it can be concluded that the comparative results of the SVM and naïve bayes algorithms, the accuracy of SVM xiaomi is 0.454 and SVM Samsung is 0.438. While the accuracy of naïve bayes xiaomi is 0.659 and naïve bayes Samsung is 0.663. Precission and recall for SVM xiaomi are 0.463 and 0.482 and for Samsung are
0.049 and 0.425. While precission and recall for naïve bayes xiaomi are 0.645 and 0.642 and for Samsung are 0.648 and 0.648.

Keyword : Stock, Support vector machine