The number of Internet users has grown dramatically in recent years, and as a result, communication between two or more people has become a relatively easy thing via the use of the Internet. In contrast to the early phases of social media, the most of them today support a wide range of languages spoken all over the globe thanks to the use of Unicode encoding [1]. As a result, many individuals choose to interact on social media platforms in their own language rather than utilizing international languages such as English or Spanish. As a result of the incorporation of languages such as Sinhala, Tamil, and Hindi onto the Internet, individuals who do not have a strong command of English are more likely to participate in social media and blogs. Users may express themselves freely and anonymously on a wide range of online communication platforms, including social media, which are widely available. However, generating and propagating hatred towards another group is a violation of one's right to free expression [2], which should be respected at all times. There have been numerous research efforts on ways to control such hate speech. The study of how to control hate speech on social media, especially with the help of areas of artificial intelligence such as machine learning and deep learning, is taking place from various angles. However, there has been a lot of research on social media about English language hate speech. As the rate at which individuals share ideas in their mother tongues on social media other than English is high, mechanisms to control hate speech shared in different languages are needed. Based on them, the study of hate speech shared in Sinhala, the most widely spoken language in Sri Lanka, is seen at an early stage. So, this study sets out to effectively detect Sinhala language online hate speeches through machine learning algorithms. The main objective of this research is to analyze how effectively the machine learning algorithm detect hate speech in the Sinhala language. Sub objectives are to find out how each machine learning algorithm reacts to hate speech in terms of accuracy, precision, recall and f1 score and to conclude which algorithm works best in detecting hate speech in Sinhala.