Seaweed is one of the marine organisms that can be found in almost coastal waters of Indonesia. Brown seaweed is a group of multicellular algae that have adapted to the marine environment. This study uses morphological identification methods for brown seaweed, further facilitated by utilizing machine learning technology. The aim of this research is to compare the identification based on morphological characteristics and by machine learning. The study focused on the North Coast of Teluk Awur and the South Coast of Krakal, Java Island, as the locations for field sample collection, utilizing three stations per water area with the method of collecting images of brown seaweed. The water quality parameters were determined as supporting data of environmental condition. The results of identification with machine learning compared with manual identification gave similar results. These show that on the North Coast, the genus Sargassum was identified with a high accuracy rate of 99.11%, while on the South Coast, the genus Sargassum was identified with an accuracy rate of 99.00%, the genus Padina with an accuracy rate of 99.15%, the genus Turbinaria 98.01%, and the genus Dictyota 96.42%. The growth of brown algae in the North Coast of Teluk Awur and the South Coast of Krakal was influenced by water quality factors such as temperature, salinity, pH, dissolved oxygen, and brightness.