According to the World Health Organization (WHO), suicide is a global issue and around 800,000 people commit suicide every year. In addition to that, WHO also reports that for every person who dies as a result of suicide, there may be more than 20 who survive an attempt. Early intervention is the most efficacious method to prevent suicides, and this involves understanding and detecting suicidal ideation. With the growth and popularity of the Internet, people have started to discuss suicidal tendencies on online platforms. By analyzing the differences between the way suicidal and non-suicidal people text online, it is possible to detect texts that contain suicidal ideation. In the present work, various natural language features of online posts, along with predicted gender and age of the person, have been leveraged to detect suicidal ideation in the text. The dataset was collected from a popular social platform called Reddit. Various supervised learning algorithms have been applied and compared. The result analysis of the present study substantiates that the models proposed are very promising in predicting suicidal ideation from the text