For centuries, ideology has been reflected in a person’s expression. The expression points out the bias or support the person holds. Nowadays, expressions are well seen on social media in the form of text. X (Formerly Twitter) has become the favoured medium for these expressions. Nepal, a politically highly influenced country where political changes have been frequent in a short period, has people’s thoughts expressed on social media. This paper presents a novel approach to finding a person’s political inclination through their Nepali tweet using machine learning techniques. By leveraging data pre-processing and XGBoost, we achieve a promising accuracy of 73%. We also discuss potential avenues for further improving accuracy, such as expanding the dataset to include other social media platforms and enhancing data pre-processing techniques.