Social media platforms have evolved into essential outlets for individuals to voice their opinions and emotions, with Twitter standing out as a prime source due to its concise nature. This project aims to delve into sentiment analysis on Twitter using linear classifier methods, with a focus on understanding public sentiment. Given the vast amount of unstructured textual data on social media, sentiment analysis becomes crucial for gauging user attitudes. Twitter's format, characterized by brief messages, presents unique challenges for sentiment analysis. This study has two primary objectives: to explore sentiment expressions on Twitter comprehensively and to develop and assess the performance of linear classifier methods in this context. By leveraging techniques like Logistic Regression, Support Vector Machines (SVM), and Naive Bayes, among others, this research endeavors to unravel the complexities of sentiment dynamics on Twitter. The project follows a systematic approach, beginning with the collection and preparation of a diverse dataset representative of various themes, trends, and user demographics on Twitter. A critical review of existing literature on sentiment analysis and linear classifiers informs the research methodology. Through this study, we aim to provide insights valuable not only academically but also practically, for individuals, businesses, and policymakers seeking to understand public sentiment.