This paper examines how sentiment analysis of social media texts can help in the betterment of various fields. It compares various research works and techniques used for analysis of sentiment. An analysis of the present methods for sentiment analysis has been done and they include natural language processing, deep learning models and transformer models. A machine learning based approach has also been implemented as an example of sentiment analysis models. The research uses a tweets dataset taken from Kaggle, this dataset contains the tweets along with their geographical location, date and time. A python library and an natural language processing technique has been used to calculate the polarity and feature extraction. Two models have been implemented to analyze the text data, and the data has been classified into three categories positive neutral and negative. The research also looks into how sentiments change over time of the day or events occurring around and also based on geography. The results have been visualized with the help of Bar charts and Line graphs. Overall the research looks into how sentiment analysis can be used to get an idea of the public perception and can be used for many fields like monitoring the brands, analysis of politics and how the sentiment changes based on factors like time of the day or external events. This research contributes to the field by offering a scalable framework for real-time sentiment tracking and helps organizations make data-driven decisions in dynamic social media environments.