In the digital age, being able to capture the sentiment of the market is a prerequisite for any rational action with respect to investments in the stock market. This has facilitated, with increasing use of online platforms, an enormous amount of daily content that is created containing sentiments which are rich but at times not easy to handle, as well. This paper addresses methods VADER and supervised machine learning algorithms such as Naive Bayes and Logistic Regression for sentiment analysis in financial news headlines to detect whether pieces of news carry positive, negative, or neutral sentiments. Word2Vec technique was used by creating word clouds for visualization about key influential terms regarding market sentiment. The datasets were implemented from DJIA and NASDAQ and performance metrics computed like accuracy, precision, recall f1 score, etc. to give more insight about how Logistic Regression outperforms other models when it came to predicting market sentiment thus making them quite useful tools serving financial analysts plus investors generally. This paper introduces a complete methodology for applying sentiment analysis in forecasting stock prices, leading to more informative and strategic decision-making.