In the current financial domain, the impact of financial news on the stock market has become increasingly significant. Its rapid dissemination speed and wide coverage have made it a key factor in influencing investor sentiment and decision-making. Therefore, this study aims to explore how to utilize Natural Language Processing (NLP) technologies to analyze the impact of financial news on the stock market. The research involves collecting multiple sources of financial news and corresponding stock market data, and employs methods such as text preprocessing, sentiment analysis, and event extraction to process and analyze the news content. By constructing machine learning models, this study attempts to quantify the sentiment orientation of financial news and the impact of specific events on stock market volatility. The research questions focus on how to effectively extract valuable information from financial news and predict its specific impact on the stock market. The results indicate that sentiment orientation and key events in financial news are significantly correlated with short-term fluctuations in the stock market, especially for news reports on specific industries or companies. Moreover, the application of deep learning models further enhances the accuracy of predicting stock market reactions. This study not only provides financial market analysts with a new analytical tool but also offers a new perspective on understanding how financial news affects the stock market.