This paper introduces a novel methodology for analyzing the impact of news articles on stock price dynamics and forecasting market trends using data science techniques. Leveraging Python’s Beautiful Soup library, we compile a comprehensive dataset from a decade’s worth of news articles sourced from the Economic Times, encompassing 100 companies across diverse sectors and macroeconomic indicators. Daily price variations for each company are collected from Yahoo Finance, spanning a 10-day window before and after news publication. To assess the impact of news articles, we utilize metrics such as average and extreme daily returns, in conjunction with sentiment analysis employing Natural Language Processing (NLP) techniques like the VADER sentiment analysis tool. We develop and validate a predictive model using March 2024 data, achieving an overall accuracy of 85.02%. This model not only forecasts the impact of news on stock prices but also anticipates market trends; with dynamic adjustments to sentiment score ranges based on incoming news articles. In conclusion, our study highlights the efficacy of data science and NLP methodologies in creating an AI-powered model for accurately assessing news impact on stock prices and predicting market trends.