Financial News Sentiment Analysis for NIFTY 50 Market Direction and Return Prediction Using FinBERT and Machine Learning

Author(s)	Ms. Sowjanya Ashwath, Ms. Aarya Singh, Mr. Arun V
Country	India
Abstract	News moves markets that much is well established. But can we actually measure how much, and in which direction? This study builds a pipeline to do exactly that for India's benchmark NIFTY 50 index, using four years of financial news covering all 50 constituent companies. Starting from 99,341 raw articles collected via the GDELT and GNEWS (April 2022 to April 2026), the corpus was cleaned down to 74,558 unique articles after removing nearly 25% duplicates from news syndication. The articles were then analysed through FinBERT, a transformer model fine-tuned on financial text, to extract daily sentiment scores. A chi-square test confirms that the sentiment-return relationship is not random (Chi² = 124.36, p = 9.92× 10⁻28). Logistic Regression then predicts which way the market will move (68.5% accuracy), while Ridge Regression estimates by how much (R² = 0.235). Together, these results show that news sentiment carries real, measurable information about NIFTY 50 movements even from a single feature.
Keywords	Sentiment Analysis, FinBERT, NIFTY 50, Stock Market Prediction, Logistic Regression, Ridge Regression, Natural Language Processing, Chi-Square Test, Weekend Effect, Indian Stock Market
Published In	Volume 8, Issue 2, March-April 2026
Published On	2026-04-16

About IJFMR Fees & Payment Current Issue Publication Archive	Submit Research Paper Track Submission Status Publication Guidelines Publication Ethics Peer Review & Plagiarism	Join as a Reviewer Editors & Reviewers Reviewer Referral Program Get Reviewer Membership Certi.	Website/Journal Policies Usage Policy Content Policies Privacy Policy

Contact Us		+91-9687-828-838	editor@ijfmr.com

International Journal For Multidisciplinary Research