• Akhil Mate
• Om Jaiswal
• Laksh Sharma
• Devang Pardeshi
• Prof. Sheetal Patangare

The fast growth of digital media has led to a massive explosion of news content that has increasingly become hard to follow, classify and decipher in real time. This paper describes an automated news crawler, news classifier, and sentiment analysis system with built-in real-time feedback notifications. The framework uses BeautifulSoup and Selenium to search the web on the static and dynamic news sources, and speech-based news is transcribed with speech recognition methods to enable the same text processing. SpaCy and NLTK are used as preprocessors, and TF-IDF and SBERT embeddings are used as feature extractors. Unsupervised clustering algorithms, including K-Means and HDBSCAN, are applied to organize the news content, which allows articles to be clustered and assigned to a specific government ministry. Transformer-based models can be fine-tuned to classify and detect sentiment, with the DistilBERT and RoBERTa models respectively providing 91.2 and 89.5 percent classification and sentiment accuracy, respectively, compared to classical models. In addition, the system will include a real-time feedback system through Gmail SMTP and Nodemailer which notifies the concerned government departments when negative news is detected. A usable interface developed on Next.js and TailwindCSS will enable stakeholders to access categorized articles, sentiment scores and use filters by ministry or language. It has been experimentally verified that fine-tuned transformer models perform much better than classical methods and are effective in contextual classification. Generally, this paper demonstrates how AI-based frameworks can make governments more responsive, facilitate the analysis of media without bias, and enable sound decisions.

News Crawling, Sentiment Analysis, Natural Language Processing (NLP),Classification,Transformer Models

"Automated News Crawling, Classification and Sentiment Analysis with Real- Time Feedback Alerts", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 12, page no.a590-a596, December-2025, Available :http://www.jetir.org/papers/JETIR2512079.pdf

"Automated News Crawling, Classification and Sentiment Analysis with Real- Time Feedback Alerts", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 12, page no. ppa590-a596, December-2025, Available at : http://www.jetir.org/papers/JETIR2512079.pdf