Real-Time News Aggregation and Sentiment Analysis Using Web Scraping and Firebase Integration

This paper presents a real-time news aggregation and sentiment analysis platform that offers users sentiment-classified news headlines. The system uses the method of web scraping for getting news headlines from online sources on an hourly basis. Then after the headlines are collected, the sentiment of each one is analyzed by the VADER that is a part of NLTK of Python. It classifies them as negative, neutral, or positive. The whole process involves the data that are sent to Firebase for storage. This technology has got the advantage of scalability and reliability. The mobile application client version is characterized with a simple navigation facility, interactive screen and sentiment tab for news of various categories. The interface of the client-side app is simple and neat, and apart from the usual login, it allows sentiment-based news tabs to IT users. The practical assessment of the sentiment analysis by means of this component is proved to have a high degree of credibility in different news categories and we observe the level of satisfaction across these categories and for a large proportion of the population. The demonstration is based not only on the high processing speed at the cost of running out of training data, learning time, or direct control over the problems to be solved, but also on the minimum distance between the output of those said methods of analysis and the user's request. Results from the interviews reveal that the system is spot-on in the way it allows people to filter the news according to the sentiment they are looking for. The result provides the system with room for further improvement, such as the use of more advanced sentiment analysis techniques and the extension of the application to add other features.