Sentiment analysis using web scraping for live news data with machine learning algorithms

Abstract

The web scraping approach extracts large portions of information in a short amount of time using its definition. Further, Web Scraping offers records retrieval, newsgathering, internet monitoring, aggressive advertising and marketing, and many more. The use of the internet for scraping makes getting access to a tremendous quantity of records online, smooth and simple. Like in manufacturing electronic and automobile industries product reviews and campaigning require customer feedback which can be easily gathered through web scraping of websites. It is quicker and less complicated than manually extracting records from websites. Web Scraping is turning into a great data access tool these days. Apart from web scraping, web crawling and data mining or web mining are also some of the areas or methods that permit the easy compilation and storage of information on the web. In this paper, some of the famous supervised machine learning algorithms like Naïve Bayes and Logistic Regression are implemented on the live news data to effectively check the impact of these algorithms on data. Further, the application of sentiment analysis is explored and analyzed by using machine learning algorithms. This naive approach helps in understanding the comments, blogs from famous news websites by dividing the opinions into various categories at great scale and depth. This paper aims to combine the supervised learning methods with web scraping techniques for deriving optimized results of news articles about accuracy, precision, and recall.