Company Permissibility for Business Activities using NLP

Abstract

As the amount of web pages available on the visible net has increased to billions of pages with trillions of pages available from the invisible net, extracting data from the internet has become a highly essential procedure in the last few years. Several tools and protocols are made available which can extract this information, and these tools have come in great demand in recent times as researchers and surfers want to discover new knowledge at an ever-increasing rate. Investing in a company requires a lot of analysis to be done before plunging in. As part of this process, the analyst manually does a search of a company name in google, reads search results to determine if there are articles negatively impacting the company and based on this research recommends the company of investment. With enormous data available on the interest, this becomes a very tedious process. In this project, techniques like web scrapping and sentiment analysis are being applied. Libraries like beautifulsoup4, scrapy are being used for scraping the data. The web scraping process is automated using RPA (Robotic Process Automation) making the web scraping process easy as well as efficient. Another technique used is the sentiment approach that applies text analysis algorithms, natural language processing (NLP), and statistics to interpret customer sentiment classifying the customer opinions into positive, negative, or neutral categories. NLTK (Natural Language Toolkit) is applied to perform sentiment analysis.