Exploring the Effectiveness of NER-based Features for Sentiment Analysis

Abstract

The advent of social media platforms and blogs has given rise to a plethora of daily comments and reviews on various subjects. The practice of assessing people's thoughts, ideas, and perceptions is known as sentiment analysis. When it comes to understanding customer preferences or public opinions on goods or services across multiple domains or themes this can prove very valuable indeed for businesses, governments, and individuals Identifying the right polarity when it comes to sentiments poses numerous obstacles for sentiment analysis experts. In order to extract implicit information from texts such as opinions, likes/ dislikes, context etc machine learning algorithms like Text mining & Natural Language Processing are deployed. This article provides a holistic view on conducting this activity along with its applications & also compares various techniques used today while highlighting their strengths & weaknesses. The key takeaways include an insight into the growing field which continues face numerous challenges.