Differentiating and Shortlisting Unauthorized Extremist Reviews

  • Kollagal Sohitha Niyse, S. Vasundra


Customer feedback is the key behind firms such as Google, Amazon's success. The quality and service of marketing are increased when the customer's input on a product is analyzed. Customers and companies know the benefits and the downsides of the product through reviews from online shopping sites (like Amazon). Sentiment analysis is one and only of the NLP's key responsibilities (Natural Language Processing). In recent years, sentiment analysis has gained a lot of attention. This paper addresses one of the core difficulties of sentimental analysis: the difficulty of sentiment-polarity classification, a generic approach for sentiment polarity classification is presented. Sentiment Analytics, sometimes referred to as Opinion Mining, is a prevalent study field for the extraction of subjective information by analyzing textual data provided by people who execute the duties of Natural Language Processing (NLP). Online artifact reviews collected from Amazon.com are the data recycled in this study. Tests are being undertaken with encouraging results, both for categorizing sentences and for categorizing reviews. Finally, our future work on sentimental analysis will also be inspired. In this case study we consider whose data point review is 4 and 5 as positive, 1 and 2 as negative review and reaming reviews we simply drop it. After that we can build the machine learning models like NB, Logistic Regression and RF models.

How to Cite
Kollagal Sohitha Niyse, S. Vasundra. (2022). Differentiating and Shortlisting Unauthorized Extremist Reviews. Design Engineering, (1), 3084 - 3095. Retrieved from http://thedesignengineering.com/index.php/DE/article/view/9373