Sentimental Analysis on Pet Supplies Amazon Product Reviews
Now as most of all the services are made digitalized the data is being stored on databases or on clouds, to analyze the product familiarity in the people and to know about the level of the firm, previous data is used. The models and techniques which are discussed in the paper are used to bring an insight about the reviews given by the people who are using those products. This paper shows the way how to process the text data, gives the overall sentiment of the text reviews given by the users. This study analyzed the reviews of the amazon pet supplies data set and built the models logistic regression and multinomial Naïve Bayes to analyze and classify them as either positive or negative statements, technique like up-sampling is also used to the make the model unbiased. In the end, the accuracy of both the models are also given which varies from one data set to another.