INTERNATIONAL JOURNAL OF SCIENTIFIC DEVELOPMENT AND RESEARCH International Peer Reviewed & Refereed Journals, Open Access Journal ISSN Approved Journal No: 2455-2631 | Impact factor: 8.15 | ESTD Year: 2016
open access , Peer-reviewed, and Refereed Journals, Impact factor 8.15
We have considered a data set from Kaggle Amazon Reviews Dataset. This data set contains data about the customer reviews of the amazon products. The data set is a .csv file containing reviewtext, review ratings, of the customers along with their review id, review province etc, and their coordinates. We have done all the preprocessing required on the data set, added a column sentiment which decides the positive or negative review to the dataset based on review rating and also we used nltk packages to text cleaning and removing unwanted words for deciding the reviews(stopwords). After the preprocessing phase we used bag of words strategy to convert the text content in reviews to numerical feature vectors, for that we used SciKit-Learn's CountVectorizer and also we use TfidfTransformer to overshadow the words which has lower average counts with same frequencies ,which has very little meaning We split the train and test data and train using machine learning models after defining the feature vector values with CountVectorizer and TfidfTransformer. To determine the model performance, use Naive Bayes analysis, Support Vector Machines, and Logistic Regression, and print the confusion matrix and accuracy
Keywords:
Cite Article:
"Analysis of product reviews using sentimental analysis", International Journal of Science & Engineering Development Research (www.ijsdr.org), ISSN:2455-2631, Vol.8, Issue 4, page no.1088 - 1092, April-2023, Available :http://www.ijsdr.org/papers/IJSDR2304180.pdf
Downloads:
000337212
Publication Details:
Published Paper ID: IJSDR2304180
Registration ID:205299
Published In: Volume 8 Issue 4, April-2023
DOI (Digital Object Identifier):
Page No: 1088 - 1092
Publisher: IJSDR | www.ijsdr.org
ISSN Number: 2455-2631
Facebook Twitter Instagram LinkedIn