IIARD International Institute of Academic Research and Development

International Journal of Engineering and Modern Technology (IJEMT )

E-ISSN 2504-8848
P-ISSN 2695-2149
VOL. 9 NO. 1 2023

---

Sentiment Analysis for E-Commerce Product Reviews

Dr. Agbakwuru Alphonsus Onyekachi, Duru Esther Chidalu

---

Abstract

Due to the rise of e-commerce, online reviews have become an important factor for customers when making purchasing decisions. Consequently, it is crucial for online retailers to comprehend customer views towards their products and services. To accomplish this, natural language processing software will be utilized to analyze the text of online reviews and classify them based on their sentiment. The resulting sentiment values will then be used to generate an overall score, which can be utilized to evaluate customer sentiment toward a specific product or service. After dividing the dataset into an 80:20 training and testing ratio, positive attitudes accounted for 50.56% of the total data, while negative sentiments accounted for 49.44%, with the machine learning algorithms being trained on the training dataset. To categorize the customer reviews dataset into positive and negative sentiments, a supervised machine learning algorithm called Random Forest classifier was utilized along with other evaluation metrics such as precision, recall, F1-score, and balanced accuracy. The model evaluation showed that the classifier's recall, precision, and f1-score were 98 %, 98%, and 98% respectively. Python programming language was used for this work, along with its libraries such as pandas, NumPy, Seaborn, and Matplotlib for data analysis and visualization, and Natural Language Tool Kit packages. This research will result in a sentiment analysis model that can accurately classify customer reviews based on sentiment and generate an overall score that represents customer sentiment toward a specific product or service

keywords:

Sentiment analysis (SA), E-commerce, Machine learning, Decision Tree, Natural Language, Evaluation metrics.

---

References:

Bhargava, M. G., & Rao, D. R. (2018). Sentimental analysis on social media data using R
programming. International Journal of Engineering & Technology, 7(2.31), 80-
84.

Bird, Steven, Ewan Klein, and Edward Loper. Natural language processing with Python 2009.
Greener, J. G., Kandathil, S. M., Moffat, L., & Jones, D. T. (2022). A guide to machine

DOWNLOAD PDF

---

Back