Smart Review Analytics: An NLP Approach To Mobile Customer Emotion Understanding

Read full text Summarise with VIRAT
Authors: Zareena Begum, Mohammad Rahin Bushra, Kondaparthy Vijetha, Mohammad  Sadhik Pasha, Mohammad  Ameerdduin
Cited by: 0
References: 21

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Mobile reviews are one of the strongest indicators of customer satisfaction, with over 90% of smartphone users relying on reviews before purchasing a device and 72% of consumers stating that a positive review increases their trust in a brand.Manual analysis of customer reviews and ratings is inefficient, error-prone, and fails to capture nuanced emotions expressed in unstructured feedback.To address these challenges, this study introduces a Natural Language Processing (NLP)-driven framework utilizing an iPhone 14 dataset containing user reviews, titles, and ratings.The pipeline begins with NLP preprocessing and Exploratory Data Analysis (EDA) to normalize and visualize the data distribution.Following this, Efficiently Learning an Encoder that Classifies Token Replacements Accurately (ELECTRA) is applied for contextual feature extraction, ensuring efficient semantic representation of textual reviews.To mitigate class imbalance in review categories, a K-Means Synthetic Minority Over-Sampling Technique (K-Means SMOTE) is used to generate synthetic samples.Unlike existing models such as the Adaptive Boosting Classifier (ABC) and the Tao Tree Classifier (TTC), the proposed system integrates an Extra Trees Classifier (ETC) for robust and scalable classification.The framework predicts bi-variate targets: Review Title and Rating, enhancing both sentiment comprehension and rating reliability.By automating review interpretation, the system provides businesses with deeper insights into customer satisfaction, product performance, and brand perception, ultimately driving improved decision-making and enhanced customer experience.

Fields of Study & Subjects

Sentiment Analysis and Opinion Mining Topic Modeling Digital Marketing and Social Media

Publication details

Journal / source
International journal of engineering science and advanced technology.
Volume
26
Issue
3
Pages
464–479
Publication date
1 March 2026
Language
EN