A Deep Convolutional Neural Network Model to Predict Consumer Recommendations using Online Reviews
- 11 August 2025
- journal article
- Published by World Scientific Pub Co Pte Ltd in International Journal of Information Technology & Decision Making
- Vol. 25 (03) , 1183-1203
Abstract
Detecting consumer perception using online reviews is challenging. Artificial Intelligence (AI) techniques have the potential to understand human perception. Sensing human psychology is essential for business growth and selecting products or services in emerging markets. This paper proposes a Deep Convolutional Neural Network (DCNN) model to predict recommendations using consumer-generated online reviews. It reinforces the capability of Deep Learning (DL) models to consider two aspects of online reviews, qualitative and quantitative, and their combinations to predictive recommendations. We have collected online reviews of airline passengers from Skytrax with the objective. We have implemented various Natural Language Processing (NLP) techniques to process the qualitative contents of online reviews. Furthermore, pre-processed data with ratings on different service aspects feeds the proposed DCNN model. To validate the performance of the proposed model, we have evaluated different performance evaluation parameters such as precision, F-score, recall, and accuracy. The experimental analysis demonstrates that the DCNN model outperforms traditional Machine Learning (ML) models for predictive recommendations. Our research indicates the power of online reviews in understanding consumer intentions for emerging market growth.Keywords
Funding Information
- Canadian Network for Research and Innovation in Machining Technology, Natural Sciences and Engineering Research Council of Canada (RGPIN-2020-05363)
This publication has 47 references indexed in Scilit:
- PlaNet: a robust deep convolutional neural network model for plant leaves disease recognitionMultimedia Tools and Applications, 2023
- Deep learning based computer-aided automatic prediction and grading system for diabetic retinopathyMultimedia Tools and Applications, 2023
- SpSAN: Sparse self-attentive network-based aspect-aware model for sentiment analysisJournal of Ambient Intelligence and Humanized Computing, 2021
- Radiologist-Level Two Novel and Robust Automated Computer-Aided Prediction Models for Early Detection of COVID-19 Infection from Chest X-ray ImagesArabian Journal for Science and Engineering, 2021
- An ensemble-based hotel recommender system using sentiment analysis and aspect categorization of hotel reviewsApplied Soft Computing, 2021
- Is position important? deep multi-task learning for aspect-based sentiment analysisApplied Intelligence, 2020
- Predicting users’ review helpfulness: the role of significant review and reviewer characteristicsSoft Computing, 2020
- What makes a helpful online review? A meta-analysis of review characteristicsElectronic Commerce Research, 2018
- Suspicious human activity recognition: a reviewArtificial Intelligence Review, 2017
- The Nature of Statistical Learning Theory1995