Analysis for Online Product Recommendation with recalling enhanced recurrent neural network-based sentiment
N. Kamal,V. Sathiya,D. Jayashree,Francis H. Shajin
DOI: https://doi.org/10.1007/s10115-024-02091-w
IF: 2.7
2024-04-24
Knowledge and Information Systems
Abstract:Recommendation system is used to filter the information according to the customer's satisfaction. Based on consumer reviews, this approach discovers and compares product scores, ratings, and rankings. Here, the data are obtained from Amazon Product Recommendation dataset. These data are supplied to pre-processing. Altered phase preserving dynamic range compression acts as a pre-processing method to eliminate unsolicited details in the related text. Ternary Pattern and Discrete Wavelet Transform is used to this preprocessed output. The output is fed to recalling enriched recurrent neural network classifier, which classifies the product suggestion as five categories: excellent, very good, good, bad, and very bad. The proposed approach is activated in MATLAB; its efficacy is evaluated under some evaluation metrics, like accuracy, precision, sensitivity, specificity, f-measure, execution time. The proposed technique achieves 21.34%, 22.54%, 25.23%, 29.56% and 26.56% high accuracy; 29.29%, 25.35%, 27.45%, 26.75% and 27.95% higher AUC value and 10.136%, 9.04%. 10.45%, 11.45% and 18.81% lower execution time compared with existing methods.
computer science, information systems, artificial intelligence