A Modified WASPAS Method for the Evaluation of E-Commerce Websites Based on Pythagorean Fuzzy Information

Web and business e-commerce site assessment is the act of comparing e-commerce sites based on their functionality, efficiency, or ability to meet the needs of a business or its clients. They are essential to keep up with the market changes and shift to provide better compliance with security requirements, and to optimize the interface to hold on to user attention. The modified versions of WASPAS method hybrid with Pythagorean fuzzy sets is exploited to perform an evaluation on the e-commerce websites like Amazon review dataset, Chnsenti Corp. and IMDB dataset using various performance indicators. The WASPAS model with balancing factor β = exp( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">j</i> ) demonstrates an average accuracy of 94.01% for Amazon review dataset, 98.19% for Chnsenti Corp. dataset and 98.83% in the case of IMBD dataset with a mean execution time of 806.60 sec, 787.10 sec and 809.19 sec, respectively. The proposed model is compared with reported state of art results methods SVM, SVM hybrid with unigram, SVM hybrid with unigram, SVM hybrid with unigram & grid search techniques, XGBoost and ERF-XGB methods with their accuracy of 50.40%, 80.81%, 80.81%, 90.1 and 98.2, respectively. The global fitness value and mean execution time from 50 independent runs is calculated for these datasets and perform statistical indicators to see the stability of the study. The proposed approach significantly enhanced the performance and reduced the computational complexity, making it a promising tool for real-time detection and prediction in efficiency of employees in various organizations as a tool of growth.