Customer preference analysis towards online shopping decisions based on optimized feature extraction

Online shopping has become an essential part of modern life, with food delivery being one of the most popular services. In this study, we investigate the factors influencing customers' preferences in online food ordering, such as easy payment, customization, and fast delivery. Analysing customer ratings and reviews is crucial in ensuring the quality of food services and promoting a healthy lifestyle. However, the trustworthiness of online reviews remains a challenge. Based on their rating, new customers can get help from online shoppers for their preferences or demands. Data are collected from customer reviews through online platforms to improve food service quality in online shopping. This work aims to predict the preferences of customers in online shopping based on optimized feature extraction using Principal Component Analysis with a Social Spider Optimization (PCA‐SSO) algorithm. Despite existing algorithms for analysing customer preferences, most may have inaccurate predictions or take time to analyse data. To overcome these challenges, this model of PCA‐SSO is proposed to predict customers' more accurate online shopping preferences. This model provides valuable implications for online shopping owners regarding customer satisfaction. Compared to the accuracy rates of KNN (78.22%), LDA (83.16%), and PCA (88.56%) algorithms, our proposed algorithm of PCA‐SSO achieved a higher accuracy of 93.54%.