Enhancing Online Food Service User Experience Through Advanced Analytics and Hybrid Deep Learning for Comprehensive Evaluation

Abstract

User experience (UX) analysis of Online Food Delivery Services (OFDS) involves features like order placement efficacy, delivery tracking reliability, ease of navigation, menu visibility, and payment process simplicity. By examining these factors, OFDS offers can optimize its platforms to improve user satisfaction, streamline ordering procedures, minimize friction points, and improve customer retention. We can gain valued visions into customer opinions and preferences by connecting sentiment analysis, recommendation systems, feature extractors, and XAI platforms. Then, this information can be employed to develop the superiority of service, personalize UX, and finally develop customer fulfilment and platform victory. This paper presents a Reptile Search Algorithm with a Hybrid DL-based UX Detection (RSAHDL-UXD) approach on OFDSs. The RSAHDL-UXD approach utilizes data preprocessing and a word2vec-based word embedding process. For UX recognition, sliced multi-head self-attention slice recurrent neural network (SMH-SASRNN) methodology is employed. Finally, the hyperparameter tuning procedure was executed using RSA. To validate the upgraded performance of the RSAHDL-UXD methodology, a wide array of models was executed on manifold online food services datasets. The experimental outcomes stated that the RSAHDL-UXD model highlighted the superior accuracy of 98.57% and 93.33% on the Swiggy and Zomato datasets, respectively.

Keywords

• Online food delivery services
• deep learning
• artificial intelligence
• reptile search algorithm
• hyperparameter tuning