A Hybrid Data-Driven Web-Based UI-UX Assessment Model

TL;DR

This paper proposes a hybrid, web-based UI-UX assessment model combining multiple algorithms like AHP, sentiment analysis, K-means, and XAI to evaluate and enhance web UI quality across various applications.

Contribution

It introduces a comprehensive, hybrid framework integrating decision-making, sentiment analysis, clustering, and explainable AI for web UI evaluation and improvement.

Findings

Developed a multi-layered assessment framework for web UI quality.

Integrated diverse algorithms for comprehensive UI evaluation.

Facilitated targeted UI improvements based on analysis results.

Abstract

Today, a large proportion of end user information systems have their Graphical User Interfaces (GUI) built with web-based technology (JavaScript, CSS, and HTML). Some of these web-based systems include: Internet of Things (IOT), Infotainment (in vehicles), Interactive Display Screens (for digital menu boards, information kiosks, digital signage displays at bus stops or airports, bank ATMs, etc.), and web applications/services (on smart devices). As such, web-based UI must be evaluated in order to improve upon its ability to perform the technical task for which it was designed. This study develops a framework and a processes for evaluating and improving the quality of web-based user interface (UI) as well as at a stratified level. The study develops a comprehensive framework which is a conglomeration of algorithms such as the multi-criteria decision making method of analytical hierarchy process (AHP) in coefficient generation, sentiment analysis, K-means clustering algorithms and explainable AI (XAI).