Methods for Automatic Web Page Layout Testing and Analysis: A Review

Abstract

Methods for automatic analysis of user interfaces are essential for a wide range of applications in computer science and software engineering. These methods are used in software security, document archiving, human-computer interaction, software engineering, and data science. Even though these methods are essential, no single research systematically lists most of the methods and their characteristics. This paper aims to give an overview of different solutions and their applications in the separate processes of automatic analysis of user interfaces. The main focus is on the techniques that analyze web page layouts and web page structure. Web pages' style, type of content, and even structure constantly (often drastically) change, as do methods that analyze them. The fact that most methods use very different datasets and web pages of various complexities are some of the reasons that the direct comparison of methods is difficult, if not impossible. Another fact is that the vast applications of methods practically solve similar problems. With these facts in mind, in the paper, we surveyed relevant scientific articles, categorized them, and provided an overview of how these methods have developed over time.

Probabilistic Programming with Programmable Variational Inference

“…Due to its centrality to many applications in computer science and beyond, there has been intense interest in automating unbiased gradient estimation for objective functions expressed as expectations, yielding several frameworks for unbiasedly differentiating first-order stochastic computation graphs [31,63,73], imperative programs with discrete randomness [3], and higher-order probabilistic programs [40]. Some works have also investigated automated computation of biased gradient estimates via smoothing [30]. However, these frameworks cannot be directly applied to variational inference problems, which require differentiating not just user code, but also traced simulators and log density evaluators of the probabilistic programs that users write.…”

Finite State GUI Testing with Test Case Prioritization Using Z-BES and GK-GRU

“…To overcome these issues, significant efforts have been made to integrate machine learning (ML) techniques, including decision tree (DC), random forest (RF), support vector machine (SVM), etc., with GUI testing. This integration has the capacity to enhance testing methodologies, improve accuracy, and hasten the identification of defects, thus improving the GUI's performance [6].…”