The Automatic Evaluation of Website Metrics and State

Abstract

This paper focuses on studying website structural and related metrics that can be used as indicators of the complexity of the website and predict maintainability requirements. The second goal of the study is to evaluate possible correlations between structural metrics and popularity (particularly in-links) metrics. Examples of some of the structural metrics evaluated in this paper include: size, complexity, and speed of page loading.While results showed that structural metrics are not good indicators of websites’ popularity, they may influence indirectly the popularity through their impact on the performance or the usability of those websites. A method is developed to evaluate the state of the website automatically and evaluate any change in that state. The study points to certain requirements that educational or higher institutes’ websites should have. Those websites should combine somewhat conflicting requirements of: high performance, particularly web page loading and speed of transaction, reliability; current, correct and up to-date information, navigability, visibility and popularity where website information should be visible internally and externally and should be easily indexed and searched for.