| Sumario: | Tree testing is a technique used in the early stages of Information Architecture to design the structure and navigation of a web application. It can also be used in later stages to evaluate navigational trees or the navigation map of existing web applications. Although Tree Testing is used frequently, its use and adoption are still largely handmade, using handcraft or spreadsheets to carry it out. This makes it challenging to perform benchmarking and comparative evaluations to assess the quality of navigational content and compare different website designs. Although some tools to assist evaluators exist, they are mainly based on standard metrics, lacking advanced representation and the possibility to compare and benchmark different evaluation projects. This paper presents a systematic solution to improve Tree Testing analysis through advanced visual, statistical, and machine-learning techniques that facilitate decision-making. Our solution includes a tool created to support evaluators and participants, featuring advanced techniques to extract and represent knowledge not included in other existing tools. This facilitates decision-making in the early (analysis) and advanced (evaluation) stages of web-content development. The tool has been evaluated with target users, obtaining successful results related to usefulness, satisfaction, ease of learning, and ease of use.
|
|---|