GRAD evaluation results

Results of the evaluation of GRAD library. A study was carried out where the participants were shown 10 sets of 4 diagrams, each generated and automatically laid out using a different tool. They were asked to grade the overall aesthetics of each diagram, as weell as perform a task and rate the difficulty of completing it. The 4 tools which were compared were PowerDesigner, MagicDraw, Papyrus and Kroki and GRAD. All used diagrams are in the archive Diagrams.zip. One folder represents data for one question. The first question's diagrams are in folder Examples 1, the second ones are in folder Examples 2 etc. The questions can be found grouped with the results in file Survey Results. The second phase of the evaluation consisted of rating the intuitiveness of a few code samples for triggering the process of laying out a diagram. GRAD was compared to JGraphX, JUNG framework and prefuse. The code samples are in file codesamples.java Finally, file background contains data of a background questionnaire.GRAD库（GRAD library）的评估结果。本研究开展了一项用户评估实验：向参与者展示10组各含4张图表的样例，每组图表均通过不同工具生成并自动完成布局。参与者需对每张图表的整体美观度进行评分，同时完成一项任务并对任务完成难度进行评级。本次对比的工具共5款，分别为PowerDesigner、MagicDraw、Papyrus、Kroki以及GRAD。所有所用图表均存储于压缩包Diagrams.zip中，每个文件夹对应一个问题的相关数据：第一个问题的图表位于Examples 1文件夹，第二个问题的图表位于Examples 2文件夹，依此类推。问题与对应的评分结果均汇总于文件Survey Results中。 评估的第二阶段为对若干用于触发图表布局流程的代码样例的直观性进行评级，本次对比对象包括GRAD、JGraphX、JUNG框架以及prefuse。相关代码样例存储于文件codesamples.java中。最后，文件background包含背景调查问卷的相关数据。