Correlation Between Scanpaths and Visual Elements of Web Pages

<p>eMINE scanpath algorithm was developed in order to identify common scanpaths in terms of visual elements of Web pages. It was integrated with the Vision Based Page Segmentation (VIPS) algorithm, which was improved and extended as part of eMINE project, on the ACTF plat- form. Once the areas of interests are created using the VIPS algorithm, eye tracking data can be imported to identify common scanpaths. To evaluate eMINE scanpath algorithm, an eye tracking study was conducted. This study shows that this algorithm is able to successfully identify common scanpaths. It also investigated that the gender and Web page familiarity affects the common scanpaths. This algorithm can be improved in the future by considering different techniques for the pre-processing of the data, by addressing the drawbacks of using the hierarchical structure and by taking into account the underlying cognitive processes.