Eye-tracking data can be used to understand how users interact with web pages. Understanding the eye-movement sequences of multiple users is a challenging task because the sequence followed by each user tends to be different. Scanpath Trend Analysis (STA) brings multiple individual eye-movement sequences together and identifies a representative sequence as a trending path. However, eye-tracking data on a web page is required to determine the trending path. Our aim here is to investigate whether we can train Machine Learning (ML) algorithms to identify trending elements on a web page without collecting eye-tracking data on that web page. This article presents our experiments with different ML classification algorithms towards achieving that goal. To validate the experiments, we used two datasets from previous research, the first one included browsing and searching tasks and the second one included browsing and synthesis tasks. Our experiments show that the k-nearest neighbors algorithm (KNN) model can successfully identify the trending elements in the first dataset for both browsing (F1=≈91%) and searching tasks (F1=≈88%). However, the second dataset’s synthesis task results were not as successful as its browsing task results. Our work here shows that a model can be created to predict the trending elements in web pages solely with web page features but the task is a critical factor in the success of prediction.

Autism Spectrum Disorder (ASD) is a neurodevelopmental disorder, often associated with social and communication challenges and whose prevalence has increased significantly over the past two decades. The variety of different manifestations of ASD makes the condition difficult to diagnose, especially in the case of highly independent adults. A large body of work is dedicated to developing new and improved diagnostic techniques, emphasising approaches that rely on objective markers. One such paradigm is investigating eye-tracking data as a promising and objective method to capture attention-related differences between people with and without autism. This study builds upon prior work in this area that focussed on developing a machine-learning classifier trained on gaze data from web-related tasks to detect ASD in adults. Using the same data, we show that a new data pre-processing approach, combined with an exploration of the performance of different classification algorithms, leads to an increased classification accuracy compared to prior work. The proposed approach to data pre-processing is stimulus-independent, suggesting that the improvements in performance shown in these experiments can potentially generalise over other studies that use eye-tracking data for predictive purposes.

Web pages are composed of different kinds of elements (menus, adverts, etc.). Segmenting pages into their elements has long been important in understanding how people experience those pages and in making those experiences “better.” Many approaches have been proposed that relate the resultant elements with the underlying source code; however, they do not consider users’ interactions. Another group of approaches analyses eye movements of users to discover areas that interest or attract them (i.e., areas of interest or AOIs). Although these approaches consider how users interact with web pages, they do not relate AOIs with the underlying source code. We propose a novel approach that integrates web page and eye tracking data driven approaches for automatic AOI detection. This approach segments an entire web page into its AOIs by considering users’ interactions and relates AOIs with the underlying source code. Based on the Adjusted Rand Index measure, our approach provides the most similar segmentation to the ground-truth segmentation compared to its individual components.

People with autism often exhibit different visual behaviours from neurotypical users. To explore how these differences are exhibited on the Web, we model visual behaviour by combining pupillary response, which is an unobtrusive measure of physiological arousal, with eye-tracking scan paths that indicate visual attention. We evaluated our approach with two populations: 19 neurotypical users and 19 users with autism. We observe differences in their visual behaviours as, in certain instances, individuals with autism exhibit a lower arousal response to affective contents. While this is consistent with the literature on autism, we confirm this phenomenon on the Web. We discuss how our modelling method can be used to identify possible UX issues such as the presence of stress, cognitive load and differences in the perception of Web elements in relation to physiological arousal.

Anecdotal evidence suggests that people with autism may have different processing strategies when accessing the web. However, limited empirical evidence is available to support this. This paper presents an eye tracking study with 18 participants with high-functioning autism and 18 neurotypical participants to investigate the similarities and differences between these two groups in terms of how they search for information within web pages. According to our analysis, people with autism are likely to be less successful in completing their searching tasks. They also have a tendency to look at more elements on web pages and make more transitions between the elements in comparison to neurotypical people. In addition, they tend to make shorter but more frequent fixations on elements which are not directly related to a given search task. Therefore, this paper presents the first empirical study to investigate how people with autism differ from neurotypical people when they search for information within web pages based on an in-depth statistical analysis of their gaze patterns.

The number of users required for usability studies has been a controversial issue over 30 years. Some researchers suggest a certain number of users to be included in these studies. However, they do not focus on eye tracking studies for analysing eye movement sequences of users (i.e., scanpaths) on web pages. We investigate the effects of the number of users on scanpath analysis with our algorithm that was designed for identifying the most commonly followed path by multiple users. Our experimental results suggest that it is possible to approximate the same results with a smaller number of users. The results also suggest that more users are required when they serendipitously browse on web pages in comparison with when they search for specific information or items. We observed that we could achieve 75% similarity to the results of 65 users with 27 users for searching tasks and 34 users for browsing tasks. This study guides researchers to determine the ideal number of users for analysing scanpaths on web pages based on their budget and time.

Eye tracking studies have widely been used in improving the design and usability of web pages and in the research of understanding how users navigate them. However, there is limited research in clustering users’ eye movement sequences (i.e., scanpaths) on web pages to identify a general direction they follow. Existing research tends to be reductionist, which means that the resulting path is so short that it is not useful. Moreover, there is little work on correlating users’ scanpaths with visual elements of web pages and the underlying source code, which means the result cannot be used for further processing. In order to address these limitations, we introduce a new concept in clustering scanpaths called Scanpath Trend Analysis (STA) that not only considers the visual elements visited by all users, but also considers the visual elements visited by the majority in any order. We present an algorithm which automatically does this trend analysis to identify a trending scanpath for multiple web users in terms of visual elements of a web page. In contrast to existing research, the STA algorithm first analyzes the most visited visual elements in given scanpaths, clusters the scanpaths by arranging these visual elements based on their overall positions in the individual scanpaths, and then constructs a trending scanpath in terms of these visual elements. This algorithm was experimentally evaluated by an eye tracking study on six web pages for two different kinds of tasks (12 cases in total). Our experimental results show that the STA algorithm generates a trending scanpath that addresses the reductionist problem of existing work by preventing the loss of commonly visited visual elements for all cases. Based on the statistical tests, the STA algorithm also generates a trending scanpath that is significantly more similar to the inputted scanpaths compared to other existing work in 10 out of 12 cases. In the remaining cases, the STA algorithm still performs significantly better than some other existing work. This algorithm contributes to behavior analysis research on the web that can be used for different purposes: for example, re-engineering web pages guided by the trending scanpath to improve users’ experience or guiding designers to improve their design.