PEAS: Predicting Eye-tracking Assisted Segmentation

Date

2023-9-04

Author

Sulayfani, Abdullah Ihsan Saleh

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

240
views

0
downloads

Cite This

Different kinds of algorithms have been proposed to identify the visual elements of web pages for different purposes such as the enhancement of web accessibility, the measurement of web page visual quality and aesthetics etc. One group of these algorithms identifies the elements by analysing the source code and visual representation of web pages whereas another group discovers the attractive elements by analysing the eye movements of users. In previous work, an approach was proposed combining these two approaches to consider both the source code and visual representation of web pages and the eye movements of users on those pages. The result of the aforementioned approach can be considered as eye-tracking assisted web page segmentation. However, since the eye-tracking data collection procedure is elaborate, time-consuming, and expensive and it is not feasible to collect eye-tracking data for each individual page, we aim develop a model to predict such segmentation without requiring eye-tracking data. An example of a potential application could be smart screen readers that prioritize reading web page segments based on which segments of the page users interact with the most. In this work, we developed multiple ML models to predict eye-tracking assisted segmentation. The used algorithms and their corresponding highest mean F1-scores are: Decision Tree 53.80%, Random Forest 64.02%, K-Nearest Neighbour (KNN) 78.74%, Support Vector Machine (SVM) 60.11%, Logistic Regression 56.91%, and Perceptron 45.23%. KNN produced the highest result. The aforementioned models were tested on a dataset that consists of 13 web pages with ground truths that were collected in a previous work.

Subject Keywords

Web Page Segmentation, Eye Tracking, Machine Learning, Areas of Interest, KNN

URI

https://hdl.handle.net/11511/105584

Collections

Northern Cyprus Campus, Thesis