No More Highlighters: Using Eye Tracking to Create Document Summaries
Have you ever been assigned an impossible amount of reading by a professor that expects nothing less than superhuman reading speed? You scan over page after page of solid text until you feel like your eyes will fall out in search of the key points that could possibly be on the quiz next class. Highlighters and those mini Post-It tabs were my favorite school supplies in college because they saved me from having to reread the thousands of pages I had been assigned over the semester when finals time came around. A team of researchers from universities in China and the US, however, have used eye tracking technology to develop every college student’s dream come true.
Imagine having a summary of all the important parts of a text automatically created after reading it. Songhua Xu, Hao Jiang, and Francis C.M. Lau have created a personalized document summarization algorithm that does just that. Using an eye-tracker to identify the words and sentences that a reader’s eyes fixate on the longest, the system can create a document summary, which is optimized according to the reader’s personal reading interests. They assume that the time a reader’s eyes spend looking at a particular word within a sentence or a sentence within a paragraph is correlated with the importance of that element to the reader. Words or sentences that receive more attention, longer gaze, are considered to have higher interest to the reader. The document summary algorithm uses information collected about the reader’s attention time on document words and sentences, and creates a personalized summary in the form of a list of sentences in descending order of interest.
At this point, the algorithm can only create a summary of a digital text the user has read all the way through. This is useful if you require a summary of a document you have already read (think cramming for exams!) for future reference. Future work proposed by this research group will explore the possibilities of generating personalized summaries of documents without the user having to read the whole text. Having personalized summaries automatically generated would be advantageous during end-of-semester crunch time and a lifesaver for those of you who suffer from chronic procrastination.
User-Oriented Document Summarization through Vision-Based Eye-Tracking
Related articles:
- Eye Tracking: The Next Stage
- Eye Tracking: The First Half
- Eye Tracking and the Architecture of Grammar
- Tips From Eye Tracking Studies on Website Design
- Eye Tracking Exploring How You Perceive Written Words
- Top 5 Eye Tracking Usability Findings
- More Eye Tracking Tips for Web Usability
- Pupil Tracking Reveals Why Color Coding Study Materials is a Good Idea
- Eye Tracking in Visual Behavior Research Gives Marketers a Competitive Edge
- Eye Tracking History: What Have You Done For Me Lately?