Imagine a friend has suffered a head injury and you learn it’s a “subdural hematoma,” which you don’t know anything about. So you turn to a search engine to look for information. What do you do next? Can your computer understand your intentions by analyzing what you click on and look at on the results page?
Rutgers LIS Distinguished Professor Nicholas Belkin and Assistant Professor Chirag Shah have been awarded a $53,000 Google Research Award to investigate just that. Their study is titled “Automatic Identification of Information Searcher Intentions During an Information Seeking Session.” Belkin, as primary investigator, and Shah, as co-investigator, will be supervising PhD student Michael Cole.
The study will examine how to automatically determine what a person engaging in information seeking on the Web intends to accomplish at each moment during their information seeking episode based on their behavior. Intentions could include learning about a topic, evaluating the relevance of retrieved items, reformulating queries in response to search results, and modification of search goals.
Belkin and his colleagues will look at the relationship between low-level behaviors, such as eye fixations, mouse movements and scrolling, and high-level intentions, such as trying to learn about a topic, determining whether a document or Web page is going to be useful, or trying to understand how to best use the search engine.
“Your computer could be working to understand your current search behavior in order to support the ongoing search. If we are successful, this research will contribute to eventually writing a program that runs on the user's computer/smart-phone/mobile device,” said Cole.
As part of their research, they will analyze logs of people’s search, which include data, such as eye-fixation patterns, in order to relate sequences of low-level behaviors to what searcher tells the researchers they are trying to accomplish during the session (high-level behaviors).
They will begin their work in September and expect to complete the study by May 2014.
Belkin, Shah, and Cole are the third LIS team to receive a Google Research Grant this year. In March, former Assistant Professor Mor Naaman also won a Google Research Award to study social traces. Cole also won a Google award with former Assistant Professor Jacek Gwizdka for a project related to his current research.