Skip to main content

University of Glasgow researchers working on new search engine to answer queries Google can't

Computer scientists at the University of Glasgow are pioneering a new kind of web search engine able to answer advanced, location and person-specific queries like "how busy is the city centre?" by collating information obtained from sensors with relevant data from social networks.

The SMART project - it stands for Search Engine for Multimedia Environment Generated Content if you fancy a mouthful - is a co-operative European research initiative with partners including IBM's Haifa Research Lab and the University of London's Imperial College.

Expected to undergo live urban testing by 2014, the engine works by matching user queries with sensory information attained through cameras and microphones, and cross-referencing the real-time findings with data from social networks like Twitter. (ed : that's a "big data" search engine then)

Questions currently impossible for engines like Google to answer - like "where is my friend going to a music event this evening?" - will be answered by matching facial recognition of the person from CCTV cameras with the individual's social media activity.

The project is being built on an open-source search engine technology known as Terrier, which the university has been developing since 2004, according to Dr Iadh Ounis of the School of Computing Science.

"The SMART engine will be able to answer high-level queries by automatically identifying cameras, microphones and other sensors that can contribute to the query, then synthesising result stemming from distrusted sources in an intelligent way," Dr Ounis says.

"SMART builds upon the existing concept of ‘smart cities,' physical spaces which are covered in an array of intelligent sensors which communicate with each other and can be searched," he adds.

While the developments represent exciting progress in the realm of search engines, there will almost certainly be privacy concerns raised as the project advances and its scope and potential uses become clearer - at the very least, it seems like the SMART engine would empower civil busybodies with a powerful new snooping tool.

Researchers at Carnegie Mellon University, for example, have demonstrated the ability to identify strangers and gain sensitive personal information using a combination of facial recognition software and social media profiles, with many law enforcement agencies now using social networks as an avenue of investigation.

The status of information held by, gathered on, or attained through social media is rapidly emerging as one of the primary areas of contention in the ongoing debate over digital privacy and freedom. Recent high-profile court rulings indicate that the online world is increasingly subject to new and legal frameworks, both new and existing.

ITProPortal wants to know what you think about the SMART engine being developed: will it turn everyone with a smartphone into Big Brother? Or is a natural improvement on existing smart technologies that will give 21st century city life a further element of convenience?