Shanxi (China) – The Chinese government has a long standing policy of censoring its Internet population. Decisions to remove content from search engines by companies like Google, for example, have been front page news. Still, those imposed limitations have not stopped scientists there from creating new technologies, which are designed to provide the foundation for more accurate searches.
One of the biggest problems users face today when they key something in a search engine is “noise”: The returned result’s often contain a wide array of possible intended meanings based on our few words. We might have searched for “scientific discovery cap”. The search engine doesn’t know us, what we were looking for or what we were hoping to find. Did we want scientific discoveries relating to ice caps, ball caps, electric capacitors or something else? Scientists from the Taiyuan University of Technology, Liu Wei and Chen Junjie, have developed a technology which the believe will solve that problem.
The technology basically conducts a search of search engines using agents. Agents are designed to crawl through search engine content looking not just at searched keyword results, but also meta data. As their algorithms compile information, they determine which results should match a user’s query most directly. The team believes this “intelligent agent search” on something like “apple” will only return fruit related results (instead of Apple computers) when that’s what we were looking for.
How does it know? The technology developed by Wei and Junjie looks deeper than just the immediately provided search query text. Their algorithms move into the area of personal data content. In short, they examine your user profile to learn something about who you are, what you do and what you’ve looked for previously. In the apple search above, their algorithm would’ve determined you were a farmer, for example. It would’ve then assumed you were searching for fruit and only returned those results. In addition, if after performing similar searches you routinely selected “Jonathan” apple results, then in future queries those entries would receive first placement in the result sets.
On one hand, the Chinese government is censoring what their population can and cannot see on the Internet. On the other hand their scientists are developing tools which, in order to function properly, need to have free access to more personal information than was ever required before.