Real Time Search User Behavior [pdf - 6 pages], by: Bernard Jansen (Penn St. Univ), Gerry Campbell (CEO, Collecta), Matthew Gregg, (Colleca_ (Apr 10) - paper presented at CHI 2010
Real Time Search receives a very careful look in this paper on user behaviour. Some points:
+ Searchers are dealing with the trivial. "Concerning search topics, the most used terms dealt with technology, entertainment, and politics, reflecting both the temporal nature of the queries and, perhaps, an early adopter user-based."
+ Searchers repeat their query over several days. It's a current awareness tool.
Good compact description of what real time search engines do
"Therefore, real time search engines (i.e., Collecta, OneRiot, CrowdEye, etc.)
employ different methods than the crawling used for conventional Web content. Real time search engines typically use some method of polling (i.e. they accept a query, send it to one or more social media platforms, retrieve and integrate the results). In this respect, real time search engines are similar to meta-search
engines. However, given the temporal nature of the content, many real time search engines do not index or store any content themselves. Additionally, as long as the query is active, real time content continues to flow into the engine in response to the query. Therefore, there is no static search engine results page generate with a set number of links."
Source: ResourceShelf posting.
Posted by Gwen at May 10, 2010 06:06 PM