Geographic intention and modification in web search

Abstract

Web searchers signal their geographic intent by using place-names in search queries. They also indicate their flexibility about geographic specificity by reformulating their queries. By examining this data we can learn to understand web searcher flexibility with respect to geographic intent. We examine aggregated data of queries with locations, and locations identified from IP addresses, to identify overall distance preferences, as well as distance preferences by search topic. We also examine query rewriting: both deliberate query rewriting, conducted in web search sessions, and automated query rewriting, with manual relevance judgements of geo-modified queries. We find geo-specification in 12.7% of user query rewrites in search sessions, and show the breakdown into sub-classes such as same-city, same-state, same-country and different-country. We also measure the dependence between US-state-name and distance-of-modified-location-from-original-location, finding that Vermont web searchers modify their locations greater distances than California web searchers. We find that automatically-modified queries are perceived as much more relevant when the geographic component is unchanged. We look at the relationship between the non-location part of a query and the distance from the user. We see that people search for child day-care near their locations and maps far from where they are located. We also give distance profiles for the top topics which cooccur with place-names in queries, which could be used to set document priors based on document proximity and query topic.

Do Expressive Geographic Queries Lead to Improvement in Retrieval Effectiveness?

“…Moreover, the ratio of geographic queries submitted to usual search engines is about 15 % according to studies described in [25,9,12], and may increase in the context of DLs. Consequently, we propose a brief review of GIR frameworks in the next section.…”

Classifying Documents According to Locational Relevance

“…In the context of search engine queries, Jones et al studied the relationship between the non-location part of an explicit geographic query and the distance of the query's location part from the issuer's IP location [25]. They found that geographic queries have a varied distance distribution and, therefore, different localization capabilities.…”

Hyper-local, directions-based ranking of places

“…[14,23]. Other work in this area [16] explores the distances that one is willing to travel in order to get to a particular place. This work can be used in our framework to automatically find a maximum threshold for travel distance for a given query.…”

“…In the offline phase, we compute a score for each place in our directory. In the online phase, we rank places relative to a user's query location and the distance the user is willing to travel [16]. The offline score can combine scores from several scorers, thus leveraging different information sources.…”