We can resolve these queries by treating distance as a ranking signal, or by providing the user a way to filter within an area.<\/p>\n\n\n\n
Distance may also not be enough – we likely need to obtain additional context clues to give good results.<\/p>\n\n\n\n
Search in an Area<\/h2>\n\n\n\n
In Solr<\/a>, you can filter to items within a circle with “ If you know more about your users, you can use the polygon search to give better results.<\/p>\n\n\n\n For instance, if you know the areas your customer normally travels (e.g. work, school, and home), you could draw an artificial bounding box around this area and rank results within it higher.<\/p>\n\n\n\n You can combine a Solr\/Elasticsearch relevance score with distance by simply dividing the score by the distance. Experimentally this seems to work well, with some caveats.<\/p>\n\n\n\n The two typical formulas for measuring distance on a sphere are Haversine and Vincenty. For most search applications these will be indistinguishable<\/a>, but Haversine will use less computation resources.<\/p>\n\n\n\ngeofilt<\/strong>\"<\/code> (filter within X miles), or filter to within a square with “bbox<\/strong><\/code>“. There is also an add-on to search 2D polygons. Unsurprisingly, Elasticsearch offers similar options<\/a>. Postgres also has a robust extension for geospatial search called PostGIS.<\/a><\/p>\n\n\n\n![\"\"](\"https:\/\/www.garysieling.com\/blog\/wp-content\/uploads\/2020\/03\/Screen-Shot-2020-03-30-at-8.11.37-AM.png\")
<\/figure>\n\n\n\nUsing Distance in Ranking<\/h2>\n\n\n\n