Skip to Navigation
Boost Search Results
-
When querying with some filtering conditions,
a basic score is calculated for each document in the results by the underlying engine. -
Providing a boost value to some fields allows you to prioritize the resulting documents.
The boost value is integrated with the basic score, making the document rank higher. -
Boosting can be achieved in the following ways:
-
At query time:
Apply a boost factor to searched terms at query time - as described in this article. -
Via index definition:
Apply a boost factor in your index definition - see this boosting indexing article.
-
-
The automatic ordering of the results by the score is configurable.
Learn more here: automatic score-based ordering -
The calculated score details of the results can be retrieved if needed.
Learn more here: get resulting score -
In this page:
Boost results - when making a full-text search
To apply boosting while running a full-text search, use the
boost() method to prioritize the preceding search() results.
employees = list(
session
# Make a dynamic full-text search Query on 'Employees' collection
.query(object_type=Employee)
# This search predicate will use the default boost value of 1
.search("Notes", "English")
# This search predicate will use a boost value of 10
.search("Notes", "Italian")
# Call 'boost' to set the boost value to previous 'search' call
.boost(10)
)
# * Results will contain all Employee documents that have
# EITHER 'English' OR 'Italian' in their 'Notes' field.
#
# * Matching documents with 'Italian' will be listed FIRST in the results,
# before those with 'English'.
#
# * Search is case-insensitive.from "Employees" where
search(Notes, "English") or boost(search(Notes, "Italian"), 10)
Boost results - when querying with where clause
boost() can be used to give different priorities to the results
returned by different where clauses.
companies = list(
session.advanced
# Make a dynamic DocumentQuery on 'Companies' collection
.document_query(object_type=Company)
# Define a 'where' condition
.where_starts_with("Name", "O")
# Call 'boost' to set the boost value of the previous 'where' predicate
.boost(10)
# Call 'or_else' so that OR operator will be used between statements
.or_else()
.where_starts_with("Name", "P")
.boost(50)
.or_else()
.where_ends_with("Name", "OP")
.boost(90)
)
# * Results will contain all Company documents that either
# (start-with 'O') OR (start-with 'P') OR (end-with 'OP') in their 'Name' field.
#
# * Matching documents the end-with 'OP' will be listed FIRST.
# Matching documents that start-with 'P' will then be listed.
# Matching documents that start-with 'O' will be listed LAST.
#
# * Search is case-insensitive.from "Companies" where
boost(startsWith(Name, "O"), 10) or
boost(startsWith(Name, "P"), 50) or
boost(endsWith(Name, "OP"), 90)
Related Articles