Back to results
US 9,286,357 B1Granted
Blending content in an output
AssigneeGoogle Inc.
InventorsAlexander Sobol, Andre Rohe, Sangsoo Sung, Lucian F. Cionca
Life of this patent
Estimating the term…
Abstract
Techniques include obtaining ranges of content relevance scores for different collections of content; generating a normalized range based on the ranges of content relevance scores; and normalizing a particular range of a particular collection of content including: generating a distribution of content relevance scores for the collection of content; identifying portions in the distribution; and generating a mapping of portions from the distribution to portions in the normalized range.
Claims18
A computer-implemented method comprising:
- performing searches of different collections of content to identify content in at least first and second collections in the different collections that is relevant to a search query, with the first collection having a first range of content relevance scores that differs from a second range of content relevance score for the second collection;
- determining where normalized content relevance scores corresponding to the identified content occur in a normalized range of content relevance scores for the at least first and second collections, with one or more unnormalized content relevance scores corresponding to the identified content being normalized based on division of a distribution of the unnormalized content relevance scores into first subsets, with a first subset including a portion of the unnormalized content relevance scores in the distribution, division of the normalized range into a plurality of second subsets, with a second subset including one or more of the normalized values, with the first subsets being mapped to the second subsets, one of the first subsets including a particular content relevance score, one of the second subsets being mapped to the one of the first subsets, and a normalized value being included in the one of the second subsets;
- ranking the identified content based, at least in part, on where the content relevance scores occur in the normalized range;
- and
- outputting at least part of the identified content based on the ranking.
2 more elements
One or more non-transitory machine-readable hardware storage devices storing instructions that are executable by one or more processing devices to perform operations comprising:
- performing searches of different collections of content to identify content in at least first and second collections in the different collections that is relevant to a search query, with the first collection having a first range of content relevance scores that differs from a second range of content relevance score for the second collection;
- determining where normalized content relevance scores corresponding to the identified content occur in a normalized range of content relevance scores for the at least first and second collections, with one or more unnormalized content relevance scores corresponding to the identified content being normalized based on division of a distribution of the unnormalized content relevance scores into first subsets, with a first subset including a portion of the unnormalized content relevance scores in the distribution, division of the normalized range into a plurality of second subsets, with a second subset including one or more of the normalized values, with the first subsets being mapped to the second subsets, one of the first subsets including a particular content relevance score, one of the second subsets being mapped to the one of the first subsets, and a normalized value being included in the one of the second subsets;
- ranking the identified content based, at least in part, on where the content relevance scores occur in the normalized range;
- and
- outputting at least part of the identified content based on the ranking.
2 more elements
A system comprising:
- one or more processing devices;
- and
- one or more non-transitory machine-readable hardware storage devices storing instructions that are executable by the one or more processing devices to perform operations comprising:
- performing searches of different collections of content to identify content in at least first and second collections in the different collections that is relevant to a search query, with the first collection having a first range of content relevance scores that differs from a second range of content relevance score for the second collection;
- determining where normalized content relevance scores corresponding to the identified content occur in a normalized range of content relevance scores for the at least first and second collections, with one or more unnormalized content relevance scores corresponding to the identified content being normalized based on division of a distribution of the unnormalized content relevance scores into first subsets, with a first subset including a portion of the unnormalized content relevance scores in the distribution, division of the normalized range into a plurality of second subsets, with a second subset including one or more of the normalized values, with the first subsets being mapped to the second subsets, one of the first subsets including a particular content relevance score, one of the second subsets being mapped to the one of the first subsets, and a normalized value being included in the one of the second subsets;
- ranking the identified content based, at least in part, on where the content relevance scores occur in the normalized range;
- and
- outputting at least part of the identified content based on the ranking.
5 more elements
Dependencies are inferred from the claim text.