# US 12,182,134 B2 · Search system ranking using a natural distribution > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,182,134 B2 - Document ID: US12182134B2 - Status: Granted - Kind: B2 - Application number: 16/870,541 - Filed: 2020-05-08 - Published: 2021-08-26 - Granted: 2024-12-31 - Earliest claimed: 2020-02-21 - Assignees: eBay Inc. - Applicants: eBay Inc. - Inventors: Manojkumar Rangasamy Kannadasan; Amit Jaiswal - Web page: https://patentgrep.com/patent/US12182134B2 ## CPC - G06F 16/24578: {using ranking} - G06F 16/9536: Search customisation based on social or collaborative filtering - G06F 16/951: Indexing; Web crawling techniques - G06F 16/9538: Presentation of query results - G06F 16/3349: Reuse of stored results of previous queries - G06F 16/338: Presentation of query results - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) ## Abstract A search engine is provided that generates search results proportional to a natural distribution. Search results identified and ranked by the search engine for a search query are redistributed so that a fixed number of top ranked search results include search result attributes proportional to the natural distribution, as determined from interaction tracking of prior search results. A natural distribution index mapping prior search queries to search result attributes and a proportional representation of interactions with the search result attributes is generated. The natural distribution is determined from the proportional representation of interactions. The search engine redistributes the fixed number of top ranked search results by minimizing a gap between the ranked set of search results for the search query and the natural distribution. ## Claims ### Claim 1 A computer-implemented method, the method comprising: receiving a search query at a search engine, the search query received from a client-computing device; ranking a set of search results for the search query based on a ranker associated with the search engine; redistributing at least a portion of the ranked set of search results proportional to a natural distribution of search result attributes for a category of search result attributes, the natural distribution determined from interaction tracking of search result attributes of prior search results and comprising a relative proportion of interactions received by the search result attributes within the category of search result attributes; and providing the redistributed ranked set of search results to the client-computing device. ### Claim 2 (depends on 1) The method of claim 1, wherein at least the portion of the ranked set of search results that is redistributed comprises a fixed number of highest ranked search results. ### Claim 3 (depends on 2) The method of claim 2, wherein the redistributed ranked set of search results is provided to the client-computing device by a webpage of the search engine, the webpage configured to return a particular number of search results, wherein the fixed number of highest ranked search results is divisible by the particular number of search results configured for the webpage. ### Claim 4 (depends on 2) The method of claim 2, wherein the fixed number of highest ranked search results comprises a first search result, and wherein the first search result is removed from the fixed number by the redistributing to obtain the natural distribution for the portion of the redistributed ranked set of search results. ### Claim 5 (depends on 1) The method of claim 1, further comprising determining the natural distribution from a natural distribution index, the natural distribution index comprising a relationship mapping of search query terms to search result attributes and a proportional representation of interactions with the search result attributes. ### Claim 6 (depends on 1) The method of claim 1, wherein the redistributing is performed by minimizing a gap between the ranked set of search results for the search query and the natural distribution determined for the search query, wherein the gap comprises a difference between a proportional representation of search result attributes within the ranked set of search results and the natural distribution. ### Claim 7 (depends on 1) The method of claim 1, wherein the natural distribution is specific to a user and is determined from the interaction tracking of the prior search results that are associated with the user. ### Claim 8 (depends on 1) The method of claim 1, wherein the redistributing is performed by a blender, and wherein the blender is a modification of the ranker. ### Claim 9 One or more computer storage media storing computer-readable instructions that, when executed by a processor, cause the processor to implement a method comprising: tracking interactions of search result attributes associated with search results provided by a search engine at a client-computing device; determining a natural distribution for a category of search result attributes, the natural distribution determined from the interaction tracking by identifying a relative proportion of interactions received by the search result attributes within the category of the search result attributes; and providing a set of search results in an order determined by the natural distribution in response to receiving a search query from the client-computing device. ### Claim 10 (depends on 9) The computer storage media of claim 9, further comprising: ranking the set of search results for the search query based on a ranker associated with the search engine to provide a ranked set of search results; and redistributing at least a portion of the ranked set of search results proportional to the natural distribution, wherein the order is based on the redistributed ranked set of search results. ### Claim 11 (depends on 10) The computer storage media of claim 10, wherein at least the portion of the ranked set of search results that is redistributed comprises a fixed number of highest ranked search results. ### Claim 12 (depends on 10) The computer storage media of claim 10, wherein the redistributing further comprises minimizing a gap between the ranked set of search results for the search query and the natural distribution, the gap comprising a difference between a proportional representation of search result attributes within the ranked set of search results and the natural distribution. ### Claim 13 (depends on 10) The computer storage media of claim 10, wherein the redistributing is performed by a blender, and wherein the blender is a modification of the ranker. ### Claim 14 (depends on 9) The computer storage media of claim 9, further comprising determining the natural distribution from a natural distribution index, the natural distribution index comprising a relationship mapping of search query terms of search queries associated with the search results to search result attributes and a proportional representation of interactions with the search result attributes. ### Claim 15 (depends on 9) The computer storage media of claim 9, wherein the natural distribution is specific to a user based on the interaction tracking of the search results associated with the user. ### Claim 16 A system comprising: at least one processor; computer storage media storing computer-readable instructions that, when executed by the at least one processor, cause the at least one processor to implement a method comprising: receiving a search query at a search engine; ranking at least a portion of a set of search results for the search query proportional to a natural distribution of search result attributes for a category of search result attributes, the natural distribution determined from interaction tracking of search result attributes associated with prior search results and comprising a relative proportion of interactions received by the search result attributes within the category of search result attributes, and wherein a proportional representation of interactions with search result attributes of the prior search results identified from the interaction tracking determines the natural distribution of search result attributes for the category; and providing the ranked set of search results at the search engine in response to the search query. ### Claim 17 (depends on 16) The system of claim 16, wherein ranking at least the portion of the set of search results further comprises: ranking the set of search results for the search query using a ranker associated with the search engine; and redistributing at least the portion of the ranked set of search results proportional to the natural distribution. ### Claim 18 (depends on 17) The system of claim 17, wherein the portion of the ranked set of search results that is redistributed is a fixed number of highest ranked search results. ### Claim 19 (depends on 18) The system of claim 18, wherein the fixed number of highest ranked search results is divisible by a particular number of search results configured to be displayed by a webpage of the search engine providing the ranked set of search results. ### Claim 20 (depends on 18) The system of claim 18, wherein the fixed number of highest ranked search results comprises a first search result, and wherein the first search result is removed from the fixed number by the redistributing to obtain the natural distribution for the portion of the ranked set of search results.