# US 12,112,365 B2 · Probabilistic determination of compatible content > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,112,365 B2 - Document ID: US12112365B2 - Status: Granted - Kind: B2 - Application number: 17/827,359 - Filed: 2022-05-27 - Published: 2022-09-08 - Granted: 2024-10-08 - Earliest claimed: 2018-04-19 - Assignees: Pinterest, Inc. - Applicants: Pinterest, Inc. - Inventors: Brian Johnson; John Milinovich; Lance Alan Riedel; Andrew Look; Mukund Narasimhan; Yu Liu - Web page: https://patentgrep.com/patent/US12112365B2 ## CPC - G06Q 30/0201: Market modelling; Market analysis; Collecting market data - G06Q 30/0631: {Recommending goods or services} - G06F 16/335: Filtering based on additional data, e.g. user or group profiles (filtering in web context G06F16/9535, G06F16/9536) - G06F 16/355: Creation or modification of classes or clusters - G06F 16/9024: {Graphs; Linked lists (G06F16/9027 takes precedence)} - G06Q 30/0282: Rating or review of business operators or products - G06N 5/022: Knowledge engineering; Knowledge acquisition - G06N 20/00: Machine learning - G06Q 10/40: Business processes related to social networking or social networking services - G06Q: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract According to aspects of the disclosed subject matter, a taste graph comprising likely content collection nodes with corresponding likely digital content items is generated through one or more analyses of a corpus of content collections that is maintained by the online content service. As should be understood, this corpus of content collections is comprised of a plurality of curated content collections, with each content collection comprising a plurality of digital content items. With this taste graph available, as a user generates (or in response to a user generating) a content collection of digital content items, reference can be made to the taste graph to identify one or more digital content items that may be added to the content collection, where the one or more digital content items have a probabilistic likelihood of being complimentary and/or compatible with the other digital content items of the content collection. ## Claims ### Claim 1 A computer-implemented method, comprising: determining, by a computing system and in response to a user created digital content collection created by a user and maintained by the computing system, a feature vector associated with the user created digital content collection, wherein the user created digital content collection includes a first plurality of digital content items having different digital content types and the feature vector is determined based at least in part on a computer-implemented analysis of the first plurality of digital content items; accessing, by the computing system, a plurality of representative digital content collections organized as a taste graph to identify a first plurality of representative digital content collections from the plurality of representative digital content collections based at least in part on the feature vector and a plurality of representative feature vectors associated with the plurality of representative digital content collections, wherein: the plurality of representative feature vectors are determined based at least in part on corresponding computer-implemented analyses of the plurality of representative digital content collections; and the taste graph is unique to the user; determining, by the computing system, a first representative digital content collection from the plurality of representative digital content collections based at least in part on a similarity between the feature vector and a first representative feature vector associated with the first representative digital content collection, wherein the first representative digital content collection includes a second plurality of digital content items having different digital content types; and identifying, by the computing system and based at least in part on the first representative digital content collection, a first digital content item from a corpus of digital content items as a recommended digital content item, wherein the first digital content item is not included in the first plurality of digital content items. ### Claim 2 (depends on 1) The computer-implemented method of claim 1, wherein the feature vector includes a plurality of feature elements. ### Claim 3 (depends on 2) The computer-implemented method of claim 2, wherein each of the plurality of feature elements corresponds to at least one of a first feature of the user created digital content collection or a second feature of at least one of the first plurality of digital content items. ### Claim 4 (depends on 2) The computer-implemented method of claim 2, wherein at least one of the plurality of feature elements is a type-specific feature element corresponding to a digital content type. ### Claim 5 (depends on 2) The computer-implemented method of claim 2, wherein each of the plurality of feature elements includes a feature-value tuple corresponding to a feature element type and a feature element value. ### Claim 6 (depends on 5) The computer-implemented method of claim 5, wherein the feature-value tuple includes a weighting. ### Claim 7 A computing system, comprising: at least one processor; and a memory storing program instructions that, when executed by the at least one processor, cause the at least one processor to at least: determine, by the computing system, a feature vector associated with a digital content collection created by a user and maintained by the computing system, wherein the digital content collection includes a first plurality of digital content items having different digital content types and the feature vector is determined based at least in part on a computer-implemented analysis of the first plurality of digital content items; access, by the computing system, a plurality of representative digital content items organized as a taste graph to identify a first representative digital content collection from a plurality of representative digital content collections based at least in part on a similarity between the feature vector and a plurality of representative feature vectors associated with the plurality of representative digital content collections, wherein the first representative digital content collection includes a second plurality of digital content items having different digital content types, wherein; the plurality of representative feature vectors are determined based at least in part on corresponding computer-implemented analyses of the plurality of representative digital content collections; and the taste graph is unique to the user; and determine, based at least in part on the first representative digital content collection, a first digital content item from a corpus of digital content items as a likely digital content item for inclusion in the digital content collection, wherein the first digital content item is not included in the first plurality of digital content items. ### Claim 8 (depends on 7) The computing system of claim 7, wherein at least a portion of the plurality of representative digital content collections are generated by the computing system. ### Claim 9 (depends on 7) The computing system of claim 7, wherein the feature vector includes a plurality of feature elements. ### Claim 10 (depends on 9) The computing system of claim 9, wherein each of the plurality of feature elements corresponds to at least one of a first feature of the digital content collection or a second feature of at least one of the first plurality of digital content items. ### Claim 11 (depends on 10) The computing system of claim 10, wherein at least one of the plurality of feature elements is a type-specific feature element corresponding to a first digital content type. ### Claim 12 (depends on 11) The computing system of claim 11, wherein the first plurality of digital content items does not include a digital content item of the first type and the type-specific feature element includes a null value. ### Claim 13 (depends on 9) The computing system of claim 9, wherein each of the plurality of feature elements includes a feature-value tuple corresponding to a feature element type and a feature element value. ### Claim 14 A computer-implemented method, comprising: determining, by a computing system and in response to a user created digital content collection created by a user and maintained by the computing system, a feature vector associated with the user created digital content collection, wherein the user created digital content collection includes a first plurality of digital content items having different digital content types and the feature vector is determined based at least in part on a computer-implemented analysis of the first plurality of digital content items; accessing, by the computing system, a plurality of representative digital content collections organized as a taste graph to identify a first representative digital content collection from the plurality of representative digital content collections based at least in part on a similarity between the feature vector and a plurality of representative feature vectors associated with the plurality of representative digital content collections, wherein: the first representative digital content collection includes a second plurality of digital content items having different digital content types and the plurality of representative feature vectors are determined based at least in part on corresponding computer-implemented analyses of the plurality of representative digital content collections; and the taste graph is unique to the user; and determining, based at least in part on the first representative digital content collection, a first digital content item from a corpus of digital content items as a likely digital content item for inclusion in the user created digital content collection, wherein the first digital content item is not included in the first plurality of digital content items. ### Claim 15 (depends on 14) The computer-implemented method of claim 14, wherein at least a portion of the plurality of representative digital content collections are generated by the computing system. ### Claim 16 (depends on 14) The computer-implemented method of claim 14, wherein: the feature vector includes a plurality of feature elements; and each of the plurality of feature elements corresponds to a feature element type, a feature element value, and a weighting. ### Claim 17 (depends on 16) The computer-implemented method of claim 16, wherein at least one of the plurality of feature elements is a type-specific feature element corresponding to a first digital content type. ### Claim 18 (depends on 14) The computer-implemented method of claim 14, wherein the feature vector and the representative feature vectors are each implemented as a sparse array of feature elements.