# US 2018/0025045 A1 · Accuracy of low confidence matches of user identifying information of an online system > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2018/0025045 A1 - Document ID: US20180025045A1 - Status: Published application - Kind: A1 - Application number: 15/215,209 - Filed: 2016-07-20 - Published: 2018-01-25 - Earliest claimed: 2016-07-20 - Assignees: Facebook, Inc. - Applicants: Facebook, Inc. - Inventors: Rituraj Kirti - Web page: https://patentgrep.com/patent/US20180025045A1 ## CPC - H04L 67/535: {Tracking the activity of the user (network monitoring arrangements H04L43/00; recording of computer activity G06F11/34)} - G06F 16/2365: {Ensuring data consistency and integrity} - G06F 16/35: Clustering; Classification - G06F 16/9035: Filtering based on additional data, e.g. user or group profiles - G06F 16/9535: Search customisation based on user profiles and personalisation - G06Q 10/10: Office automation; Time management - G06Q 10/40: Business processes related to social networking or social networking services - G06Q 30/02: Marketing; Price estimation or determination; Fundraising - G06Q 30/0201: Market modelling; Market analysis; Collecting market data - G06Q 30/0241: Advertisements - G06Q 30/0269: {based on user profile or attribute} - G06Q 30/06: Buying, selling or leasing transactions - H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M) - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - 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 ## Abstract An online system receives third party user identifying information. The online system accesses accuracy measures associated with each of a plurality of the user identifying information sets. The online system identifies high accuracy sets of user identifying information that include the one or more types of user identifying information included in the received types of the third party user identifying information. The online system identifies as high confidence matches those local users of the online system having the high accuracy sets of user identifying information that match a corresponding set of third party user identifying information for the plurality of third party users. The online system also identifies as low confidence matches, and identifies as similar matches those of the low confidence matches that have a measure of similarity to one or more of the high confidence matches that is beyond a threshold measure of similarity. ## Claims ### Claim 1 A method comprising: receiving, at an online system, one or more types of third party user identifying information describing each of a plurality of third party users; accessing, for a plurality of user identifying information sets maintained by the online system, accuracy measures associated with each of the plurality of the user identifying information sets, where the accuracy measures are based at least in part on local users having local user identifying information in a set matching third party user identifying information in a set associated with at least one third party user; identifying high accuracy sets of user identifying information that include the one or more types of user identifying information included in the received types of the third party user identifying information and associated with accuracy measures above a threshold level; identifying as high confidence matches those local users of the online system having the high accuracy sets of user identifying information that match a corresponding set of third party user identifying information for the plurality of third party users; identifying as low confidence matches those local users of the online system having sets of user identifying information that are not a high accuracy set that match a corresponding set of third party user identifying information for the plurality of third party users; identifying as similar matches those of the low confidence matches that have a measure of similarity to one or more of the high confidence matches that is beyond a threshold measure of similarity, the measure of similarity based at least in part on one or more characteristics of the low confidence matches matching the identified one or more characteristics associated with each of the high confidence matches; and providing a first selection of the high confidence matches and a second selection of the similar matches to a third party system. ### Claim 2 (depends on 1) The method of claim 1, wherein the sets of user identifying information include combinations of one or more of: addresses, unique identifiers, network addresses, date of births, postal codes, physical addresses, geolocations, names, genders, and phone numbers. ### Claim 3 (depends on 1) The method of claim 1, wherein the identifying high accuracy sets of user identifying information further comprises: identifying the received types of user identifying information in the third party user identifying information; identifying sets of the received types of user identifying information that provide user identifying information for at least one third party user; and identifying, as the high accuracy sets of user identifying information, the sets of the received types of user identifying information that match the sets of user identifying information in the accessed data that are associated with accuracy measures above the threshold level. ### Claim 4 (depends on 1) The method of claim 1, wherein the identifying as high confidence matches further comprises: identifying as high accuracy matched users those local users of the online system having sets of user identifying information that match the high accuracy sets of user identifying information; matching the user identifying information of one or more of the high accuracy matched users with the third party user identifying information; and identifying as high confidence matches those high accuracy matched users with matches to the third party user identifying information. ### Claim 5 (depends on 1) The method of claim 1, wherein the identifying as low confidence matches further comprises: identifying as low accuracy matched users those local users of the online system having sets of user identifying information that do not match the high accuracy sets of user identifying information; matching the user identifying information of one or more of the low accuracy matched users with the third party user identifying information; and identifying as low confidence matches those low accuracy matched users with matches to the third party user identifying information. ### Claim 6 (depends on 1) The method of claim 1, wherein the identifying as similar matches further comprises: determining that a set of one or more of the similar matches match the third party user identifying information for the same third party user; and filtering out similar matches from the set of one or more of the similar matches to include only one similar match with the highest measure of similarity of the similar matches in the set of one or more similar matches. ### Claim 7 (depends on 1) The method of claim 1, further comprising: determining a score for the similar matches based on the measure of similarity for each similar match; determining the score for each high confidence match based on the measure of accuracy of the high accuracy set of user identifying information used to match the high confidence user; and providing, to the third party system, a first indication of scores for each similar match and a second indication of scores for each high confidence match. ### Claim 8 (depends on 1) The method of claim 1, further comprising: determining a bid value for each similar matches, the bid value for each similar match proportional to a bid value for a high confidence match based on the measure of similarity of the similar match; and providing the bid values to the third party system. ### Claim 9 (depends on 1) The method of claim 1, wherein the high confidence matches and the similar matches are limited to those users of the online system within a particular geographic region. ### Claim 10 (depends on 1) The method of claim 1, wherein the third party user identifying information is hashed, and wherein the user identifying information of local users is matched to the third party user identifying information by hashing the user identifying information of the local users and comparing the hash to the hash of the third party user identifying information. ### Claim 11 A computer program product comprising a non-transitory computer readable storage medium having instructions encoded thereon that, when executed by a processor, cause the processor to: receive, at an online system, one or more types of third party user identifying information describing each of a plurality of third party users; access, for a plurality of user identifying information sets maintained by the online system, accuracy measures associated with each of the plurality of the user identifying information sets, where the accuracy measures are based at least in part on local users having local user identifying information in a set matching third party user identifying information in a set associated with at least one third party user; identify high accuracy sets of user identifying information that include the one or more types of user identifying information included in the received types of the third party user identifying information and associated with accuracy measures above a threshold level; identify as high confidence matches those local users of the online system having the high accuracy sets of user identifying information that match a corresponding set of third party user identifying information for the plurality of third party users; identify as low confidence matches those local users of the online system having sets of user identifying information that are not a high accuracy set that match a corresponding set of third party user identifying information for the plurality of third party users; identify as similar matches those of the low confidence matches that have a measure of similarity to one or more of the high confidence matches that is beyond a threshold measure of similarity, the measure of similarity based at least in part on one or more characteristics of the low confidence matches matching the identified one or more characteristics associated with each of the high confidence matches; and provide a first selection of the high confidence matches and a second selection of the similar matches to a third party system. ### Claim 12 (depends on 11) The computer program product of claim 11, wherein the sets of user identifying information include combinations of one or more of: addresses, unique identifiers, network addresses, date of births, postal codes, physical addresses, geolocations, names, genders, and phone numbers. ### Claim 13 (depends on 11) The computer program product of claim 11, having further instructions, that when executed by a processor, cause the processor to: identify the received types of user identifying information in the third party user identifying information; identify sets of the received types of user identifying information that provide user identifying information for at least one third party user; and identify, as the high accuracy sets of user identifying information, the sets of the received types of user identifying information that match the sets of user identifying information in the accessed data that are associated with accuracy measures above the threshold level. ### Claim 14 (depends on 11) The computer program product of claim 11, having further instructions, that when executed by a processor, cause the processor to: identify as high accuracy matched users those local users of the online system having sets of user identifying information that match the high accuracy sets of user identifying information; match the user identifying information of one or more of the high accuracy matched users with the third party user identifying information; and identify as high confidence matches those high accuracy matched users with matches to the third party user identifying information. ### Claim 15 (depends on 11) The computer program product of claim 11, having further instructions, that when executed by a processor, cause the processor to: identify as low accuracy matched users those local users of the online system having sets of user identifying information that do not match the high accuracy sets of user identifying information; match the user identifying information of one or more of the low accuracy matched users with the third party user identifying information; and identify as low confidence matches those low accuracy matched users with matches to the third party user identifying information. ### Claim 16 (depends on 11) The computer program product of claim 11, having further instructions, that when executed by a processor, cause the processor to: determine that a set of one or more of the similar matches match the third party user identifying information for the same third party user; and filter out similar matches from the set of one or more of the similar matches to include only one similar match with the highest measure of similarity of the similar matches in the set of one or more similar matches. ### Claim 17 (depends on 11) The computer program product of claim 11, having further instructions, that when executed by a processor, cause the processor to: determine a score for the similar matches based on the measure of similarity for each similar match; determine the score for each high confidence match based on the measure of accuracy of the high accuracy set of user identifying information used to match the high confidence user; and provide, to the third party system, a first indication of scores for each similar match and a second indication of scores for each high confidence match. ### Claim 18 (depends on 11) The computer program product of claim 11, having further instructions, that when executed by a processor, cause the processor to: determine a bid value for each similar matches, the bid value for each similar match proportional to a bid value for a high confidence match based on the measure of similarity of the similar match; and provide the bid values to the third party system. ### Claim 19 (depends on 11) The computer program product of claim 11, wherein the high confidence matches and the similar matches are limited to those users of the online system within a particular geographic region. ### Claim 20 (depends on 11) The computer program product of claim 11, wherein the third party user identifying information is hashed, and wherein the user identifying information of local users is matched to the third party user identifying information by hashing the user identifying information of the local users and comparing the hash to the hash of the third party user identifying information.