# US 12,111,850 B2 · Genealogical entity resolution system and method > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,111,850 B2 - Document ID: US12111850B2 - Status: Granted - Kind: B2 - Application number: 17/715,649 - Filed: 2022-04-07 - Published: 2022-07-21 - Granted: 2024-10-08 - Earliest claimed: 2017-10-24 - Assignees: Ancestry.com Operations Inc. - Applicants: Ancestry.com Operations Inc. - Inventors: Tyler Folkman; Rey Furner - Web page: https://patentgrep.com/patent/US12111850B2 ## CPC - G06F 16/285: {Clustering or classification} - G06F 16/906: Clustering; Classification - G06F 16/2246: {Trees, e.g. B+trees} - G06F 16/288: {Entity relationship models} - G06N 20/00: Machine learning - G06F 17/11: for solving equations {, e.g. nonlinear equations, general mathematical optimization problems (optimization specially adapted for a specific administrative, business or logistic context G06Q10/04)} - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract Systems and methods for determining whether two tree persons in a genealogical database correspond to the same real-life individual. Embodiments include obtaining, from a tree database, a first tree person from a first genealogical tree and a second tree person from a second genealogical tree. Embodiments also include identifying a plurality of familial categories. Embodiments further include, for each familial category of the plurality of familial categories, extracting a first quantity of features for each of the tree persons in the familial category, generating a first similarity score for each possible pairing of tree persons, identifying a representative pairing based on a maximum first similarity score, and extracting a second quantity of features for each of the tree persons in the representative pairing. Embodiments may also include generating a second similarity score based on the second quantity of features. ## Claims ### Claim 1 A computer-implemented method for genealogical entity resolution comprising: at an online system comprising memory and one or more processors: obtaining, from a genealogical tree database, a first tree person from a first genealogical tree and a second tree person from a second genealogical tree, the first genealogical tree and the second genealogical tree each comprising a plurality of interconnected tree persons corresponding to individuals that are related to each other; identifying a familial category such that the familial category of the first genealogical tree comprises at least one tree person and the familial category of the second genealogical tree comprises a plurality of tree persons; extracting a first feature for the at least one tree person from the first genealogical tree and a corresponding first feature for each of the plurality of tree persons from the second genealogical tree; generating a plurality of similarity scores, each similarity score based on a similarity of the first feature for the at least one tree person from the first genealogical tree to the corresponding first feature for respective tree persons of the plurality of tree persons from the second genealogical tree; and identifying a representative pairing of tree persons within the familial category of the first and second genealogical trees, the representative pairing having a maximum similarity score of the plurality of similarity scores. ### Claim 2 (depends on 1) The method of claim 1, further comprising: extracting a second feature for each of the tree persons in the representative pairing; generating a second similarity score based on the second feature for each of the tree persons in the representative pairing for the familial category; determining that the first tree person and the second tree person correspond to a same individual based on the second similarity score; and modifying one or more node connections of a cluster database such that a first node of the cluster database which corresponds to the first tree person connects directly with a second node of the cluster database which corresponds to the second tree person. ### Claim 3 (depends on 1) The method of claim 1, further comprising: obtaining, from a cluster database, an additional tree person sharing a cluster with a tree person in the first genealogical tree or the second genealogical tree, wherein the cluster database comprises a plurality of nodes corresponding to tree persons and node connections therebetween and identifies groups of tree persons from different genealogical trees which correspond to a same individual; and causing the additional tree person to be included in the familial category. ### Claim 4 (depends on 1) The method of claim 1, wherein the familial category is ego. ### Claim 5 (depends on 1) The method of claim 1, wherein the familial category is one of spouse, children, father, and mother. ### Claim 6 (depends on 5) The method of claim 5, wherein the familial category is a children category, and three representative pairings of tree persons in the first and second genealogical trees are determined based on a highest three similarity scores of the plurality of similarity scores. ### Claim 7 (depends on 1) The method of claim 1, wherein the plurality of similarity scores is generated for each familial category of a plurality of familial categories including two or more of ego, spouse, father, mother, and children. ### Claim 8 (depends on 1) The method of claim 1, wherein each similarity score of the plurality of similarity scores is generated based on a weighted sum of a metric function being weighted by machine-learned feature weights. ### Claim 9 A non-transitory computer readable storage media comprising instructions that, when executed by one or more processors, cause the one or more processors to perform operations for genealogical entity resolution comprising: obtaining, from a genealogical tree database, a first tree person from a first genealogical tree and a second tree person from a second genealogical tree, the first genealogical tree and the second genealogical tree each comprising a plurality of interconnected tree persons corresponding to individuals that are related to each other; identifying a familial category such that the familial category of the first genealogical tree comprises at least one tree person and the familial category of the second genealogical tree comprises a plurality of tree persons; extracting a first feature for the at least one tree person from the first genealogical tree and a corresponding first feature for each of the plurality of tree persons from the second genealogical tree; generating a plurality of similarity scores, each similarity score based on a similarity of the first feature for the at least one tree person from the first genealogical tree to the corresponding first feature for respective tree persons of the plurality of tree persons from the second genealogical tree; and identifying a representative pairing of tree persons within the familial category of the first and second genealogical trees, the representative pairing having a maximum similarity score of the plurality of similarity scores. ### Claim 10 (depends on 9) The non-transitory computer readable storage media of claim 9, further comprising: extracting a second feature for each of the tree persons in the representative pairing; generating a second similarity score based on the second feature for each of the tree persons in the representative pairing for the familial category; determining that the first tree person and the second tree person correspond to a same individual based on the second similarity score; and modifying one or more node connections of a cluster database such that a first node of the cluster database which corresponds to the first tree person connects directly with a second node of the cluster database which corresponds to the second tree person. ### Claim 11 (depends on 10) The non-transitory computer readable storage media of claim 10, wherein the first feature comprises a first plurality of features and the second feature comprises a second plurality of features, the second plurality of features being greater than the first plurality of features. ### Claim 12 (depends on 9) The non-transitory computer readable storage media of claim 9, further comprising: obtaining, from a cluster database, an additional tree person sharing a cluster with a tree person in the first genealogical tree or the second genealogical tree, wherein the cluster database comprises a plurality of nodes corresponding to tree persons and node connections therebetween and identifies groups of tree persons from different genealogical trees which correspond to a same individual; and causing the additional tree person to be included in the familial category. ### Claim 13 (depends on 9) The non-transitory computer readable storage media of claim 9, wherein the first similarity score is generated based on a weighted sum of a metric function being weighted by feature weights. ### Claim 14 (depends on 13) The non-transitory computer readable storage media of claim 13, wherein the metric function comprises a metric between 0 and 1 and determined using one of Jaro-Winkler distance or cosine similarity. ### Claim 15 A genealogical entity-resolution system comprising: one or more processors; one or more computer readable storage media comprising instructions that, when executed by the one or more processors, cause the one or more processor to perform operations including: obtaining, from a genealogical tree database, a first tree person from a first genealogical tree and a second tree person from a second genealogical tree, the first genealogical tree and the second genealogical tree each comprising a plurality of interconnected tree persons corresponding to individuals that are related to each other; identifying a familial category such that the familial category of the first genealogical tree comprises at least one tree person and the familial category of the second genealogical tree comprises a plurality of tree persons; extracting a first feature for the at least one tree person from the first genealogical tree and a corresponding first feature for each of the plurality of tree persons from the second genealogical tree; generating a plurality of similarity scores, each similarity score based on a similarity of the first feature for the at least one tree person from the first genealogical tree to the corresponding first feature for respective tree persons of the plurality of tree persons from the second genealogical tree; and identifying a representative pairing of tree persons within the familial category of the first and second genealogical trees, the representative pairing having a maximum similarity score of the plurality of similarity scores. ### Claim 16 (depends on 15) The genealogical entity-resolution system of claim 15, further comprising: extracting a second feature for each of the tree persons in the representative pairing; generating a second similarity score based on the second feature for each of the tree persons in the representative pairing for the familial category; determining that the first tree person and the second tree person correspond to a same individual based on the second similarity score; and modifying one or more node connections of a cluster database such that a first node of the cluster database which corresponds to the first tree person connects directly with a second node of the cluster database which corresponds to the second tree person. ### Claim 17 (depends on 16) The genealogical entity-resolution system of claim 16, wherein the second feature is a plurality of features greater than 50. ### Claim 18 (depends on 16) The genealogical entity-resolution system of claim 16, wherein determining that the first tree person and the second tree person correspond to a same individual comprises comparing the second similarity score to a predetermined threshold. ### Claim 19 (depends on 15) The genealogical entity-resolution system of claim 15, wherein the first similarity is score is generated based on a weighted sum of a metric function being weighted by machine-learning generated feature weights. ### Claim 20 (depends on 15) The genealogical entity-resolution system of claim 15, wherein the first feature comprises a first plurality of features and the second feature comprises a second plurality of features, the second plurality of features being greater than the first plurality of features.