# US 12,111,852 B2 · Aggregation of noisy datasets into master firmographic database > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,111,852 B2 - Document ID: US12111852B2 - Status: Granted - Kind: B2 - Application number: 18/225,863 - Filed: 2023-07-25 - Published: 2023-12-07 - Granted: 2024-10-08 - Earliest claimed: 2020-06-29 - Assignees: 6Sense Insights, Inc. - Applicants: 6Sense Insights, Inc. - Inventors: Tai Vo; Nitin Vijayvargiya; Daniel Hsiung; Premal Shah; Viral Bajaria; Akshara Palakodety - Web page: https://patentgrep.com/patent/US12111852B2 ## CPC - G06F 16/285: {Clustering or classification} - G06F 16/211: {Schema design and management} - G06F 16/215: Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors - G06F 16/2272: {Management thereof} - G06F 16/24556: {Aggregation; Duplicate elimination} - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) ## Abstract Aggregation of noisy datasets into a master firmographic database. In an embodiment, firmographic records are received from a plurality of sources, and normalized into a common schema. One or more firmographic records may be cleaned by replacing a value of one or more fields in those firmographic record(s) with a value of those field(s) in another firmographic record. The firmographic records may then be clustered, and each of the clusters may be collapsed into a single conflated firmographic record based on a voting process. A master identifier may be generated for each conflated firmographic record, and the conflated firmographic records may be merged into a master firmographic database that is indexed by master identifiers. ## Claims ### Claim 1 A method comprising using at least one hardware processor to: receive data comprising a plurality of firmographic records from a plurality of sources, wherein each of the plurality of firmographic records comprises a plurality of fields; normalize the plurality of firmographic records into a common schema; clean the plurality of firmographic records by replacing a value of each of one or more of the plurality of fields in one or more of the plurality of firmographic records with a value of that field in another one of the plurality of firmographic records; cluster the plurality of firmographic records into a plurality of clusters, wherein each of the plurality of clusters comprises a subset of the plurality of firmographic records; for each of the plurality of clusters, collapse the subset of firmographic records in that cluster into a single conflated firmographic record based on a voting process within that cluster; generate a master identifier for each conflated firmographic record; and merge the conflated firmographic records into a master firmographic database, comprising a plurality of mastered firmographic records, indexed by the master identifiers. ### Claim 2 (depends on 1) The method of claim 1, further comprising using the at least one hardware processor to, after receiving the data and prior to clustering the plurality of firmographic records, filter the plurality of firmographic records according to one or more exclusion criteria. ### Claim 3 (depends on 2) The method of claim 2, wherein the one or more exclusion criteria comprise an exclusion list that comprises at least one value for each of one or more of the plurality of fields, and wherein firmographic records containing the at least one value for a respective one of the one or more fields are filtered out of the plurality of firmographic records. ### Claim 4 (depends on 1) The method of claim 1, further comprising using the at least one hardware processor to: identify a subset of the plurality of mastered firmographic records that is associated with a user; and provide the identified subset to the user. ### Claim 5 (depends on 1) The method of claim 1, wherein normalizing the plurality of firmographic records into a common schema comprises standardizing values of one or more of the plurality of fields in the plurality of firmographic records. ### Claim 6 (depends on 1) The method of claim 1, wherein cleaning the plurality of firmographic records comprises: classifying each of the plurality of firmographic records into one of a plurality of categories, wherein the plurality of categories comprises a strong category and a weak category; and, for each of one or more of the plurality of firmographic records that are classified into the weak category, replace the value of each of one or more of the plurality of fields in that firmographic record with the value of that field in one of the plurality of firmographic records that is classified into the strong category. ### Claim 7 (depends on 6) The method of claim 6, wherein the plurality of categories further comprises a neutral category, and wherein none of the values of the plurality of fields in the plurality of firmographic records that are classified into the neutral category are replaced during the cleaning. ### Claim 8 (depends on 1) The method of claim 1, wherein clustering the plurality of firmographic records into a plurality of clusters comprises, for each of a plurality of values of a key, grouping the plurality of firmographic records having that value of the key into a cluster. ### Claim 9 (depends on 8) The method of claim 8, wherein the plurality of fields comprises a domain name and a company name, and wherein, for each of the plurality of firmographic records, the value of the key for that firmographic record is determined as: when the firmographic record comprises a value for the domain name, the value for the domain name; and, when the firmographic record does not comprise a value for the domain name, a value for the company name. ### Claim 10 (depends on 1) The method of claim 1, wherein clustering the plurality of firmographic records into a plurality of clusters comprises, for each of a plurality of values of a key and each of a plurality of values of at least one of the plurality of fields, grouping the plurality of firmographic records having that value of the key and that value of the at least one field into a cluster. ### Claim 11 (depends on 10) The method of claim 10, wherein the at least one field comprises one or more of a country, state, or city. ### Claim 12 (depends on 1) The method of claim 1, wherein the plurality of fields comprises a domain name and a company name, and wherein generating the master identifier for each conflated firmographic record comprises: when the conflated firmographic record comprises a value for the domain name, generating the master identifier as a function of the value for the domain name; and, when the conflated firmographic record does not comprise a value for the domain name, generating the master identifier as a function of a value for the company name. ### Claim 13 (depends on 1) The method of claim 1, further comprising using the at least one hardware processor to, before merging the conflated firmographic records into the master firmographic database, filtering out each of one or more of the conflated firmographic records based on a reliability of the plurality of sources that contributed the subset of firmographic records that were collapsed into that conflated firmographic record. ### Claim 14 (depends on 1) The method of claim 1, wherein the plurality of sources comprise one or both of a customer relationship management system or a marketing automation platform. ### Claim 15 (depends on 1) The method of claim 1, wherein each of the plurality of firmographic records comprises a value for each of one or more attributes of a company. ### Claim 16 (depends on 1) The method of claim 1, wherein cleaning the plurality of firmographic records comprises filling in a blank value in one of the plurality of fields in at least one of the plurality of firmographic records with a non-blank value, based on a histogram of values of that one field for the plurality of firmographic records. ### Claim 17 (depends on 1) The method of claim 1, wherein the plurality of firmographic records are clustered at a plurality of levels, and wherein the plurality of levels comprise one or more of a global level, a country level, a state level, or a city level. ### Claim 18 (depends on 1) The method of claim 1, wherein, for each of the plurality of clusters, the voting process within that cluster is based both on frequency of values for the plurality of fields within the subset of firmographic records and a reliability of each of the plurality of sources from which the subset of firmographic records were received. ### Claim 19 A system comprising: at least one hardware processor; and one or more software modules that are configured to, when executed by the at least one hardware processor, receive data comprising a plurality of firmographic records from a plurality of sources, wherein each of the plurality of firmographic records comprises a plurality of fields, normalize the plurality of firmographic records into a common schema, clean the plurality of firmographic records by replacing a value of each of one or more of the plurality of fields in one or more of the plurality of firmographic records with a value of that field in another one of the plurality of firmographic records, cluster the plurality of firmographic records into a plurality of clusters, wherein each of the plurality of clusters comprises a subset of the plurality of firmographic records, for each of the plurality of clusters, collapse the subset of firmographic records in that cluster into a single conflated firmographic record based on a voting process within that cluster, generate a master identifier for each conflated firmographic record, and merge the conflated firmographic records into a master firmographic database, comprising a plurality of mastered firmographic records, based on the master identifiers. ### Claim 20 A non-transitory computer-readable medium having instructions stored therein, wherein the instructions, when executed by a processor, cause the processor to: receive data comprising a plurality of firmographic records from a plurality of sources, wherein each of the plurality of firmographic records comprises a plurality of fields; normalize the plurality of firmographic records into a common schema; clean the plurality of firmographic records by replacing a value of each of one or more of the plurality of fields in one or more of the plurality of firmographic records with a value of that field in another one of the plurality of firmographic records; cluster the plurality of firmographic records into a plurality of clusters, wherein each of the plurality of clusters comprises a subset of the plurality of firmographic records; for each of the plurality of clusters, collapse the subset of firmographic records in that cluster into a single conflated firmographic record based on a voting process within that cluster; generate a master identifier for each conflated firmographic record; and merge the conflated firmographic records into a master firmographic database, comprising a plurality of mastered firmographic records, based on the master identifiers.