# US 12,001,439 B2 · Information service for facts extracted from differing sources on a wide area network > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,001,439 B2 - Document ID: US12001439B2 - Status: Granted - Kind: B2 - Application number: 17/320,132 - Filed: 2021-05-13 - Published: 2022-09-15 - Granted: 2024-06-04 - Earliest claimed: 2009-01-21 - Assignees: Recorded Future, Inc. - Applicants: Recorded Future, Inc. - Inventors: Staffan Truvé; Christopher Ahlberg - Web page: https://patentgrep.com/patent/US12001439B2 ## CPC - G06F 16/24578: {using ranking} - H04W 4/20: Services signaling; Auxiliary data signalling, i.e. transmitting data via a non-traffic channel - G06F 16/2477: {Temporal data queries} - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - H04W: WIRELESS COMMUNICATION NETWORKS (broadcast communication H04H; communication systems using wireless links for non-selective communication, e.g. wireless extensions H04M1/72) ## Abstract In one general aspect, a wide area network fact information service method is disclosed. This method includes storing a plurality of canonical fact entries, storing one or more fact descriptor entries for each of the canonical fact entries, and ranking the canonical fact entries relative to each other based on the descriptor entries. ## Claims ### Claim 1 A wide area network fact information service method performed in connection with software running on a processor and non-transitory storage, including: electronically storing a plurality of different machine-readable canonical fact entries in the non-transitory storage, wherein the canonical fact entries in the plurality of stored canonical fact entries each corresponds to a different corresponding occurrence and each includes an electronically stored occurrence timepoint for its corresponding occurrence, wherein the occurrence timepoints each identify a time at which its corresponding occurrence occurred, storing one or more machine-readable fact descriptor entries for each of the different machine-readable canonical fact entries in the non-transitory storage, electronically ranking the machine-readable fact descriptor entries relative to each other and storing the ranking in the non-transitory storage, receiving feedback about the stored electronically ranked ranking, adjusting the stored electronically ranked ranking of the machine-readable fact descriptor entries relative to each other based on the received feedback, and making predictions about future events based on the adjusted electronically ranked ranking of the machine-readable fact descriptor entries. ### Claim 2 (depends on 1) The method of claim 1 wherein the step of ranking is a continuous process that recalculates the rank of canonical fact entries based on new descriptor entries. ### Claim 3 (depends on 2) The method of claim 2 further including, after the step of adjusting in response to user interaction with a user interface element, electronically accessing one or more of the machine-readable fact descriptor entries from the nontransitory storage. ### Claim 4 (depends on 1) The method of claim 1 wherein the step of ranking is an iterative process that recalculates the rank of canonical fact entries based on new descriptor entries. ### Claim 5 (depends on 1) The method of claim 1 wherein the step of ranking is a parallel process that recalculates the rank of canonical fact entries based on new descriptor entries. ### Claim 6 (depends on 1) The method of claim 1 wherein the ranking is based on a predetermined ranking for entities associated with the canonical fact entries. ### Claim 7 (depends on 6) The method of claim 6 wherein the entity ranking is based on a number of descriptor entries. ### Claim 8 (depends on 6) The method of claim 6 wherein the entity ranking is based on a number of descriptor entries within a category. ### Claim 9 (depends on 6) The method of claim 6 wherein the entity ranking is based on a sentiment/attitude value. ### Claim 10 (depends on 6) The method of claim 6 wherein the entity ranking is based on co-occurrence with other facts and their rankings. ### Claim 11 (depends on 1) The method of claim 1 wherein the ranking is based on a ranking of source credibility for data source providers associated with the descriptor entries. ### Claim 12 (depends on 11) The method of claim 11 wherein a same source provider can have different credibility values for different sources that it provides. ### Claim 13 (depends on 11) The method of claim 11 wherein a same source provider can have different credibility values for different categories of sources that it provides. ### Claim 14 (depends on 11) The method of claim 11 wherein the credibility ranking is based on at least one user interest measure. ### Claim 15 (depends on 1) The method of claim 1 wherein the ranking is based on a proximity measure for the descriptor entries. ### Claim 16 (depends on 15) The method of claim 15 wherein the proximity measure is a temporal proximity measure. ### Claim 17 (depends on 15) The method of claim 15 wherein proximity measures are combined using a fading factor. ### Claim 18 (depends on 1) The method of claim 1 wherein the ranking is based on a plurality of similarity measures that express a similarity between the canonical fact entries and/or fact descriptor entries. ### Claim 19 (depends on 18) The method of claim 18 wherein the similarity measures relate to fact type and temporal overlap. ### Claim 20 (depends on 18) The method of claim 18 wherein the similarity measures operate according to a hierarchy. ### Claim 21 (depends on 1) The method of claim 1 wherein the ranking is based on publication times. ### Claim 22 (depends on 1) The method of claim 1 wherein the step of ranking is performed for sources of fact descriptors. ### Claim 23 (depends on 22) The method of claim 22 wherein at least some of the derived occurrence timepoints occur in the future. ### Claim 24 (depends on 22) The method of claim 22 wherein the occurrence timepoint for at least some of the canonical fact entries is derived from content meaning on the network. ### Claim 25 (depends on 24) The method of claim 24 wherein at least some of the derived occurrence timepoints occur in the future. ### Claim 26 (depends on 1) The method of claim 1 wherein the occurrence timepoint for at least some of the canonical fact entries is derived from content meaning on the network. ### Claim 27 (depends on 26) The method of claim 26 wherein at least some of the derived occurrence timepoints occur in the future. ### Claim 28 (depends on 1) The method of claim 1 wherein at least some of the occurrence timepoints occur in the future. ### Claim 29 (depends on 1) The method of claim 1 further including, after the step of adjusting in response to user interaction with a user interface element, electronically accessing one or more of the machine-readable fact descriptor entries from the nontransitory storage. ### Claim 30 (depends on 1) The method of claim 1 wherein the making predictions includes making predictions of dates of occurrence of future events. ### Claim 31 A wide area network fact information service system including a processor and non-transitory storage, including: non-transitory machine-readable canonical fact entry storage, wherein a plurality of different machine-readable canonical fact entries stored in the canonical fact entry storage each corresponds to a different corresponding occurrence and each includes an electronically stored occurrence timepoint for its corresponding occurrence, wherein the occurrence timepoint identifies a time at which its corresponding occurrence occurred, machine-readable fact descriptor entry storage for storing one or more fact machine-readable descriptor entries for each of the different machine-readable canonical fact entries, a electronic ranker for ranking the stored fact descriptor entries relative to each other and storing them in the non-transitory storage, wherein the ranker is responsive to received feedback about the stored electronically ranked ranking to adjust the stored electronically ranked ranking of the machine-readable fact descriptor entries relative to each other based on the received feedback, and a prediction engine responsive to the adjusted electronically ranked ranking of the stored machine-readable fact descriptor entries and operative to make predictions about future events based on the adjusted electronically ranked ranking of the stored machine-readable fact descriptor entries. ### Claim 32 A wide area network fact information service system, including a processor and non-transitory storage, including: non-transitory means for storing a plurality of different machine-readable canonical fact entries, wherein a plurality of machine-readable canonical fact entries stored in the means for storing each corresponds to a different corresponding occurrence and each includes an occurrence timepoint for its corresponding occurrence, wherein the occurrence timepoint identifies a time at which its corresponding occurrence occurred, non-transitory means for storing one or more machine-readable fact descriptor entries for each of the different machine-readable canonical fact entries, and means for electronically ranking the machine-readable fact descriptor entries relative to each other and storing the ranking, wherein the means for ranking is responsive to received feedback about the stored electronically ranked ranking to adjust the stored electronically ranked ranking of the machine-readable fact descriptor entries relative to each other based on the received feedback, and means responsive to the means for electronically ranking for making predictions about future events based on the adjusted electronically ranked ranking of the stored machine-readable fact descriptor entries.