# US 12,111,851 B1 · Dynamic classification of time-series categorical data > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,111,851 B1 - Document ID: US12111851B1 - Status: Granted - Kind: B1 - Application number: 18/188,404 - Filed: 2023-03-22 - Published: 2024-09-26 - Granted: 2024-10-08 - Earliest claimed: 2023-03-22 - Assignees: Fujitsu Limited - Applicants: Fujitsu Limited - Inventors: Yusuke Hida - Web page: https://patentgrep.com/patent/US12111851B1 ## CPC - G06F 16/285: {Clustering or classification} - G06F 16/906: Clustering; Classification - G06F 16/24575: {using context} - G06F 16/2477: {Temporal data queries} - G06F 16/29: Geographical information databases - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) ## Abstract A method may include obtaining position data and associated time entries for multiple individuals. The method may also include, for each of the individuals, associating instances of the position data with categories of locations to generate a context vector, where the context vector includes text strings describing a context of a given individual. The method may also include, for each of the individuals, squeezing the context vector to combine consecutive categories of locations to generate a count vector, where the count vector includes the text strings and a count of the consecutive categories of locations. The method may also include classifying each of the individuals based on a comparison of the count vector to a known class of individuals, and adjusting at least one factor directing flow of multiple people in a different manner based on the classification. ## Claims ### Claim 1 A method comprising: obtaining position data and associated time entries for a plurality of individuals; for each of the plurality of individuals: associating instances of the position data with categories of locations to generate a context vector, the context vector including text strings describing a context of a given individual; and squeezing the context vector to combine consecutive categories of locations to generate a count vector, the count vector including the text strings and a count of the consecutive categories of locations; classifying each of the plurality of individuals based on a comparison of the count vector to a known class of individuals; and adjusting at least one factor directing flow of a plurality of people in a different manner based on the classification. ### Claim 2 (depends on 1) The method of claim 1, wherein the classifying each of the plurality of individuals includes comparing the count vector to a set of ground truth models associated with a set of known classes of individuals including the known class of individuals. ### Claim 3 (depends on 2) The method of claim 2, wherein the classifying each of the plurality of individuals comprises: measuring a distance between the count vector of a given individual and each of the ground truth models; and selecting the known class of individuals from the set of known classes of individuals based on the known class of individuals having a lowest absolute distance between the count vector of the given individual and the ground truth model of the known class of individuals. ### Claim 4 (depends on 1) The method of claim 1, further comprising: continuing to receive additional position data and additional associated time entries; associating instances of the additional position data with the categories of locations to generate an expanded context vector that includes both the position data and the additional position data; squeezing the expanded context vector to generate an expanded count vector; and re-classifying at least one of the plurality of individuals based on a second comparison of the expanded count vector to the known class of individuals. ### Claim 5 (depends on 4) The method of claim 4, wherein the squeezing the expanded context vector and the re-classifying are performed only after receiving a threshold amount of the additional position data. ### Claim 6 (depends on 1) The method of claim 1, wherein the context vector is normalized over time. ### Claim 7 (depends on 6) The method of claim 6, wherein the normalization over time includes selecting one of the text strings for a given unit of time. ### Claim 8 (depends on 1) The method of claim 1, wherein the categories of locations include at least one of at home, moving, and at work. ### Claim 9 (depends on 8) The method of claim 8, wherein the known class of individuals includes at least one of works from home, works a full day in an office, works morning hours in the office, and works afternoon hours in the office. ### Claim 10 (depends on 1) The method of claim 1, wherein the categories of locations include at least one of moving by train and moving by automobile. ### Claim 11 (depends on 1) The method of claim 1, wherein the at least one factor includes at least one of a routing algorithm when providing directions to the plurality of people, a quantity of buses running on a given route, a frequency of trains running on a given line, a bus fare for the given route, a train fare for the given route, and a toll on a given road. ### Claim 12 One or more non-transitory computer-readable media containing instructions that, when executed by one or more processors, cause a system to perform operations, the operations comprising: obtaining position data and associated time entries for a plurality of individuals; for each of the plurality of individuals: associating instances of the position data with categories of locations to generate a context vector, the context vector including text strings describing a context of a given individual; and squeezing the context vector to combine consecutive categories of locations to generate a count vector, the count vector including the text strings and a count of the consecutive categories of locations; classifying each of the plurality of individuals based on a comparison of the count vector to a known class of individuals; and generating instructions to adjust at least one factor directing flow of a plurality of people in a different manner based on the classification. ### Claim 13 (depends on 12) The non-transitory computer-readable media of claim 12, wherein the classifying each of the plurality of individuals includes comparing the count vector to a set of ground truth models associated with a set of known classes of individuals including the known class of individuals. ### Claim 14 (depends on 13) The non-transitory computer-readable media of claim 13, wherein the classifying each of the plurality of individuals comprises: measuring a distance between the count vector of a given individual and each of the ground truth models; and selecting the known class of individuals from the set of known classes of individuals based on the known class of individuals having a lowest absolute distance between the count vector of the given individual and the ground truth model of the known class of individuals. ### Claim 15 (depends on 12) The non-transitory computer-readable media of claim 12, wherein the operations further comprise: continuing to receive additional position data and additional associated time entries; associating instances of the additional position data with the categories of locations to generate an expanded context vector that includes both the position data and the additional position data; squeezing the expanded context vector to generate an expanded count vector; and re-classifying at least one of the plurality of individuals based on a second comparison of the expanded count vector to the known class of individuals. ### Claim 16 (depends on 15) The non-transitory computer-readable media of claim 15, wherein the squeezing the expanded context vector and the re-classifying are performed only after receiving a threshold amount of the additional position data. ### Claim 17 (depends on 12) The non-transitory computer-readable media of claim 12, wherein the context vector is normalized over time. ### Claim 18 (depends on 17) The non-transitory computer-readable media of claim 17, wherein the normalization over time includes selecting one of the text strings for a given unit of time. ### Claim 19 (depends on 12) The non-transitory computer-readable media of claim 12, wherein the categories of locations include at least one of at home, moving, and at work. ### Claim 20 (depends on 12) The non-transitory computer-readable media of claim 12, wherein the at least one factor includes at least one of a routing algorithm when providing directions to the plurality of people, a quantity of buses running on a given route, a frequency of trains running on a given line, a bus fare for the given route, a train fare for the given route, and a toll on a given road.