# US 11,451,930 B2 · Dynamic hotspot prediction method and system > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 11,451,930 B2 - Document ID: US11451930B2 - Status: Granted - Kind: B2 - Application number: 17/071,192 - Filed: 2020-10-15 - Published: 2022-04-21 - Granted: 2022-09-20 - Earliest claimed: 2020-10-15 - Assignees: Conduent Business Services, LLC - Applicants: Conduent Business Services, LLC - Inventors: Saikat Saha; Prakash Chandra; Sumit Das; Bindusree Yadamakanti; Vinuta Gayatri - Web page: https://patentgrep.com/patent/US11451930B2 ## CPC - H04W 4/029: Location-based management or tracking services - G06F 3/0481: based on specific properties of the displayed interaction object or a metaphor-based environment, e.g. interaction with desktop elements like windows or icons, or assisted by a cursor's changing behaviour or appearance - G06F 16/29: Geographical information databases - G06N 7/01: Probabilistic graphical models, e.g. probabilistic networks - G06Q 10/06315: {Needs-based resource requirements planning or analysis} - G06N 20/00: Machine learning - H04W: WIRELESS COMMUNICATION NETWORKS (broadcast communication H04H; communication systems using wireless links for non-selective communication, e.g. wireless extensions H04M1/72) - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS - 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 A method and system is presented for dynamically predicting hotspots (predictive of a high volume of event occurrences) in a geographic location. The method calculates a hotspot probability measure for each grid within this geographic location by summing probability of predicted volume bucket with probabilities of all higher buckets (buckets are ordinal here). The method also calculates a hotspot score for each grid based on the corresponding hotspot probability measure, selects a pre-defined number of grids having the highest hotspot scores as predicted hotspots, and generates an alert indicating the predicted hotspots and displaying the alert on a predictive GUI, showing the multiple grids. ## Claims ### Claim 1 A method for dynamically predicting hotspots having a high volume of crime in a geographic location and generating a corresponding alert, the method comprising: an implementation on a server that includes at least one processor for performing the steps including, collecting raw occurrence data representing event occurrences within a preselected geographical area; performing spatial aggregation to map the collected raw occurrence data to multiple grids within the pre-selected geographical area; deriving aggregated data over a time period for the grids based on the mapped raw occurrence data, the aggregated data including an aggregated event count; building a data driven volume predictive model, where volume data is grouped into a number of pre-defined buckets, such that each pre-defined bucket represents a range of volumes with a lowest pre-defined bucket representing lowest volumes to a highest pre-defined bucket representing highest volumes, and for each grid, a corresponding probability of each bucket is determined based on the aggregated event count and a predictive volume is chosen as a highest probability bucket; calculating a hotspot probability measure for each grid by summing the probability of the highest probability bucket with probabilities of all higher pre-defined buckets; calculating a hotspot score for each grid based on the corresponding hotspot probability measure and a predicted bucket weight; selecting a predetermined N number of grids having the highest hotspot score as predicted hotspot; and generating an alert indicating the predicted hotspots and displaying the alert on a predictive GUI showing the multiple grids. ### Claim 2 (depends on 1) The method of claim 1, further comprising performing the step of sorting the grids in descending order of realized crime volume for a given time window and selecting N number of top grids as true hotspots. ### Claim 3 (depends on 1) The method of claim 1, further comprising displaying an evaluation of predictive accuracy. ### Claim 4 (depends on 3) The method of claim 3, further comprising generating a validation GUI through comparison of the predicted hotspots and data collected during a prediction timeframe. ### Claim 5 (depends on 4) The method of claim 4, further comprising generating the validation GUI by creating a first display icon for accurately predicted hotspots, a second display icon for falsely predicted hotspots and a third display icon for unpredicted hotspots and displaying the icons within the pre-selected geographical area. ### Claim 6 (depends on 2) The method of claim 2, further comprising storing the generated alerts in a front database. ### Claim 7 (depends on 6) The method of claim 6, further comprising utilizing a REST API to retrieve and display the generated alerts on a hotspot dashboard of the predictive GUI. ### Claim 8 (depends on 1) The method of claim 1, wherein calculating the hotspot score comprises summing a predicted bucket weight with the hotspot probability measure. ### Claim 9 (depends on 1) The method of claim 1, further comprising collecting the raw occurrence data from a remote database using ETL steps. ### Claim 10 A system for dynamically predicting hotspots of an event in a geographic location and generating corresponding alerts, the system comprising: a communication interface configured to collect raw occurrence data from a remote database using ETL steps; and a server including at least one processor and a memory storing instructions for execution by the processor for performing steps including, performing spatial aggregation to map the collected raw occurrence data to multiple grids within the pre-selected geographical area; deriving aggregated data over a time period for the grids based on the mapped raw occurrence data, the aggregated data including an aggregated event count; implementing a data driven volume predictive model, where volume data is grouped into a number of pre-defined buckets, such that each pre-defined bucket represents a range of volumes with a lowest pre-defined bucket representing lowest volumes to a highest pre-defined bucket representing highest volumes, and for each grid, determining a corresponding probability of each bucket based on the aggregated event count and selecting a predictive volume as a highest probability bucket; calculating a hotspot probability measure for each grid by summing the probability of the highest probability bucket with probabilities of all higher pre-defined buckets; calculating a hotspot score for each grid based on the corresponding hotspot probability measure and a predicted bucket weight; selecting a predetermined N number of grids having the highest hotspot score as predicted hotspot; and generating an alert indicating the predicted hotspot and a display interface displaying the alert on a predictive GUI showing the multiple grids. ### Claim 11 (depends on 10) The system of claim 10, wherein the processor further performs the step of sorting the grids in descending order of realized event volume for a given time window and selects a predetermined N number of top grids as true hotspots. ### Claim 12 (depends on 10) The system of claim 10, further comprising providing an evaluation of predictive accuracy. ### Claim 13 (depends on 12) The system of claim 12, wherein the processor generates a validation GUI through comparison of the predicted hotspots and data collected during a prediction timeframe. ### Claim 14 (depends on 13) The system of claim 13, wherein the validation GUI comprises a first display icon representing accurately predicted hotspots, a second display icon representing falsely predicted hotspots and a third display icon for unpredicted hotspots, wherein the icons are displayed within the pre-selected geographical area. ### Claim 15 (depends on 11) The system of claim 11, further comprising a front database storing the generated alerts. ### Claim 16 (depends on 15) The system of claim 15, further comprising a REST API for retrieving and displaying the generated alerts on a hotspot dashboard of the predictive GUI. ### Claim 17 (depends on 10) The system of claim 10, wherein calculating the hotspot score comprises summing a predicted bucket weight with the hotspot probability measure. ### Claim 18 (depends on 10) The system of claim 10, wherein the predicted hotspot predicts a high volume of an event in a particular geographic location during a prediction timeframe. ### Claim 19 A non-transitory computer readable medium storing instructions for execution by a processor, wherein the instructions upon execution cause the processor to perform steps including: collecting raw occurrence data representing event occurrences within a preselected geographical area; performing spatial aggregation to map the collected raw occurrence data to multiple grids within the pre-selected geographical area; deriving aggregated data over a time period for the grids based on the mapped raw occurrence data, the aggregated data including an aggregated event count; building a data driven volume predictive model, where volume data is grouped into a number of pre-defined buckets, such that each pre-defined bucket represents a range of volumes with a lowest pre-defined bucket representing lowest volumes to a highest pre-defined bucket representing highest volumes, and for each grid, determining a corresponding probability of each bucket based on the aggregated event count and selecting the predictive volume as a highest probability bucket; calculating a hotspot probability measure for each grid by summing the probability of the highest probability bucket with probabilities of all higher pre-defined buckets; calculating a hotspot score for each grid based on the corresponding hotspot probability measure and a predicted bucket weight; selecting a predetermined N number of grids having the highest hotspot score as predicted hotspots; and generating an alert indicating the predicted hotspots and displaying the alert on a predictive GUI showing the multiple grids. ### Claim 20 (depends on 19) The computer readable medium of claim 19, wherein the instructions further cause the processor to sort the grids in descending order of realized event volume for a given time window and select a predetermined N number of top grids as true hotspots.