Machine learning artificial intelligence system for predicting popular hours

Abstract

A system for generating a graphical user interface in a client device. The system may include a processor in communication with the client device and a database. The processor may execute: receiving a request for occupancy information of a specified merchant; obtaining a plurality of credit card authorizations associated with the merchant; generating a posted transaction array based on the credit card authorizations; removing outlier members of the posted transaction array by applying a threshold filter; generating a transaction frequency array based on the posted transaction array, the transaction frequency array comprising weekdays and aggregated transactions associated with the weekdays; modifying the transaction frequency array by applying a transformation to the aggregated transactions; generating a smoothed array by applying a kernel density estimate to the transaction frequency array; and generating a graphical user interface displaying information in the smoothed array.

1 .- 20 . (canceled) 
     
     
         21 . A system for generating a graphical user interface in a client device comprising:
 at least one processor in communication with the client device and a database; and   a storage medium storing instructions that configure the at least one processor to execute operations comprising:
 receiving, from the client device, a request for occupancy information of a specified merchant; 
 obtaining, from the database in response to the request, a plurality of credit card proximity transactions associated with the merchant, the proximity transactions comprising at least one of NFC, chip-read, or swiped transactions; 
 generating a posted transaction array based on the credit card proximity transactions, the posted transaction array comprising a plurality of time intervals and numbers of transactions for the time intervals; 
 generating a transaction frequency array based on the posted transaction array, the transaction frequency array comprising weekdays and aggregated transactions associated with the weekdays, wherein generating the transaction frequency array comprises modifying the time intervals of the posted transaction array by subtracting a processing time; 
 generating a smoothed array by applying a kernel density estimate to the transaction frequency array; and 
 generating a graphical user interface displaying information in the smoothed array. 
   
     
     
         22 . The system of  claim 21 , the operations further comprising modifying the transaction frequency array though minmax normalization. 
     
     
         23 . The system of  claim 21 , wherein generating the transaction frequency array comprises:
 determining an aggregation window comprising an initial date and a final date, the initial date being a predetermined number of days before a current day, and the final date being the current day; and   eliminating, from the posted transaction array, transactions outside the aggregation window.   
     
     
         24 . The system of  claim 21 , wherein the kernel density estimate comprises at least one of a Gaussian kernel or an Epanechnikov kernel. 
     
     
         25 . The system of  claim 24 , wherein the kernel density estimate is an Epanechnikov kernel with a bandwidth of 2. 
     
     
         26 . The system of  claim 21 , wherein:
 the graphical user interface comprises a first graphical user interface displaying a weekday selection option; and   the operations further comprise:
 receiving a day selection from the client device; and 
 generating a second graphical user interface displaying a histogram corresponding to the day selection. 
   
     
     
         27 . The system of  claim 26 , wherein the first graphical user interface displays a hyperlink to a merchant website. 
     
     
         28 . The system of  claim 26 , wherein the second graphical user interface highlights a peak popular hour in the histogram by at least one of a different color or a different shape. 
     
     
         29 . The system of  claim 21 , wherein generating the transaction frequency array comprises:
 determining predicted hours of operation based on the posted transaction array; and   eliminating time intervals outside the predicted hours of operation from the posted transaction array.   
     
     
         30 . The system of  claim 29 , wherein: the graphical user interface displays a histogram in a first window; and the graphical user interface displays the predicted hours of operation in a second window different from the first window. 
     
     
         31 . The system of  claim 29 , wherein determining hours of operation comprises:
 determining a plurality of model results for time intervals in the posted transaction array using a plurality of prediction models, the prediction models outputting corresponding model results; and   determining prediction indications for the time intervals in the posted transaction array by tallying the model results.   
     
     
         32 . The system of  claim 31 , wherein the prediction models comprise decision tree models configured for random forest analysis. 
     
     
         33 . The system of  claim 21 , wherein generating the posted transaction array comprises applying a linear regression model. 
     
     
         34 . The system of  claim 21 , wherein the processing time is based on numbers of transactions for the time intervals. 
     
     
         35 . The system of  claim 21 , wherein the operations further comprise:
 generating a posted transaction array based on the transaction frequency array, the posted transaction array comprising a plurality of time intervals and numbers of transactions for the time intervals;   determining a plurality of model results for time intervals in the posted transaction array using a plurality of prediction models, each prediction model outputting a corresponding model result; and   determining an hours of operation prediction by tallying model results; and   the graphical user interface further comprises a details window displaying the hours of operation prediction.   
     
     
         36 . The system of  claim 26 , wherein
 the histogram comprises a plurality of buttons, each button associated with a day of the week; and   the instructions further configure the at least one processor to execute instructions of displaying a second graphical user interface when a user selects one of the plurality of buttons.   
     
     
         37 . The system of  claim 26 , wherein the second graphical user interface comprises a second details window comprising a second histogram displaying time intervals for a selected day of the week and predicted occupancies for time intervals. 
     
     
         38 . The system of  claim 21 , wherein the graphical user interface further comprises a recommendations window displaying an alternative merchant associated with a category of the specified merchant and located within a threshold distance of the specified merchant. 
     
     
         39 . A non-transitory computer-readable medium storing instructions that, when executed by a processor, cause the processor to operate a system for generating a graphical user interface in a client device, the instructions comprising:
 receiving, from the client device, a request for occupancy information of a specified merchant;   obtaining, from a database in response to the request, a plurality of credit card proximity transactions associated with the merchant, the proximity transactions comprising at least one of NFC, chip-read, or swiped transactions;   generating a posted transaction array based on the credit card proximity transactions, the posted transaction array comprising a plurality of time intervals and numbers of transactions for the time intervals;   generating a transaction frequency array based on the posted transaction array, the transaction frequency array comprising weekdays and aggregated transactions associated with the weekdays, wherein generating the transaction frequency array comprises modifying the time intervals of the posted transaction array by subtracting a processing time;   generating a smoothed array by applying a kernel density estimate to the transaction frequency array; and   generating a graphical user interface displaying information in the smoothed array.   
     
     
         40 . A computer-implemented method for generating a graphical user interface in a client device, the method comprising:
 receiving, from the client device, a request for occupancy information of a specified merchant;   obtaining, from a database in response to the request, a plurality of credit card proximity transactions associated with the merchant, the proximity transactions comprising at least one of NFC, chip-read, or swiped transactions;   generating a posted transaction array based on the credit card proximity transactions, the posted transaction array comprising a plurality of time intervals and numbers of transactions for the time intervals;   generating a transaction frequency array based on the posted transaction array, the transaction frequency array comprising weekdays and aggregated transactions associated with the weekdays, wherein generating the transaction frequency array comprises modifying the time intervals of the posted transaction array by subtracting a processing time;   generating a smoothed array by applying a kernel density estimate to the transaction frequency array; and   generating a graphical user interface displaying information in the smoothed array.