Machine learning based customized notification generation and authentication system

Abstract

A system for employing machine learning to generate customized notifications for a user is provided. The system may identify a landmark in proximity to a user's mobile device and may obtain information associated with products or services offered by the landmark. While the mobile device is within a predetermined distance of the landmark, a machine learning model may be employed to generate a customized notification, such as an offer—e.g., a discount or a special financing offer for one of the identified products—generated specifically for use by the user based on the user's spending and/or financing history. The customized notification may be transmitted to the mobile device with a security key for accessing the offer. In response to receiving an indication that the security key was selected and the product purchased, the user may be authenticated and the offer may be applied to an account associated with the user.

What is claimed is: 
     
         1 . A method comprising:
 inputting, by a computing device and to a machine learning model, a history associated with a first user, a location of a mobile device of the first user and a location of a first landmark, wherein the machine learning model is trained based on training data comprising:
 training offer information indicating whether a training offer has been redeemed by a training user, 
 product information for the training offer, and 
 a distance between a training landmark and a location of a computing device associated with the training user when the training offer was generated; 
   receiving, by the computing device, as output from the machine learning model and based on the history associated with the first user, a likelihood that the first user would redeem an offer for a first product associated with the first landmark;   based on the likelihood that the first user would redeem the offer for the first product, transmitting, from the computing device and to the mobile device of the first user, a notification of the offer, wherein the notification of the offer includes a security key for accessing the offer;   receiving, by the computing device, an indication of a selection of the security key;   authenticating, based on determining that identification information of a device sending the indication of the selection of the security key matches identification information associated with one or more devices of the first user and based on a distance between the first landmark and the device sending the indication of the selection of the security key, the first user; and   applying, based on authentication of the first user, the offer to an account associated with the first user.   
     
     
         2 . The method of  claim 1 , further comprising:
 determining, utilizing a web script configured to identify key words on a web page associated with the first landmark, one or more products associated with the first landmark comprising the first product.   
     
     
         3 . The method of  claim 1 , further comprising:
 determining, utilizing optical character recognition (OCR) to process an image of a receipt for a transaction made by the first user, one or more products associated with the history of the first user.   
     
     
         4 . The method of  claim 1 , further comprising:
 after determining that the mobile device is within a predetermined distance of the first landmark:
 receiving, from the mobile device, one or more operating parameters comprising an internet protocol (IP) address of the mobile device; and 
 based on determining that a location associated with the IP address does not match the location of the mobile device, sending, to the mobile device, a notification to disable a virtual private network (VPN) service on the mobile device. 
   
     
     
         5 . The method of  claim 1 , wherein authenticating the first user comprises:
 receiving, in response to the selection of the security key, security key information comprising information related to the offer;   querying, based on the security key information, a database to determine a user for whom the notification of the offer was generated, wherein the user for whom the notification of the offer was generated is the first user; and   querying the database to determine identification information associated with one or more devices of the first user.   
     
     
         6 . The method of  claim 5 , wherein the one or more devices of the first user comprise the mobile device of the first user or a different device of the first user. 
     
     
         7 . The method of  claim 1 , wherein authenticating the first user further comprises:
 updating the machine learning model with information associated with the offer comprising a distance between the training landmark and the location of the training user when the security key associated with the training offer was selected by the training user computing device; and   determining, using the updated machine learning model and based on the distance between the first landmark and the device sending the indication of the selection of the security key, a likelihood that the first user has redeemed the offer,   wherein applying the offer to the account associated with the first user is further based on the likelihood that the first user has redeemed the offer.   
     
     
         8 . The method of  claim 1 , further comprising:
 updating, after receiving the indication of the selection of the security key and after authenticating the first user, the machine learning model with information associated with the offer, wherein the information associated with the offer further comprises one or more of: the first product, the first landmark, a date and time of transmission of the notification of the offer, a date and time of the selection of the security key, a date and time associated with the information indicating that the first user purchased the first product, a value associated with the offer, a type of the offer, or a price of the first product at the first landmark.   
     
     
         9 . The method of  claim 1 , wherein the history of the first user comprises information indicating one or more of:
 spending patterns associated with the first user,   products preferred by the first user,   merchants preferred by the first user, or   average spending of the first user for different categories of products.   
     
     
         10 . The method of  claim 1 , further comprising:
 receiving, from the mobile device, information indicating that the first user performed web browsing for the first product; and   generating the notification of the offer based on the information indicating that the first user performed the web browsing for the first product.   
     
     
         11 . The method of  claim 1 , further comprising:
 querying a database associated with the first landmark to determine a stock level of one or more products associated with the first landmark; and   generating the notification based on the stock level of the one or more products.   
     
     
         12 . A computing device comprising:
 one or more processors; and   memory storing instructions that, when executed by the one or more processors, cause the computing device to:   input, to a machine learning model, a history associated with a first user, a location of a mobile device of the first user and a location of a first landmark;   receive, as output from the machine learning model and based on the history associated with the first user, a likelihood that the first user would redeem an offer for a first product associated with the first landmark;   based on the likelihood that the first user would redeem the offer for the first product, transmit, to the mobile device of the first user, a notification of the offer, wherein the notification of the offer includes a security key for accessing the offer;   receive an indication of a selection of the security key;   authenticate, based on determining that identification information of a device sending the indication of the selection of the security key matches identification information associated with one or more devices of the first user and based on a distance between the first landmark and the device sending the indication of the selection of the security key, the first user; and   apply, based on authentication of the first user, the offer to an account associated with the first user.   
     
     
         13 . The computing device of  claim 12 , wherein the instructions, when executed by the one or more processors, cause the computing device to:
 train, using training data, the machine learning model to determine a likelihood that a training user would redeem a training offer, wherein the training data comprises:
 training offer information indicating whether the training offer has been redeemed by the training user, 
 product information for the training offer, and 
 a distance between a training landmark and a location of a computing device associated with the training user when the training offer was generated. 
   
     
     
         14 . The computing device of  claim 12 , wherein the instructions, when executed by the one or more processors, cause the computing device to:
 determine, utilizing a web script configured to identify key words on a web page associated with the first landmark, one or more products associated with the first landmark comprising the first product.   
     
     
         15 . The computing device of  claim 12 , wherein the instructions, when executed by the one or more processors, cause the computing device to:
 determine, utilizing optical character recognition (OCR) to process an image of a receipt for a transaction made by the first user, one or more products associated with the history of the first user.   
     
     
         16 . The computing device of  claim 12 , wherein the instructions, when executed by the one or more processors, cause the computing device to:
 receive, from the mobile device, one or more operating parameters comprising an internet protocol (IP) address of the mobile device; and   based on determining that a location associated with the IP address does not match the location of the mobile device, send, to the mobile device, a notification to disable a virtual private network (VPN) service on the mobile device.   
     
     
         17 . One or more non-transitory, computer-readable media storing instructions that, when executed by one or more processors of a computing device, cause the computing device to:
 input, to a machine learning model, a history associated with a first user, a location of a mobile device of the first user and a location of a first landmark;   receive, as output from the machine learning model and based on the history associated with the first user, a likelihood that the first user would redeem an offer for a first product associated with the first landmark;   based on the likelihood that the first user would redeem the offer for the first product, transmit, to the mobile device of the first user, a notification of the offer;   authenticate, based on determining that identification information of a device sending a selection of the notification to redeem the offer matches identification information associated with one or more devices of the first user and based on a distance between the first landmark and the device sending the selection, the first user; and   apply, based on authentication of the first user, the offer to an account associated with the first user.   
     
     
         18 . The one or more non-transitory, computer-readable media of  claim 17 , wherein the notification of the offer includes a security key for accessing the offer, and wherein the instructions, when executed by the one or more processors, cause the computing device to receive an indication of a selection of the security key; and wherein authenticate the first user is further based on determining that the identification information of a device sending the indication of the selection of the security key matches the identification information associated with one or more devices of the first user and based on a distance between the first landmark and the device sending the indication of the selection of the security key. 
     
     
         19 . The one or more non-transitory, computer-readable media of  claim 18 , wherein the instructions, when executed by the one or more processors, cause the computing device to authenticate the first user by causing the computing device to:
 receive, in response to the selection of the security key, security key information comprising information related to the offer;   query, based on the security key information, a database to determine a user for whom the notification of the offer was generated, wherein the user for whom the notification of the offer was generated is the first user; and   query the database to determine identification information associated with one or more devices of the first user.   
     
     
         20 . The one or more non-transitory, computer-readable media of  claim 17 , wherein the instructions, when executed by the one or more processors, cause the computing device to:
 generate the notification of the offer for the first product based on:
 analyzing, using the machine learning model, the history associated with the first user to generate one or more of a product score and a spending score; 
 generating, based on one or more of the product score and the spending score, a ranked list of one or more notifications associated with one or more products; and 
   selecting, from the ranked list of the one or more notifications, the notification of the offer for the first product.