| CPC G06Q 30/0206 (2013.01) [G06N 20/00 (2019.01); G06Q 30/0625 (2013.01); G06Q 30/0629 (2013.01); G06Q 30/0641 (2013.01)] | 20 Claims |
|
1. A method comprising:
training a machine learning model, implemented via a plurality of connected nodes that are part of a plurality of players of an artificial neural network, to output a customized price indicator for a particular merchant by adjusting one or more machine learning model parameters associated with at least one connected node of the plurality of connected nodes that are part of the plurality of layers of the artificial neural network so as to train an ability of the trained machine learning model to output the customized price indicator for the particular merchant;
detecting a message, in a messaging application executing on a client device, that comprises an indication of one or more merchants, wherein the message is part of a group messaging session comprising different messages associated with a plurality of users;
retrieving, based on the indication of the one or more merchants and for each of the one or more merchants, merchant data that indicates at least one cost associated with a corresponding merchant;
generating one or more group price indicators by determining a group price indicator for each merchant of the one or more merchants by:
determining, for each corresponding user of the plurality of users and using a plurality of identifiers for the plurality of users, spending habit data that indicates spending habits for the corresponding user at one or more different merchants;
providing, as input to the machine learning model, the merchant data and the spending habit data for each of the plurality of users;
determining a plurality of customized price indicators by determining, based on one or more outputs of the machine learning model received in response to the input, a customized price indicator for each user that indicates a comparison of a predicted cost of a first merchant relative to an average past expenditure, by each user, at the one or more different merchants; and
combining each of the plurality of customized price indicators for each user to yield a group price indicator for the merchant; and
causing display, in the messaging application executing on the client device, of the one or more group price indicators for each of the one or more merchants.
|