| CPC G06Q 10/1091 (2013.01) [G06Q 10/063114 (2013.01); G06Q 10/1053 (2013.01); G06Q 2220/00 (2013.01)] | 19 Claims |
|
1. An apparatus for assorter quantification, the apparatus comprising:
at least a processor; and
a memory communicatively connected to the at least a processor and including instructions configuring the at least a processor to:
receive a transfer request, wherein receiving the transfer request comprises authenticating the assorter and the endpoint, wherein authenticating the transfer request comprises authenticating an endpoint identity of the endpoint by a biometric authentication, wherein authenticating an endpoint identity comprises:
receive a fingerprint scan from a biometric sensor associated with the endpoint and authenticating the fingerprint scan received from the endpoint to authenticate the endpoint identity;
scan a user fingerprint as a function of a fingerprint biometric sensor;
scan a user face as a function of a facial biometric sensor;
identify a user typing behavior using a video capture device;
authenticate an assorter identity as a function of the fingerprint scan, the user face scan and the user typing behavior wherein authenticating the user comprises:
calculating a confidence level for the assorter identity, wherein calculating the confidence level further comprises a statistical measure of reliability; and
comparing the confidence level to an authentication threshold;
receive an assorter-linked data set associated with an assorter and an endpoint-linked data set associated with an endpoint;
identify an assortment activity as a function of the assorter-linked data set and the endpoint-linked data set; and
determine a quantification action as a function of the assortment activity, wherein determining a quantification action comprises:
training a machine learning model as a function of training data and a machine learning algorithm, wherein training the machine learning model further comprises:
generating a degree of match between at least two pairs of fuzzy sets, using a classifier derived from training data, wherein multiple fuzzy matches can be performed, wherein the training data further comprises at least an assortment activity input and outputs at least a quantification action, wherein outputting at least a quantification action further comprises applying weighted values to the at least an assortment activity input and correlating the weighted values of the at least an assortment activity to adjacent layers of the at least a quantification action;
computing an overall degree of match by averaging the degree of match between the at least two pairs of fuzzy sets to measure similarity between the assorter-linked data set and the endpoint-linked data set; and
generating, using the trained machine learning model, the quantification action; and
transferring a payment to the assorter as a function of the quantification action.
|