Neural network for optimizing display of hotels on a user interface

1. A system for implementing a neural network for optimal display of hotels on a user interface, the system comprising at least one processor and a memory storing computer program instructions which when executed by the at least one processor cause the at least one processor to perform operations comprising:

storing a neural network having an input layer, a hidden layer, and an output layer, wherein

the input layer comprises hotel neurons representing at least one hotel, and traveller neurons representing at least one traveller,

the hidden layer comprises a plurality of neurons, and receives a first value from the input layer, and

the output layer comprises display neurons representing at least one display, and receives a second value from the hidden layer;

presenting a training dataset to the neural network, the training dataset comprising travel data;

presenting a testing dataset after introduction of the training dataset, the testing dataset testing the performance and accuracy of the neural network;

determining, by the neural network, a combination of hotels to display to maximize a probability that a given hotel will be selected by a traveller, the combination of hotels to display determined by:

receiving a request by a traveller to book a hotel at a location;

determining a candidate set of hotels h 1 , h 2 , and h 3 , in a vicinity of the location;

determining a past purchase decision percentage by the traveller for each of a first paired combination h 1 and h 2 , a second paired combination h 1 and h 3 , and a third paired combination h 2 and h 3 , each respective past purchase decision percentage based on a prior booking choice of the traveller when presented with a first display having the first paired combination, a second display having the second paired combination, and a third display having the third paired combination;

determining objective point totals, based on employer preference, comprising (i) a first objective point total for hotel h 1 , (ii) a second objective point total for hotel h 2 , and (iii) a third objective point total for hotel h 3 ;

determining expected objective point totals, comprising:

(i) for the first paired combination, a first expected objective point total comprising a first sum of (a) the past purchase decision percentage of h 1 multiplied by the objective point total of h 1 , (b) the past purchase decision percentage of h 2 multiplied by the objective point total of h 2 , and (c) the past purchase decision percentage of h 3 multiplied by the objective point total of h 3 ;

(ii) for the second paired combination, a second expected objective point total comprising a second sum of (a) the past purchase decision percentage of h 1 multiplied by the objective point total of h 1 , (b) the past purchase decision percentage of h 2 multiplied by the objective point total of h 2 , and (c) the past purchase decision percentage of h 3 multiplied by the objective point total of h 3 ;

(iii) for the third paired combination, a third expected objective point total comprising a third sum of (a) the past purchase decision percentage of h 1 multiplied by the objective point total of h 1 , (b) the past purchase decision percentage of h 2 multiplied by the objective point total of h 2 , and (c) the past purchase decision percentage of h 3 multiplied by the objective point total of h 3 ; and

displaying to the traveller the paired combination having a highest expected objective point total.