US 11,893,608 B2
System and method providing business insights based on cluster success analytics for products based businesses
Shlomi Medalion, Tel Aviv (IL); Yair Horesh, Tel Aviv (IL); Yehezkel Shraga Resheff, Tel Aviv (IL); Sigalit Bechler, Tel Aviv (IL); Oren Sar Shalom, Tel Aviv (IL); and Daniel Ben David, Tel Aviv (IL)
Assigned to INTUIT INC., Mountain View, CA (US)
Filed by Intuit Inc., Mountain View, CA (US)
Filed on Mar. 13, 2020, as Appl. No. 16/818,268.
Prior Publication US 2021/0287261 A1, Sep. 16, 2021
Int. Cl. G06Q 30/0282 (2023.01); G06Q 30/0202 (2023.01); G06F 40/30 (2020.01); G06F 18/23 (2023.01); G06F 17/16 (2006.01); G06N 3/084 (2023.01); G06F 18/214 (2023.01)
CPC G06Q 30/0282 (2013.01) [G06F 17/16 (2013.01); G06F 18/2148 (2023.01); G06F 18/23 (2023.01); G06F 40/30 (2020.01); G06N 3/084 (2013.01); G06Q 30/0202 (2013.01)] 25 Claims
OG exemplary drawing
 
1. A computer-implemented method for recommending business factors to a merchant, said method comprising:
training a network to identify related merchant-vendor pairs by:
embedding, by a first encoder, a plurality of merchant descriptions to a vector space to obtain a plurality of merchant vectors, each merchant description corresponding to a vendor;
embedding, by a second encoder, each corresponding vendor to the vector space to obtain a plurality of vendor vectors;
calculating a relation metric for each description-vendor pair;
training a neural network with the description-vendor pairs and the corresponding relation metrics to predict whether a new vendor and a new merchant description are related;
training the first encoder to embed related merchant descriptions to similar regions in the vector space; and
training the second encoder to embed related vendors to the similar regions in the vector space;
receiving a request for recommended business factors from a device;
receiving merchant data associated with a merchant from the device, the merchant data comprising vendor data associated with the merchant;
embedding the merchant data to a vector space to obtain a merchant vector, the vector space comprising a plurality of other vectors associated with other merchants;
calculating, using the trained network, a relation metric between the merchant vector and a vector of the plurality of other vectors, the vector being associated with a second merchant, the relation metric representing a degree of relation between the merchant and the second merchant;
determining that the relation metric is above a pre-defined threshold value; and
when it is determined that the relation metric is above the pre-defined threshold value, sending merchant data associated with the second merchant to the device, the merchant data associated with the second merchant comprising business factors associated with the second merchant.