US 12,141,215 B2
Graph-based prediction for contact suggestion in a location sharing system
Pierre Leveau, Paris (FR); and Antoine Sinton, Courteuil (FR)
Assigned to Snap Inc., Santa Monica, CA (US)
Filed by Snap Inc., Santa Monica, CA (US)
Filed on Aug. 15, 2023, as Appl. No. 18/450,110.
Application 18/450,110 is a continuation of application No. 16/353,604, filed on Mar. 14, 2019, granted, now 11,868,414.
Prior Publication US 2023/0385347 A1, Nov. 30, 2023
This patent is subject to a terminal disclaimer.
Int. Cl. G06F 16/9535 (2019.01); G06F 3/0484 (2022.01); G06F 16/9536 (2019.01); G06N 3/08 (2023.01); G06N 5/022 (2023.01); G06N 20/00 (2019.01)
CPC G06F 16/9535 (2019.01) [G06N 20/00 (2019.01); G06F 3/0484 (2013.01); G06N 3/08 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A system comprising:
one or more processors; and
a memory storing instructions that, when executed by the one or more processors, configure the system to perform operations comprising:
accessing, from a database coupled to a server computer, user data of a plurality of users;
generating, using one or more processors of the server computer, a directed graph of the plurality of users,
wherein a weight of a weighted edge between two users of the plurality of users is based, at least partially, on an amount of time the two users have spent together;
computing, by the one or more processors, a first score for each user of the plurality of users, the first score of each user being computed using an edge-weighted edge-based ranking algorithm based on the directed graph,
wherein for each user u of the plurality of users, the first score is dependent on edge-weighted values for each user v included in a set containing all users linked to the user u, divided by a number of links from the second user v;
computing, by the one or more processors, using a machine learning model, a second score for each user of the plurality of users, the second score of each user being, at least partially, based on the first score of the user, the second score for each user being computed based on a first probability of a first user of the plurality of users sending a connection request to the user, and on a second probability of the user accepting the connection request received from the first user;
generating, by the one or more processors, a ranked contact suggestion list of one or more users of the plurality of users, the one or more users being ranked based on their respective second score, and the one more users for presenting as contact suggestions to the first user; and
causing display, on a client device of the first user, of the ranked contact suggestion list.