US 12,086,817 B2
Personalized alert generation based on information dissemination
Marisa Affonso Vasconcelos, Sao Paulo (BR); Mu Qiao, Belmont, CA (US); Nicholas Linck, San Jose, CA (US); Yuya Jeremy Ong, San Jose, CA (US); Claudio Santos Pinhanez, Sao Paulo (BR); and Rogerio Abreu de Paula, Sao Paulo (BR)
Assigned to International Business Machines Corporation, Armonk, NY (US)
Filed by International Business Machines Corporation, Armonk, NY (US)
Filed on Mar. 31, 2021, as Appl. No. 17/218,297.
Prior Publication US 2022/0318823 A1, Oct. 6, 2022
Int. Cl. G06Q 30/0201 (2023.01); G06Q 50/00 (2024.01)
CPC G06Q 30/0201 (2013.01) [G06Q 50/01 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A system, said system comprising:
a memory; and
a processor in communication with said memory, said processor being configured to perform operations comprising:
analyzing historical interactions by a user within a network, wherein said historical interactions include historical user interaction data selected from the group consisting of user scroll speed and user clicks over time, and wherein said interactions are included in a corpus;
generating a user profile for said user, wherein said user profile includes an historical use by said user, and wherein said user profile is included in said corpus;
training a machine learning algorithm using said corpus;
monitoring current session interactions by said user within said network during a current user session, said current session interactions selected from the group consisting of user scroll speed and user clicks over time;
identifying an attempt by said user to share a post via said network during said current user session;
extracting a context from said post;
detecting a deviation of said attempt by said user to share said post from said historical use by said user based on a comparison of said current session interactions to said historical user interaction data;
determining said deviation indicates a current user temperament;
quantifying a user post partiality, wherein said user post partiality quantifies a partiality that said user has for said post, and wherein said user post partiality is based on said interactions, said user profile, and said context;
generating a personalized alert to evaluate said post with said machine learning algorithm, wherein said personalized alert is personalized to said user to engage said user in analytical thinking, wherein said personalized alert is personalized to said user based on said corpus, said context, and said user post partiality, and wherein said personalized alert includes a question for said user;
adapting said personalized alert to said user based on said current user temperament, wherein said adapting said personalized alert includes adjusting a complexity of said question based on said current user temperament; and
prompting said user with said adapted personalized alert during said current user session on said network before said user shares said post.