US 12,223,122 B1
Method of system of providing local personalized emoticon recommendation
Rajat Aggarwal, Noida (IN)
Assigned to Microsoft Technology Licensing, LLC, Redmond, WA (US)
Filed by Microsoft Technology Licensing, LLC, Redmond, WA (US)
Filed on Nov. 30, 2023, as Appl. No. 18/524,785.
Int. Cl. G06F 3/023 (2006.01); G06F 40/284 (2020.01); G06F 40/40 (2020.01); H04L 51/04 (2022.01); H04L 51/216 (2022.01)
CPC G06F 3/0237 (2013.01) [G06F 40/284 (2020.01); G06F 40/40 (2020.01); H04L 51/04 (2013.01); H04L 51/216 (2022.05)] 20 Claims
OG exemplary drawing
 
1. A client device comprising:
a processor; and
a memory in communication with the processor, the memory comprising executable instructions that, when executed by the processor alone or in combination with other processors, cause the client device to perform functions of:
receiving input data at an application including a local emoticon recommendation engine on the client device of a user that offers use of emoticons in communications between the user and other users;
pre-processing the input data via the local emoticon recommendation engine by removing at least one of stop words, punctuation, and special characters, to generate one or more text segments from the input data;
following the pre-processing, tokenizing the one or more text segments via the local emoticon recommendation engine to generate one or more tokenized words from the one or more text segments;
tagging the tokenized words via the local emoticon recommendation engine by querying a natural language library to generate a part-of-speech (POS) tag set for each of the one or more tokenized words;
providing the generated POS tag set to a local machine learning (ML) model trained for providing emoticon recommendations based on the POS tag set and a library of emoticons;
executing the local ML model on the client device to provide one or more emoticon recommendations;
displaying the one or more emoticon recommendations to the user as user selectable recommendations; and
implementing ongoing training on the local ML model, wherein training the local ML model includes associating each POS tag set in a list of POS tag sets with ranked emoticons.