US 12,406,138 B2
System for providing intelligent part of speech processing of complex natural language
Michael L. Mattera, Coral Springs, FL (US); John A. Maliani, Pembroke Pines, FL (US); Robert L Vanderwall, Weston, FL (US); Dionny Santiago, Weston, FL (US); Brian R. Muras, Weston, FL (US); Keith A. Briggs, Coral Springs, FL (US); David Adamo, Sunrise, FL (US); and Tariq King, Pembroke Pines, FL (US)
Assigned to UKG INC., Weston, FL (US)
Filed by UKG INC., Weston, FL (US)
Filed on Dec. 23, 2022, as Appl. No. 18/088,411.
Application 18/088,411 is a continuation of application No. 16/827,089, filed on Mar. 23, 2020, granted, now 11,537,793.
Application 16/827,089 is a continuation of application No. 15/994,869, filed on May 31, 2018, granted, now 10,599,767, issued on Mar. 24, 2020.
Prior Publication US 2023/0132002 A1, Apr. 27, 2023
This patent is subject to a terminal disclaimer.
Int. Cl. G06F 40/253 (2020.01); G06F 40/242 (2020.01); G06F 40/284 (2020.01)
CPC G06F 40/253 (2020.01) [G06F 40/242 (2020.01); G06F 40/284 (2020.01)] 21 Claims
OG exemplary drawing
 
1. A system for providing intelligent part of speech processing of complex natural language, comprising:
a memory that stores instructions; and
a processor that executes the instructions to cause the processor to be configured to:
identify one or more non-text items in an input, wherein the non-text items correspond to an emoji, a GIF, an emoticon, video content, audio content, media content, or a combination thereof;
generate a first modified input, by replacing at least a portion of the identified non-text items in the input with one or more tokens that correspond to the replaced portion of the identified non-text items;
tag parts-of-speech for each non-text item and the one or more tokens in the first modified input to generate tagged part-of-speech text for the first modified input, wherein tagging comprises:
determining the parts-of-speech for each non-text item and the one or more tokens, using a series of semantic libraries, wherein each semantic library is a repository comprising media content and information comprising relationships and associations between concepts, keywords, words, content, and their meanings; and
determining whether the non-text items match and/or correlate with an initial set of concepts, keywords, content, terms, and/or identifiers in at least one of the semantic libraries;
replace, in the tagged part-of-speech text for the first modified input, the one or more tokens with the non-text items originally identified in the input to generate a second modified input;
update at least one agglomerated machine learning model utilized by the system based on analyzing the tagged part-of-speech text of the second modified input; and
generate at least one analysis or interpretation associated with the input using the updated at least one agglomerated model.