| CPC G06Q 30/0282 (2013.01) [G06F 40/295 (2020.01); G06N 7/01 (2023.01); G06F 40/30 (2020.01); G06N 3/044 (2023.01); G06N 3/045 (2023.01); G06N 3/08 (2013.01); G06N 20/00 (2019.01); G06N 20/10 (2019.01)] | 20 Claims |
|
1. A method of troubleshooting a software application, comprising:
receiving natural language text generated by a plurality of different sources of information;
pre-processing the natural language text by cleaning and vectorizing the natural language text;
extracting a plurality of negative reviews from the natural language text by executing a first machine learning model (MLM), wherein a first input to the first MLM is the natural language text and a first output of the first MLM is first probabilities that the plurality of negative reviews have negative sentiments, and wherein the plurality of negative reviews comprise instances of the natural language text having corresponding negative sentiment probabilities above a threshold value;
categorizing the plurality of negative reviews by executing a second MLM, wherein a second input to the second MLM is the plurality of negative reviews, and wherein a second output of the second MLM is second probabilities that the plurality of negative reviews are assigned to a plurality of categories;
identifying, using a name recognition controller and based on categorizing, a name of a software application in the plurality of negative reviews and sorting the plurality of negative reviews into a subset of negative reviews relating to the name; and
adjusting the software application, named by the name recognition controller, based on the subset of negative reviews.
|