US 12,354,039 B2
Detection and classification of impediments
Vinu Varghese, Bangalore (IN); Anto Jose, Thodupuzha (IN); Balaji Janarthanam, Chennai (IN); Selvakuberan Karuppasamy, Chennai (IN); Saumyabrata Mitra, Kolkata (IN); Karthik Srikantamurthy, Mysuru (IN); Ravi Kant Gaur, Bangalore (IN); Ragav Devanathan, Chennai (IN); and Nirav Jagdish Sampat, Mumbai (IN)
Assigned to Accenture Global Solutions Limited, Dublin (IE)
Filed by Accenture Global Solutions Limited, Dublin (IE)
Filed on Jun. 21, 2022, as Appl. No. 17/808,039.
Prior Publication US 2023/0410004 A1, Dec. 21, 2023
Int. Cl. G06Q 10/0631 (2023.01)
CPC G06Q 10/06312 (2013.01) [G06Q 10/063114 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method performed by one or more processors for artificial intelligence based detection and classification of impediments in a network, the method comprising:
by the one or more processors, receiving from a first communication channel and a second communication channel via the network a first dataset associated with a first project, the first dataset including first content received from the first communication channel and second content received from the second communication channel;
by the one or more processors, performing, by an active learning module, a first active learning cycle in which data from the first dataset is labeled and then used to train a first model to classify data points as either technical or non-technical based on a classification feature set;
by the one or more processors, performing, by the active learning module, after the first active learning cycle, a second active learning cycle in which the first model is refined to improve an accuracy of the first model in classifying the data points as either technical or non-technical based on the classification feature set;
by the one or more processors, identifying at least a first work item in the first dataset;
by the one or more processors, detecting, with respect to a first data point in the first dataset, a first impediment that is associated with the first work item;
by the one or more processors, classifying, using the first model as refined in the second active learning cycle, the first impediment as technical or non-technical;
by the one or more processors, selecting a first stakeholder of the first project based on the classification of the first impediment;
by the one or more processors, generating a message of the first impediment based on the selection of the first stakeholder; and
by the one or more processors, transmitting the message of the first impediment to a user device associated with the first stakeholder.