公开(公告)号：US20230419042A1

公开(公告)日：2023-12-28

申请号：US17937606

申请日：2022-10-03

申请人： UnitedHealth Group Incorporated

发明人： Rajesh SABAPATHY ,  Chirag MITTAL ,  Gourav AWASTHI ,  Aditya Teja JOSYULA

IPC分类号： G06F40/30 ,  G06N5/02

CPC分类号： G06F40/30 ,  G06N5/022

摘要： There is a need for more effective, efficient, and accurate computer text comprehension. This need is addressed by applying unique text processing techniques to identify and remove irrelevant sentences from a narrative. The text processing techniques include a machine-learning based model that is trained using automatically generated training data that is tailored to a particular circumstance. A method for machine narrative comprehension includes receiving a narrative data object comprising one or more sentences; determining, using a machine-learning based irrelevant classifier model, a relevance of at least one of the one or more sentences; responsive to a determination that at least one sentence is irrelevant, generating a pertinent summary by removing the at least one sentence from the narrative; and generating, based at least in part on the pertinent summary, an output indicia data object for the narrative data object.

公开(公告)号：US20220277315A1

公开(公告)日：2022-09-01

申请号：US17249383

申请日：2021-03-01

申请人： UnitedHealth Group Incorporated

发明人： Chirag MITTAL ,  Arjun RANA ,  Rajesh SABAPATHY ,  Erin CONWAY ,  Sudeep VISHNUMURTHY ,  Arushi PARGANIHA

IPC分类号： G06Q30/00 ,  G06F16/951 ,  G06F16/908 ,  G06Q40/08 ,  G06Q10/10

摘要： Methods, apparatuses, systems, computing entities, and/or the like are provided. An example method may include retrieving a scheme data object; generating a tabular data object; generating a graph data object including a plurality of nodes and a plurality of edges; calculating, based on the graph data object, a plurality of node attributes for the plurality of nodes and a plurality of edge attributes for the plurality of edges; identifying at least one irregular member node from the plurality of member nodes; identifying at least one irregular referring provider node from the plurality of referring provider nodes; identifying at least one irregular servicing provider node from the plurality of servicing provider nodes based at least in part on the at least one irregular member node and the at least one irregular referring provider node; and performing one or more responsive actions.

公开(公告)号：US20230419051A1

公开(公告)日：2023-12-28

申请号：US17937616

申请日：2022-10-03

申请人： UnitedHealth Group Incorporated

发明人： Rajesh SABAPATHY ,  Chirag MITTAL ,  Gourav AWASTHI ,  Aditya Teja JOSYULA

IPC分类号： G06F40/56 ,  G06F40/35 ,  G06F40/247

CPC分类号： G06F40/56 ,  G06F40/35 ,  G06F40/247

摘要： There is a need for more effective and efficient predictive natural language summarization. This need is addressed by applying hybrid extractive and abstractive summarization techniques in a unique processing pipeline to generate a cohesive and comprehensive summary of a multi-party interaction. A method for generating the summary of a multi-party interaction includes receiving a multi-party interaction transcript data object comprising a plurality of interaction utterances from at least two participants; using an extractive summarization model to identify a key sentence of the multi-party interaction transcript data object; identifying an interaction utterance from the multi-party interaction transcript data object that corresponds to the key sentence; generating a contextual summary for the multi-party interaction transcript data object based at least in part on the interaction utterance; and generating a reported contextual summary for the multi-party interaction transcript data object based at least in part on the contextual summary.