	US 12,219,453 B2
	Dynamic communication routing based on consistency weighting and routing rules
Matan Barak, Ra'anana (IL); Efim Dimenstein, Bnei Atarot (IL); and Shlomo Lahav, Ramat-Gan (IL)
Assigned to LIVEPERSON, INC., New York, NY (US)
Filed by LIVEPERSON, INC., New York, NY (US)
Filed on Mar. 20, 2023, as Appl. No. 18/123,528.
Application 18/123,528 is a continuation of application No. 17/095,320, filed on Nov. 11, 2020, granted, now 11,638,195.
Application 17/095,320 is a continuation of application No. 16/166,297, filed on Oct. 22, 2018, granted, now 10,869,253, issued on Dec. 15, 2020.
Application 16/166,297 is a continuation of application No. 15/171,525, filed on Jun. 2, 2016, granted, now 10,142,908, issued on Nov. 27, 2018.
Claims priority of provisional application 62/169,726, filed on Jun. 2, 2015.
Prior Publication US 2023/0232305 A1, Jul. 20, 2023
Int. Cl. H04W 76/10 (2018.01); G06Q 10/107 (2023.01); G06Q 30/01 (2023.01); G06Q 30/02 (2023.01); H04W 8/26 (2009.01); H04W 40/02 (2009.01); H04W 88/02 (2009.01)

CPC H04W 40/02 (2013.01) [G06Q 10/107 (2013.01); G06Q 30/01 (2013.01); G06Q 30/02 (2013.01); H04W 8/26 (2013.01); H04W 76/10 (2018.02); H04W 88/02 (2013.01)]

21 Claims

1. A computer-implemented method, comprising:

detecting an interaction between a network device and a terminal device, wherein the interaction corresponds to a set of messages transmitted during a communications session between the network device and the terminal device, and wherein the interaction corresponds to a trend in a particular dimension;

identifying one or more topics associated with the interaction, wherein the one or more topics are identified based on content corresponding to the set of messages;

determining one or more inferred sentiment scores associated with the interaction, wherein the one or more inferred sentiment scores are determined based on the set of messages;

generating one or more message indices based on the one or more topics and the one or more inferred sentiment scores, wherein the one or more message indices include subsets of the set of messages;

identifying a message index from the one or more message indices, wherein the message index is identified in accordance with a rule;

generating a message chronicle, wherein the message chronicle includes one or more messages associated with the message index identified from the one or more message indices; and

analyzing the message chronicle to characterize a presence of the trend in the particular dimension.