System and method for electronic chat production

A computer-implemented method comprising:

• receiving, by a computer processor, an electronic chat, the electronic chat embodying a set of electronic chat messages;
• generating a set of models that model time gaps between adjacent messages from the set of electronic chat messages, wherein the set of models includes at least a first model corresponding to a single conversation and one or more second models that each correspond to two or more conversations, and wherein generating the set of models is performed without regard to the content of any of the electronic chat messages;
• and
• in response to selecting one of the one or more second models as a best model based on one or more criteria, adaptively splitting, by the computer processor, the set of electronic chat messages into a set of conversations without regard to the content of any of the electronic chat messages, each conversation in the set of conversations comprising a subset of electronic chat messages from the set of electronic chat messages, wherein adaptively splitting the set of electronic chat messages into a set of conversations further comprises:
• identifying a plurality of split points based on the time gaps between the adjacent messages from the set of electronic chat messages, wherein the identifying is performed based on providing information relating to the time gaps to a machine learning model trained to determine which of the time gaps correspond to a given split point between respective conversations based on distributions of time gaps associated with a plurality of electronic messages within the respective conversations;
• and
• storing, by the computer processor, each conversation from the set of conversations as a separate document.

An e-discovery system comprising:

• a computer processor;
• a non-transitory, computer-readable medium embodying thereon a set of computer instructions executable by the computer processor, the set of computer instructions comprising instructions for:
• sending a chat query to a remote electronic chat service;
• receiving an electronic chat responsive to the chat query, the electronic chat embodying a set of electronic chat messages;
• generating a set of models that model time gaps between adjacent messages from the set of electronic chat messages, wherein the set of models includes at least a first model corresponding to a single conversation and one or more second models that each correspond to two or more conversations, and wherein generating the set of models is performed without regard to the content of any of the electronic chat messages;
• and
• in response to selecting one of the one or more second models as a best model based on one or more criteria, adaptively splitting the set of electronic chat messages into a set of conversations without regard to the content of any of the electronic chat messages, each conversation in the set of conversations comprising a subset of electronic chat messages from the set of electronic chat messages, wherein adaptively splitting the set of electronic chat messages into a set of conversations further comprises:
• identifying a plurality of split points based on the time gaps between the adjacent messages from the set of electronic chat messages, wherein the identifying is performed based on providing information relating to the time gaps to a machine learning model trained to determine which of the time gaps correspond to a given split point between respective conversations based on time distributions of time gaps associated with a plurality of electronic messages within the respective conversations;
• and
• storing each conversation from the set of conversations as a separate document.