Chat discourse convolution

1. A method, comprising:

performing, with a computer, a lexicographic analysis of content of a chat message of a chat discourse and storing a result of the lexicographic analysis in a database;

selecting a hash technique for generating a hash of the content of the chat message based on a likelihood that the hash of the content of the chat message will be expanded for use in content analysis, wherein the hash technique is performed using machine learning and the likelihood that the hash of the content of the chat message will be expanded for use in content analysis is determined by monitoring which hashes of other processed chat messages have been expanded for purposes of content analysis;

generating the hash of the content of the chat message using the selected hash technique;

creating a leximarker comprising the hash of the content of the chat message and a reference to the stored result of the lexicographic analysis;

generating a modified version of the chat message by replacing the content of the chat message with the leximarker;

electronically storing the modified version of the chat message in place of the chat message as part of the chat discourse; and

using the leximarker to retrieve the result of the lexicographic analysis and, based on the retrieved result, performing content analysis of the chat discourse, wherein the content analysis of the chat discourse is performed without expanding the hash of the content of the chat message to obtain the content of the chat message in an original, unmodified form.

2. The method of claim 1 , wherein a size of the hash varies based on a text length of the chat message and an expected amount of lexicographical information to be extracted from the chat message.

3. The method of claim 1 , wherein the generating the hash comprises determining an amount of hashing of the content of the chat message based on comparing a leximarker contextual value of the chat message with a leximarker contextual threshold.

4. A system, comprising:

a computer having at least one processor programmed to initiate executable operations, the executable operations including:

selecting a hash technique for generating a hash of the content of the chat message based on a likelihood that the hash of the content of the chat message will be expanded for use in content analysis, wherein the hash technique is performed using machine learning and the likelihood that the hash of the content of the chat message will be expanded for use in content analysis is determined by monitoring which hashes of other processed chat messages have been expanded for purposes of content analysis;

generating the hash of the content of the chat message using the selected hash technique;

creating a leximarker comprising the hash of the content of the chat message and a reference to the stored result of the lexicographic analysis;

generating a modified version of the chat message by replacing the content of the chat message with the leximarker;

electronically storing the modified version of the chat message in place of the chat message as part of the chat discourse; and

using the leximarker to retrieve the result of the lexicographic analysis and, based on the retrieved result, performing content analysis of the chat discourse, wherein the content analysis of the chat discourse is performed without expanding the hash of the content of the chat message to obtain the content of the chat message in an original, unmodified form.

5. The system of claim 4 , wherein the generating the hash comprises determining an amount of hashing of the content of the chat message based on comparing a leximarker contextual value of the chat message with a leximarker contextual threshold.

6. The system of claim 4 , wherein a size of the hash varies based on a text length of the chat message and an expected amount of lexicographical information to be extracted from the chat message.

7. A computer program product, comprising:

a computer-readable storage medium having program code stored thereon, the program code executable by computer hardware to initiate operations including:

performing a lexicographic analysis of content of a chat message of a chat discourse and storing a result of the lexicographic analysis in a database;

selecting a hash technique for generating a hash of the content of the chat message based on a likelihood that the hash of the content of the chat message will be expanded for use in content analysis, wherein the hash technique is performed using machine learning and the likelihood that the hash of the content of the chat message will be expanded for use in content analysis is determined by monitoring which hashes of other processed chat messages have been expanded for purposes of content analysis;

generating the hash of the content of the chat message using the selected hash technique;

creating a leximarker comprising the hash of the content of the chat message and a reference to the stored result of the lexicographic analysis;

generating a modified version of the chat message by replacing the content of the chat message with the leximarker;

electronically storing the modified version of the chat message in place of the chat message as part of the chat discourse; and

using the leximarker to retrieve the result of the lexicographic analysis and, based on the retrieved result, performing content analysis of the chat discourse, wherein the content analysis of the chat discourse is performed without expanding the hash of the content of the chat message to obtain the content of the chat message in an original, unmodified form.

8. The computer program product of claim 7 , wherein wherein a size of the hash varies based on a text length of the chat message and an expected amount of lexicographical information to be extracted from the chat message.

9. The computer program product of claim 7 , wherein the generating the hash comprises determining an amount of hashing of the content of the chat message based on comparing a leximarker contextual value of the chat message with a leximarker contextual threshold.