ARTIFICIAL INTELLIGENCE FOR CONTEXT CLASSIFIER

1 . A message context classifier for sorting out messages that match predefined fields-of-interest, comprising:

means for receiving streams of electronic messages having as yet undetermined contents, contexts, and sentiments communicated between senders and recipients;

means for separating said streams of electronic messages into individual messages and missives from identifiable senders to identifiable recipients;

means for deconstructing each said message and missive into its constituent words;

means for disambiguating said constituent words by finding commonalities between them that exist for particular and predefined subject categories and fields-of-interest;

means for estimating a most probable subject category and field-of-interest for an instant message and missive;

means for evaluating a particular sentiment conveyed by the sender in said instant message and missive as confined to an estimate of a most probable subject category and field-of-interest;

means for classifying said instant missive as warranting the attention of a user according to a match between a user input selection and the results obtained by the means for estimating a most probable subject category and field-of-interest and the means for evaluating a particular sentiment; and

means for outputting instant messages and missives classified in a particular way.

2 . The message context classifier for sorting out messages that match predefined fields-of-interest of claim 1 , further comprising:

means for identifying individual senders of said messages and profiling their behaviors and classifications of previous messages; and

means for weighting an estimate of a most probable subject category and field-of-interest for an instant missive according to a behavior profile of corresponding individual senders.

3 . The message context classifier for sorting out messages that match predefined fields-of-interest of claim 1 , further comprising:

means for identifying individual recipients of said messages and profiling their behaviors and classifications of previous messages directed to them; and

means for weighting an estimate of a most probable subject category and field-of-interest for an instant missive according to a behavior profile of corresponding individual recipients.

4 . The message context classifier for sorting out messages that match predefined fields-of-interest of claim 1 , further comprising:

means for identifying peer groups of senders of said messages and profiling their behaviors and classifications of previous messages sent by any of them; and

means for weighting an estimate of a most probable subject category and field-of-interest for an instant missive according to a behavior profile of corresponding senders in their relevant peer group.

5 . The message context classifier for sorting out messages that match predefined fields-of-interest of claim 1 , further comprising:

means for ranking said constituent words by finding their frequencies of common usage that exist for particular and predefined subject categories and fields-of-interest; and

means for estimating a most probable subject category and field-of-interest for an instant missive using word rankings.

6 . The message context classifier for sorting out messages that match predefined fields-of-interest of claim 1 , further comprising:

means for ranking said constituent words as always used, commonly used, rarely used, and never used by a sorting their frequencies of common usage that exist for particular and predefined subject categories and fields-of-interest; and

means for estimating a most probable subject category and field-of-interest for an instant missive using word rankings.

7 . The message context classifier for sorting out messages that match predefined fields-of-interest of claim 1 , further comprising:

means for ranking said constituent words as always-used, commonly-used, rarely-used, and never-used by a sorting their frequencies of common usage that exist for particular and predefined subject categories and fields-of-interest; and

means for eliminating from further consideration a subject category and field-of-interest for an instant missive if a word ranking is returned as never used;

wherein, a reduction of the uncertainty is obtained about the subject category and field-of-interest for an instant missive.

8 . The message context classifier for sorting out messages that match predefined fields-of-interest of claim 1 , further comprising:

means for continuously weighing said constituent words from always-used to never-used according to statistics of their frequencies in common usage that exist for particular and predefined subject categories and fields-of-interest; and

means for eliminating from further consideration a subject category and field-of-interest for an instant missive if the weight of a constituent word is returned as not exceeding an adjustable threshold;

wherein, a reduction of the uncertainty is obtained about the subject category and field-of-interest for an instant missive.

9 . The message context classifier for sorting out messages that match predefined fields-of-interest of claim 1 , further comprising:

a smart agent assigned for each word in a vocabulary of words expected to be encountered in said messages, and each including attributes describing alternative meanings that can be ascribed to the particular word.

10 . The message context classifier for sorting out messages that match predefined fields-of-interest of claim 1 , further comprising:

a smart agent assigned for each context, content, and sentiment that can be conveyed by words expected to be encountered in said messages, and each including attributes describing alternatives that can be ascribed to it.

11 . The message context classifier for sorting out messages that match predefined fields-of-interest of claim 1 , further comprising:

a smart agent assigned for each sender communicating context, content, and sentiments in said messages, and each including behavior profiles continuously updated from ongoing message classifications;

wherein, such behavior profiles are used to decide on a more probable context, content, and sentiment of an instant message.

12 . The message context classifier for sorting out messages that match predefined fields-of-interest of claim 1 , further comprising:

a smart agent assigned for each group of peers of senders communicating context, content, and sentiments in said messages, and each including behavior profiles continuously updated from ongoing message classifications for them individually;

wherein, such behavior profiles are used to decide on a more probable context, content, and sentiment of an instant message.

13 . The message context classifier for sorting out messages that match predefined fields-of-interest of claim 1 , further comprising:

a smart agent assigned for each recipient receiving context, content, and sentiments in said messages, and each including behavior profiles continuously updated from ongoing message classifications received by them individually;

wherein, such behavior profiles are used to decide on a more probable context, content, and sentiment of an instant message.

14 . The message context classifier for sorting out messages that match predefined fields-of-interest of claim 1 , further comprising:

a smart agent assigned for each group of peers of recipients receiving context, content, and sentiments in said messages, and each including behavior profiles continuously updated from ongoing message classifications received by them individually;

wherein, such behavior profiles are used to decide on a more probable context, content, and sentiment of an instant message.

15 . The message context classifier for sorting out messages that match predefined fields-of-interest of claim 1 , further comprising:

means for identifying individual senders of said messages and profiling their behaviors and classifications of previous messages;

means for weighting an estimate of a most probable subject category and field-of-interest for an instant missive according to a behavior profile of corresponding individual senders;

means for identifying individual recipients of said messages and profiling their behaviors and classifications of previous messages directed to them;

means for weighting an estimate of a most probable subject category and field-of-interest for an instant missive according to a behavior profile of corresponding individual recipients;

means for identifying peer groups of senders of said messages and profiling their behaviors and classifications of previous messages sent by any of them;

means for weighting an estimate of a most probable subject category and field-of-interest for an instant missive according to a behavior profile of corresponding senders in their relevant peer group;

means for ranking said constituent words by their frequencies of common usage that exist for particular and predefined subject categories and fields-of-interest;

means for estimating a most probable subject category and field-of-interest for an instant missive using word rankings;

means for ranking said constituent words as always used, commonly used, rarely used, and never used by a sorting their frequencies of common usage that exist for particular and predefined subject categories and fields-of-interest;

means for estimating a most probable subject category and field-of-interest for an instant missive using word rankings;

means for ranking said constituent words as always-used, commonly-used, rarely-used, and never-used by a sorting their frequencies of common usage that exist for particular and predefined subject categories and fields-of-interest;

means for eliminating from further consideration a subject category and field-of-interest for an instant missive if a word ranking is returned as never used, wherein, a reduction of the uncertainty is obtained about the subject category and field-of-interest for an instant missive;

means for continuously weighing said constituent words from always-used to never-used according to statistics of their frequencies in common usage that exist for particular and predefined subject categories and fields-of-interest;

means for eliminating from further consideration a subject category and field-of-interest for an instant missive if the weight of a constituent word is returned as not exceeding an adjustable threshold, wherein, a reduction of the uncertainty is obtained about the subject category and field-of-interest for an instant missive;

a smart agent assigned for each word in a vocabulary of words expected to be encountered in said messages, and each including attributes describing alternative meanings that can be ascribed to the particular word;

a smart agent assigned for each context, content, and sentiment that can be conveyed by words expected to be encountered in said messages, and each including attributes describing alternatives that can be ascribed to it;

a smart agent assigned for each sender communicating context, content, and sentiments in said messages, and each including behavior profiles continuously updated from ongoing message classifications, wherein, such behavior profiles are used to decide on a more probable context, content, and sentiment of an instant message;

a smart agent assigned for each group of peers of senders communicating context, content, and sentiments in said messages, and each including behavior profiles continuously updated from ongoing message classifications for them individually, wherein, such behavior profiles are used to decide on a more probable context, content, and sentiment of an instant message;

a smart agent assigned for each recipient receiving context, content, and sentiments in said messages, and each including behavior profiles continuously updated from ongoing message classifications received by them individually, wherein, such behavior profiles are used to decide on a more probable context, content, and sentiment of an instant message; and

a smart agent assigned for each group of peers of recipients receiving context, content, and sentiments in said messages, and each including behavior profiles continuously updated from ongoing message classifications received by them individually, wherein, such behavior profiles are used to decide on a more probable context, content, and sentiment of an instant message.

16 . (canceled)

17 . An artificial intelligence system, comprising:

a computer network server connected to receive and analyze millions of simultaneous text and/or voice messages written by humans to be read and understood by humans;

means for key, or otherwise important words in sentences to be recognized and arrayed, wherein each such word is contributed to a qualia generator that spawns the word into its possible contexts, themes, or other reasonable ambiguities that can exist at the level of sentences, paragraphs, and missives;

a thesaurus-like table for expanding each word into a spread of discrete definitions, wherein, several such spreads are used as templates on the others to find petals that exhibit a convergence of meaning; and

means for deconstructing each paragraph is into sub-contexts that are appropriate within the overall theme once the context of a whole missive has been predicted;

wherein, particular contexts identified are useful to trigger an actionable output.