Method and apparatus for summarizing one or more text messages using indicative summaries
View Patent ↗A method and apparatus are provided for summarizing a text message, such as an email message or a transcribed audio message. A portion of each text message, such as a sentence, is extracted as an indicative summary of the text message based on a degree of overlap of words in the sentence with a set of words, such as words in the message subject or words in a related root message. The extracted portion is based on a score for each portion of the text message, such as a sentence. An interface is also provided for presenting the indicative summaries of a set of related text messages to a user.
1. A method comprising:
receiving, at a server, a message comprising a subject, a first sentence and a second sentence;
processing the message to yield a processed subject based on the subject, a processed first sentence based on the first sentence, and a processed second sentence based on the second sentence, wherein processing the message comprises identifying words in the message that are of a predefined word type; and
selecting exactly one of the first sentence or the second sentence as a summary text, which summarizes the message, based on:
(i) a first number of words of the predefined word type in the processed first sentence;
(ii) a second number of words of the predefined word type in the processed second sentence;
(iii) first overlapping words, of the predefined word type, occurring both in the processed subject and the processed first sentence; and
(iv) second overlapping words, of the predefined word type, occurring both in the processed subject and the processed second sentence.
2. The method of claim 1 , wherein the summary text is a first summary text, and wherein processing the message further yields a processed message based on the message, the method further comprising:
receiving a follow-up message to the message, the follow-up message comprising a third sentence and a fourth sentence;
processing the follow-up message to yield a processed third sentence based on the third sentence and a processed fourth sentence based on the fourth sentence; and
selecting only one of the third sentence or the fourth sentence as a second summary text, which summarizes the follow-up message, based on:
(i) third overlapping words, of the predefined word type, occurring both in the processed message and the processed third sentence; and
(ii) fourth overlapping words, of the predefined word type, occurring both in the processed message and the processed fourth sentence.
3. The method of claim 1 , wherein the predefined word type comprises an information-bearing word type.
4. The method of claim 3 , wherein the information-bearing word type comprises one of verbs, nouns, adjectives, or adverbs.
5. The method of claim 1 , further comprising selecting exactly one of the first sentence or the second sentence for each of a plurality of messages.
6. The method of claim 5 , wherein the plurality of messages is a message thread.
7. The method of claim 1 , wherein processing the message further comprises substituting a word in the message with a non-inflected lexical form of the word.
8. The method of claim 1 , wherein selecting the exactly one of the first sentence or the second sentence is further based on:
(v) a third number of the first overlapping words; and
(vi) a fourth number of the second overlapping words.
9. The method of claim 8 , wherein selecting the exactly one of the first sentence or the second sentence comprises:
computing a first score for the first sentence, the first score equaling the third number divided by the first number;
computing a second score for the second sentence, the second score equaling the fourth number divided by the second number;
when the first score is higher than the second score, selecting the first sentence as the summary text; and
when the second score is higher than the first score, selecting the second sentence as the summary text.
10. The method of claim 1 , wherein processing the message further comprises removing one of a quotation or a signature block from the message.
11. The method of claim 1 , wherein the message is a transcribed audio message.
12. A system comprising:
a processor;
a computer-readable storage medium storing instructions which, when executed by the processor, cause the processor to perform operations comprising:
receiving a message comprising a subject, a first sentence and a second sentence;
processing the message to yield a processed subject based on the subject, a processed first sentence based on the first sentence, and a processed second sentence based on the second sentence, wherein processing the message comprises identifying words in the message that are of a predefined word type; and
selecting exactly one of the first sentence or the second sentence as a summary text, which summarizes the message, based on:
(i) a first number of words of the predefined word type in the processed first sentence;
(ii) a second number of words of the predefined word type in the processed second sentence;
(iii) first overlapping words, of the predefined word type, occurring both in the processed subject and the processed first sentence; and
(iv) second overlapping words, of the predefined word type, occurring both in the processed subject and the processed second sentence.
13. The system of claim 12 , wherein the summary text is a first summary text, wherein processing the message further yields a processed message based on the message, and wherein the computer-readable storage medium stores additional instructions which, when executed by the processor, perform further operations comprising:
receiving a follow-up message to the message, the follow-up message comprising a third sentence and a fourth sentence;
processing the follow-up message to yield a processed third sentence based on the third sentence and a processed fourth sentence based on the fourth sentence; and
selecting only one of the third sentence or the fourth sentence as a second summary text, which summarizes the follow-up message, based on:
(i) third overlapping words, of the predefined word type, occurring both in the processed message and the processed third sentence; and
(ii) fourth overlapping words, of the predefined word type, occurring both in the processed message and the processed fourth sentence.
14. The system of claim 12 , wherein the predefined word type comprises an information-bearing word type.
15. The system of claim 14 , wherein the information-bearing word type comprises one of verbs, nouns, adjectives, or adverbs.
16. The system of claim 12 , wherein the computer-readable storage medium stores additional instructions which, when executed by the processor, perform further operations comprising selecting exactly one of the first sentence or the second sentence for each of a plurality of messages.
17. The system of claim 16 , wherein the plurality of messages is a message thread.
18. The system of claim 12 , wherein processing the message further comprises substituting a word in the message with a non-inflected lexical form of the word.
19. The system of claim 12 , wherein selecting the exactly one of the first sentence or the second sentence is further based on:
(v) a third number of the first overlapping words;
(vi) a fourth number of the second overlapping words.
20. The system of claim 19 , wherein selecting the exactly one of the first sentence or the second sentence comprises:
computing a first score for the first sentence, the first score equaling the third number divided by the first number;
computing a second score for the second sentence, the second score equaling the fourth number divided by the second number;
when the first score is higher than the second score, selecting the first sentence as the summary text; and
when the second score is higher than the first score, selecting the second sentence as the summary text.
21. The system of claim 20 , wherein processing the message further comprises removing one of a quotation or a signature block from the message.
22. A computer-readable storage device storing instructions which, when executed by a processor, to cause the processor to perform operations comprising:
receiving, at a server, a message comprising a subject, a first sentence and a second sentence;
processing the message to yield a processed subject based on the subject, a processed first sentence based on the first sentence, and a processed second sentence based on the second sentence, wherein processing the message comprises identifying words in the message that are of a predefined word type; and
selecting exactly one of the first sentence or the second sentence as a summary text, which summarizes the message, based on:
(i) a first number of words of the predefined word type in the processed first sentence;
(ii) a second number of words of the predefined word type in the processed second sentence;
(iii) first overlapping words, of the predefined word type, occurring both in the processed subject and the processed first sentence; and
(iv) second overlapping words, of the predefined word type, occurring both in the processed subject and the processed second sentence.