Systems and methods for interactive topic-based text summarization

• US 7,451,395 B2
• Filed: 12/16/2002
• Issued: 11/11/2008
• Est. Priority Date: 12/16/2002
• Status: Expired due to Fees

• Current Assignee: Palo Alto Research Center, Inc. (Xerox Holdings Corp.)
• Original Assignee: Palo Alto Research Center, Inc. (Xerox Holdings Corp.)
• Inventors: Chen, Francine R., Zaenen, Annie E., Brants, Thorsten H.
• Primary Examiner(s): Hong; Stephen
• Assistant Examiner(s): Patel; Manglesh M
• Application Number:
  US10/319,508
• Publication Number:
  US 20040122657A1
• Time in Patent Office: 2,157 Days
• Field of Search: 715/530, 715/903, 715/501.1, 715/254, 715/200, 715/255
• US Class Current: 715/254
• CPC Class Codes: G06F 40/131 Fragmentation of text files, e.g. creating reusable text-blocks; Linking to fragments, e.g. using XInclude; Namespaces

Abstract

Techniques for determining interactive topic-based summarization are provided. A text to be summarized is segmented. Discrete keyword, key-phrase, n-gram, sentence and other sentence constituent based summaries are generated based on statistical measures for each text segment. Interactive topic-based summaries are displayed with human sensible omitted text indicators such as alternate colors, fonts, sounds, tactile elements or other human sensible display characteristics useful in indicating omitted text. Individual and/or combinations of discrete keyword, key-phrase, n-gram, sentence, noun phrase and sentence constituent based summaries are dynamically displayed to provide an overview of topic and subtopic development within a text. A hierarchical and interactive display of texts based on the use of discrete sentence constituent based summaries which associates expansible and contractible displayed text provides contextualized access to an interactive topic-based text summary and to an original text.

Claims

1. A method of determining interactive topic-based text summaries comprising the steps of:

dynamically determining a text to summarize;

dynamically determining at least one text segment of the determined text to summarize;

dynamically determining at least one sentence constituent based summary type; and

dynamically determining at least one discrete interactive topic-based text summary for a determined segment based on the determined sentence constituent based summary type and a statistical measure of text segment characterization and text segment differentiation;

dynamically determining a subtopic development by recursively repeating the step of dynamically determining at least one sentence constituent based summary type and the step of dynamically determining at least one discrete interactive topic-based text summary, the subtopic development arranged as a hierarchy including at least one subtopic of the at least one discrete interactive topic-based text summary;

dynamically determining at least one text segment related to a subtopic of the topic-based text summary to summarize based on dynamic user selections;

storing in an electronic memory storage at least one discrete interactive topic-based text summary based on the at least one text segment of the determined text and the at least one determined sentence constituent based summary type;

storing the subtopic development in the electronic memory storage, and

dynamically displaying at least one discrete interactive subtopic-based text summary based on the at least one text segment related to a subtopic of the topic-based text summary and wherein a human sensible display characteristic indicates omitted text for the subtopic-based text summary.

2. The method of claim 1, wherein the statistical measure is a measure of determined segment characterization and differentiation from at least one other segment.

3. The method of claim 2, wherein the at least one other segment is at least one of the n-preceding and m-following segments and where n+m>

0.

4. The method of claim 2, wherein the sentence constituent based summary type is at least one of a keyword, a key-phrase, an n-gram, a sentence and a noun phrase.

5. The method of claim 2, wherein the measure of text region characterization and other region differentiation is at least one of mutual information, log odds ratio, X², G² distributions.

6. The method of claim 1, wherein the sentence constituent based summary type is determined based on at least one of a device characteristic, a user preference and a user selection.

7. The method of claim 5, further comprising the steps of:

dynamically determining at least one text segment to summarize based on dynamic user selections;

dynamically displaying at least one discrete interactive topic-based text summary based on the at least one dynamically determined text segment and wherein a human sensible display characteristic indicates omitted text;

dynamically determining at least one text segment related to a subtopic of the topic-based text summary to summarize based on dynamic user selections; and

dynamically displaying at least one discrete interactive subtopic-based text summary based on the at least one text segment related to a subtopic of the topic-based text summary and wherein a human sensible display characteristic indicates omitted text for the subtopic-based text summary.

8. The method of claim 7, wherein the human sensible display characteristic is at least one of a visual, auditory, tactile, olfactory and taste characteristic.

9. The method of claim 8, wherein the at least one visual characteristic is at least one of a text font, color, italics, bolding and placement.

10. The method of claim 1, wherein the statistical measure is at least one of four components of mutual information based on presence in a segment and presence in other segments.

11. The method of claim 1, wherein the statistical measure is at least two of four components of mutual information based on presence in a segment and presence in other segments.

12. The method of claim 1, further comprising the step of normalizing key-phrase probabilities based on key-phrase length.

13. A system for determining interactive topic-based text summaries comprising:

an input/output circuit for receiving, a dynamically selected sentence constituent based summary type, a dynamically selected text to summarize;

a processor that retrieves the selected text from the input/output circuit, segments the text into at least one text segment, dynamically determines text segments to summarize into discrete interactive topic-based summaries, and recursively determines segments of the text segments to summarize into discrete interactive subtopic-based summaries;

a probability and latent class determining circuit for determining distributions of the selected sentence constituent type in the segmented text and in at least one text segment;

a circuit for dynamically determining at least one text segment related to a subtopic of the topic-based text summary to summarize based on dynamic user selections;

a statistical measure determining circuit for determining a statistical measure of at least one text segment; and

wherein the processor determines and prepares for display a discrete interactive topic-based summary for at least one text segment based on the received sentence constituent based summary type, the determined distributions and a statistical measure of the segment; and

a display circuit for dynamically displaying at least one discrete interactive subtopic-based text summary based on the at least one text segment related to a subtopic of the topic-based text summary and wherein a human sensible display characteristic indicates omitted text for the subtopic-based text summary.

14. The system of claim 13, wherein the statistical measure is a measure of text segment characterization and other segment differentiation.

15. The system of claim 14, wherein other segments are at least one of n-preceding text segments of a determined text segment and m-following text segments and where n+m>

0.

16. The system of claim 14, wherein the sentence constituent based summary type is at least one of a keyword, a key-phrase, an n-gram, a sentence and a noun phrase.

17. The system of claim 14, wherein the measure of text segment characterization and other segment differentiation is at least one of mutual information, log odds ratio, X², G² distributions.

18. The system of claim 13, wherein the sentence constituent based summary type is determined based on at least one of a device characteristic, a user preference and a user selection.

19. The system of claim 17, further comprising:

a display circuit that displays at least one discrete interactive topic-based text summary based on a dynamic selection of the selected text segment, uses a human sensible display characteristic to indicate omitted text, displays at least one discrete interactive subtopic-based text summary based on the topic-based text summary, and uses the human sensible display characteristic to indicate omitted text for the subtopic-based text summary.

20. The system of claim 19, wherein the human sensible display characteristic is at least one of a visual, auditory, tactile, olfactory and taste characteristic.

21. The system of claim 20, wherein the visual characteristic is at least one of text font, text color, text italics, text bolding and placement.

22. The system of claim 13, wherein the statistical measure is at least one of four components of mutual information based on presence in a segment and presence in other segments.

23. The system of claim 13, wherein the statistical measure is at least two of four components of mutual information based on presence in a segment and presence in other segments.

24. The system of claim 13, further comprising the step of:

normalizing key-phrase probabilities based on key-phrase length.

25. Computer readable storage medium comprising:

computer readable program code embodied on the computer readable storage medium, the computer readable program code usable to program a computer to a method of determining interactive topic-based text summaries comprising the steps of;

dynamically determining a text to summarize;

dynamically determining at least one text segment of the determined text to summarize;

dynamically determining at least one sentence constituent based summary type;

dynamically determining at least one discrete interactive topic-based text summary based on the determined sentence constituent based summary type and a statistical measure of text segment characterization and text segment differentiation;

dynamically determining a subtopic development by recursively repeating the step of dynamically determining at least one sentence constituent based summary type and the step of dynamically determining at least one discrete interactive topic-based text summary, the subtopic development arranged as a hierarchy including at least one subtopic of the at least one discrete interactive topic-based text summary;

dynamically determining at least one text segment related to a subtopic of the topic-based text summary to summarize based on dynamic user selections;

storing in an electronic memory storage at least one discrete interactive topic-based text summary based on the at least one text segment of the determined text and the at least one determined sentence constituent based summary type;

storing the subtopic development in the electronic memory storage; and

dynamically displaying at least one discrete interactive subtopic-based text summary based on the at least one text segment related to a subtopic of the topic-based text summary and wherein a human sensible display characteristic indicates omitted text for the subtopic-based text summary.

26. A method of determining interactive topic-based text summaries comprising the steps of:

dynamically determining a text to summarize;

dynamically determining at least one text segment to summarize based on dynamic user selections;

dynamically determining at least one sentence constituent based summary type; and

dynamically determining at least one discrete interactive topic-based text summary for a determined segment based on the determined sentence constituent based summary type and a statistical measure of text segment characterization and text segment differentiation;

dynamically determining a subtopic development by recursively repeating the step of dynamically determining at least one sentence constituent based summary type and the step of dynamically determining at least one discrete interactive topic-based text summary, the subtopic development arranged in a hierarchical order including at least one subtopic of the at least one discrete interactive topic-based text summary;

dynamically determining at least one text segment related to a subtopic of the topic-based text summary to summarize based on dynamic user selections;

storing in an electronic memory storage at least one discrete interactive topic-based text summary based on the at least one text segment of the determined text and the at least one determined sentence constituent based summary type;

storing the subtopic development in the electronic memory storage;

dynamically displaying at least one discrete interactive topic-based text summary based on the at least one dynamically determined text segment and wherein a first human sensible display characteristic placed by the discrete interactive topic-based text summary indicates omitted text for the topic-based text summary;

dynamically displaying the omitted text for the topic-based text summary based on selection by a user of the first human sensible display characteristic placed by the discrete interactive topic-based text summary; and

dynamically displaying in the hierarchical order the subtopic development, the step of dynamically displaying the subtopic development comprising;

dynamically displaying, by each displayed at least one discrete interactive topic-based text summary, each discrete interactive subtopic-based text summary based on the at least one text segment related to a subtopic of the topic-based text summary, wherein a second human sensible display characteristic placed by the discrete interactive subtopic-based text summary indicates omitted text for the subtopic-based text summary; and

dynamically displaying the omitted text for the subtopic-based text summary based on selection by the user of the second human sensible display characteristic placed by the discrete interactive subtopic-based text summary.