System and method for automatic, unsupervised contextualized content summarization of single and multiple documents

1. A computer-implemented method for generating a context-sensitive summarization of a corpus of content, the context-sensitive summarization being performed by an automated summarization system executing on a hardware processor of an information processing system, the method comprising:

receiving an input corpus to the automated summarization system via a network;

processing the input corpus to extract phrases via the automated summarization system;

using the phrases to generate a grammatical analysis of the input corpus via the automated summarization system;

using the grammatical analysis to determine the thematic relevance of sentences in the input corpus via the automated summarization system;

ranking the sentences according to their thematic relevance to form a context-based ranking via the automated summarization system;

using the context-based ranking to construct a context graph for the phrases via the automated summarization system;

using the context graph to generate a content summarization of the input corpus via the automated summarization system, the content summarization comprising ranked sentences;

using a stochastic matrix combined with a damping factor to determine an order of the ranked sentences in the content summarization via the automated summarization system; and,

generating an output summary to the network via the automated summarization system, the output summary simplifying information search and limiting a need to access original source content.

2. The method of claim 1 , wherein:

sentences are ranked by performing a text ranking operation, the text ranking operation exploiting a source text structure to identify key phrases.

3. The method of claim 1 , wherein:

the damping factor comprises a random surfer model damping factor.

4. The method of claim 1 , further comprising:

limiting the number of ranked sentences in the content summarization according to a content summarization size limit.

5. The method of claim 1 , further comprising:

performing heuristic post-processing to remove duplicate and similar ranked sentences from the content summarization.

6. The method of claim 1 , wherein:

the input corpus comprises multiple documents; and

the ranked sentences in the content summarization reference their corresponding document in the input corpus.

7. A system comprising:

a hardware processor;

a data bus coupled to the processor; and

a computer-usable medium embodying computer program code, the computer-usable medium being coupled to the data bus, the computer program code used for generating a context-sensitive summarization of a corpus of content, the context-sensitive summarization being performed by an automated summarization system executing on the hardware processor of the system, and comprising instructions executable by the processor and configured for:

receiving an input corpus to the automated summarization system via a network;

processing the input corpus to extract phrases via the automated summarization system;

using the phrases to generate a grammatical analysis of the input corpus via the automated summarization system;

using the grammatical analysis to determine the thematic relevance of sentences in the input corpus via the automated summarization system;

ranking the sentences according to their thematic relevance to form a context-based ranking via the automated summarization system;

using the context-based ranking to construct a context graph for the phrases via the automated summarization system;

using the context graph to generate a content summarization of the input corpus via the automated summarization system, the content summarization comprising ranked sentences;

using a stochastic matrix combined with a damping factor to determine an order of the ranked sentences in the content summarization; and,

generating an output summary to the network via the automated summarization system, the output summary simplifying information search and limiting a need to access original source content.

8. The system of claim 7 , wherein:

sentences are ranked by performing a text ranking operation, the text ranking operation exploiting a source text structure to identify key phrases.

9. The system of claim 7 , wherein:

the damping factor comprises a random surfer model damping factor.

10. The system of claim 7 , further comprising:

limiting the number of ranked sentences in the content summarization according to a content summarization size limit.

11. The system of claim 7 , further comprising:

performing heuristic post-processing to remove duplicate and similar ranked sentences from the content summarization.

12. The system of claim 7 , wherein:

the input corpus comprises multiple documents; and

the ranked sentences in the content summarization reference their corresponding document in the input corpus.

13. A non-transitory, computer-readable storage medium embodying computer program code used for generating a context-sensitive summarization of a corpus of content, the context-sensitive summarization being performed by an automated summarization system executing on a hardware processor, the computer program code comprising computer executable instructions configured for:

receiving an input corpus to the automated summarization system via a network;

processing the input corpus to extract phrases via the automated summarization system;

using the phrases to generate a grammatical analysis of the input corpus via the automated summarization system;

using the grammatical analysis to determine the thematic relevance of sentences in the input corpus via the automated summarization system;

ranking the sentences according to their thematic relevance to form a context-based ranking via the automated summarization system;

using the context-based ranking to construct a context graph for the phrases via the automated summarization system;

using the context graph to generate a content summarization of the input corpus via the automated summarization system, the content summarization comprising ranked sentences; and,

using a stochastic matrix combined with a damping factor to determine an order of the ranked sentences in the content summarization; and,

generating an output summary to the network via the automated summarization system, the output summary simplifying information search and limiting a need to access original source content.

14. The non-transitory, computer-readable storage medium of claim 13 , wherein:

sentences are ranked by performing a text ranking operation, the text ranking operation exploiting a source text structure to identify key phrases.

15. The non-transitory, computer-readable storage medium of claim 13 , wherein:

the damping factor comprises a random surfer model damping factor.

16. The non-transitory, computer-readable storage medium of claim 13 , further comprising:

limiting the number of ranked sentences in the content summarization according to a content summarization size limit.

17. The non-transitory, computer-readable storage medium of claim 13 , further comprising:

performing heuristic post-processing to remove duplicate and similar ranked sentences from the content summarization.

18. The non-transitory, computer-readable storage medium of claim 13 , wherein:

the input corpus comprises multiple documents; and

the ranked sentences in the content summarization reference their corresponding document in the input corpus.

19. The non-transitory, computer-readable storage medium of claim 13 , wherein the computer executable instructions are deployable to a client system from a server system at a remote location.

20. The non-transitory, computer-readable storage medium of claim 13 , wherein the computer executable instructions are provided by a service provider to a user on an on-demand basis.