Generating a Summary Based on Readability

• US 20140324883A1
• Filed: 04/25/2013
• Published: 10/30/2014
• Est. Priority Date: 04/25/2013
• Status: Active Grant

• Current Assignee: Micro Focus LLC (Open Text Corporation)
• Original Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Inventors: Deolalikar, Vinay
• Granted Patent: US 9,727,641 B2
• Time in Patent Office: Days
• US Class Current: 707/748
• CPC Class Codes: G06F 16/345 Summarisation for human users

Abstract

A technique to generate a summary of a set of sentences. Each sentence in the set can be evaluated based on a criterion, such as informativeness of the sentence. The sentences may also be evaluated for readability based on a readability measure. Sentences can be selected for inclusion in the summary based on the evaluations.

Claims

1. A method for generating a readable summary, comprising:

extracting, using a processor, a set of sentences from a document;

scoring each sentence of the set of sentences based on an informativeness measure;

scoring each sentence of the set of sentences based on a readability measure; and

selecting sentences in the set for inclusion in a summary based on readability scores and informativeness scores.

2. The method of claim 1, further comprising ranking sentences in the set based on readability scores and informativeness scores.

3. The method of claim 2, wherein the sentences selected for inclusion in the summary are the top ranked sentences until a maximum limit for the summary is reached.

4. The method of claim 2, wherein ranking sentences in the set comprises:

identifying a subset of sentences having an informativeness score above a threshold; and

ranking the subset of sentences based on the sentences'"'"' readability scores.

5. The method of claim 1, further comprising:

identifying a sentence in the set having an informativeness score greater by a first threshold than the remaining sentences in the set but having a readability score below a second threshold;

generating a concatenated sentence by concatenating at least one contextual sentence to the identified sentence, the contextual sentence being a sentence preceding or following the identified sentence in the document; and

selecting the concatenated sentence for inclusion in the summary.

6. The method of claim 1, wherein the sentences are selected using a linear program optimization that maximizes informativeness and readability of the resulting summary as measured by the informative scores and readability scores of the sentences.

7. A system for generating a readable summary, comprising:

an extraction module to extract a plurality of sentences from a document;

a scoring module to identify sentences for inclusion in a summary of the document based on a criterion;

a readability module to evaluate the readability of the identified sentences according to a readability measure; and

a summarizer to select sentences based in part on the evaluated readability of the identified sentences.

8. The system of claim 7, wherein the scoring module is configured to assign a score to each identified sentence indicating its suitability for inclusion in the summary based on the criterion.

9. The system of claim 8, wherein the criterion is informativeness.

10. The system of claim 8, further comprising a ranking module to rank the identified sentences based on the evaluated readability and the assigned scores.

11. The system of claim 10, wherein the summarizer is configured to select sentences based on the ranking generated by the ranking module.

12. The system of claim 7, wherein the summarizer comprises a context module to add a contextual sentence to a selected sentence. wherein the contextual sentence is a sentence preceding or following the selected sentence in the document.

13. A non-transitory computer readable storage medium storing instructions that, when executed by a processor, cause a computer to:

evaluate each sentence of a set of sentences for informativeness;

evaluate each sentence of the set of sentences for readability based on a readability measure; and

generate a summary by adding sentences to the summary based on the informativeness and readability evaluations.

14. The storage medium of claim 13, further storing instructions that cause a computer to:

rank sentences in the set based on the informativeness and readability evaluations; and

select sentences for addition to the summary based on the ranking.

15. The storage medium of claim 13, wherein the instructions to rank the sentences cause the processor to:

identify a subset of sentences having evaluated informativeness above a threshold; and

rank the subset of sentences based on the readability evaluations of the subset of sentences.