Abstract generation method and device
A technology for summarization and sentence generation, applied in the field of information processing, it can solve the problems of not considering word frequency vectors, not covering sentence information, affecting sentence information, etc., to achieve the effect of improving the accuracy and coverage of generalization
Inactive Publication Date: 2018-06-19
NEUSOFT CORP
View PDF10 Cites 4 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0003] Because the TF-IDF-based abstract extraction technology calculates the words in the sentence through TF-IDF, and takes topN as the score of the abstract sentence, especially when the sentence is long, it cannot consider all the word frequency vectors. If the words with low word frequency in the sentence are more However, discarding all these low-frequency words will affect the information contained in the sentence. Therefore, the abstract extracted in this way often cannot cover the sentence information, which affects the summary accuracy and coverage of the content of the document center.
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0077] Embodiments of the present invention are described in detail below, examples of which are shown in the drawings, wherein the same or similar reference numerals designate the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the figures are exemplary and are intended to explain the present invention and should not be construed as limiting the present invention.
[0078] The method and device for generating an abstract according to the embodiments of the present invention will be described below with reference to the accompanying drawings.
[0079] An abstract is a concise and coherent essay that accurately and comprehensively reflects the central content of a document. Currently, related abstract extraction techniques are usually based on TF-IDF to score sentences.
[0080] Because the TF-IDF-based abstract extraction technology calculates the words in the sentence through TF-IDF, and ta...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention provides an abstract generation method and device. The method comprises the steps of obtaining a sentence vector of each sentence, selecting candidate abstract sentences from all the sentences according to the sentence vectors, acquiring first scores of the candidate abstract sentences, selecting abstract sentences from the candidate abstract sentences according to the first scores,and further using the abstract sentences to generate an abstract of an article. Since the sentence vectors can better preserve sentence information, the first scores indicate the sentence quality of the candidate abstract sentences in a unit length, and double screening is carried out on all the sentences in the article through the sentence vectors and the first scores to obtain the abstract sentences, so that the accuracy of summarizing the main content of the article through the abstract is improved, and the coverage degree is increased.
Description
technical field [0001] The present invention relates to the technical field of information processing, in particular to a method and device for generating an abstract. Background technique [0002] An abstract is a concise and coherent essay that accurately and comprehensively reflects the central content of a document. At present, related abstract extraction technology is usually based on Term Frequency-Inverse Document Frequency (TF-IDF) to calculate the score of words, so as to score the sentence by the words in the sentence. [0003] Because the TF-IDF-based abstract extraction technology calculates the words in the sentence through TF-IDF, and takes topN as the score of the abstract sentence, especially when the sentence is long, it cannot consider all the word frequency vectors. If the words with low word frequency in the sentence are more However, discarding all these low-frequency words will affect the information contained in the sentence. Therefore, the abstract e...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06F17/30G06F17/27
CPCG06F16/345G06F40/279
Inventor 杜森
Owner NEUSOFT CORP