Title selection method and system of read robot

A robot and title technology, applied in the field of natural language processing research, can solve problems such as time-consuming, data sparsity, complexity, and insignificant effect improvement, and achieve the effect of reducing reasoning costs.

Active Publication Date: 2018-03-23
SHANXI UNIV
View PDF3 Cites 11 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] The feature-based method needs to manually construct a large number of different features, which consumes a lot of time. Although the end-to-end neural network model can automatically learn features, due to the sparsity of data and the complexity of the problem, the effect improvement is not obvious.

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 more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Title selection method and system of read robot
  • Title selection method and system of read robot
  • Title selection method and system of read robot

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0054] figure 2 Concrete flowchart for the relationship construction unit of the present invention. image 3 It is a schematic diagram of solving the reading comprehension title multiple-choice questions of the present invention. This embodiment reads the title selection method of the robot, and the specific steps are as follows:

[0055] 101. Obtain reading comprehension chapters, question stems and their options (such as Image 6 shown).

[0056] 102. Paragraph topic sentence extraction method based on relevant factors Extract the topic sentences of each paragraph in the reading comprehension text to obtain the main points of the text. It comprehensively evaluates the value of the sentence reflecting the theme, so as to more accurately extract the topic sentence of the paragraph and obtain the main points of the chapter.

[0057] 103. According to the chapter points obtained before, use the pre-trained Word2Vec word vector to represent the obtained chapter points and op...

Embodiment 2

[0072] The title selection system of the reading robot in this embodiment includes five parts: a chapter point extraction unit, a title and chapter point correlation analysis unit, a title structure analysis unit, a correlation matrix and title structure fusion unit, and an answer selection unit.

[0073] The chapter key point extraction unit is used to obtain the options corresponding to the reading comprehension chapter, question stem, and question stem. The method of extracting the topic sentence of the paragraph based on relevant factors extracts the topic sentence of each paragraph in the reading comprehension chapter. The topic sentence of each paragraph is a chapter respectively. main points;

[0074] The correlation analysis unit between the title and the main points of the chapter is used to analyze the correlation between the title and the main points of the chapter, that is, to calculate the similarity value of each title and each main point of the chapter respective...

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

No PUM Login to view more

Abstract

The invention belongs to the field of natural language processing research, and discloses a title selection method and system of a read robot. The title selection method includes the following steps:1, a chapter main point extraction unit; 2, a title and chapter main point correlation analysis unit; 3, a title structure unit; 4, a relevancy matrix and title structure fusion unit; and 5, an answerselection unit for sorting options and selecting a suitable answer. The title selection method provides a tile and chapter main point correlation analysis method firstly, can construct a title and chapter main point based relevancy matrix by analyzing the relevancy between a title and a chapter main point, and is integrated with a title structure feature on the basis, and can select a title mostrelevant with a chapter. The method complies with a solution law, can solve the problem of title selection and question answering of the read robot, and can effectively fill in the gap of intelligentanswering of Chinese reading comprehension title selection questions.

Description

technical field [0001] The invention belongs to the field of natural language processing research, and in particular relates to a title selection method and system for a reading robot. Background technique [0002] The research and development of reading robots has become a research hotspot in recent years, attracting more and more attention from scholars and companies at home and abroad, and has achieved good development. [0003] In 2011, an IBM supercomputer (Watson) defeated the two most outstanding human contestants on the quiz show (Jeopardy). Todai Robot, an artificial intelligence robot developed by the National Institute of Informatics in Japan, has been able to pass the university entrance examination. In the 2015 Japanese university entrance examination, Todai Robot scored 511 points, 90 points higher than the average score of candidates. More points. In 2016, Google's artificial intelligence robot "AlphaGo" defeated the human representative Go player Li Shishi ...

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
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G06F17/27
CPCG06F40/258
Inventor 李茹关勇郭少茹张旗王智强柴清华
Owner SHANXI UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap