Cascading-type composition generating method

A cascading, compositional technology, applied in unstructured text data retrieval, text database clustering/classification, instruments, etc.

Active Publication Date: 2018-04-27
语仓科技(北京)有限公司
View PDF2 Cites 24 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Therefore, it is difficult to directly use existing topic analy

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
  • Cascading-type composition generating method
  • Cascading-type composition generating method
  • Cascading-type composition generating method

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0030] Specific implementation mode 1: This implementation mode provides a method for generating a cascading composition, including:

[0031] Step 1. According to the input material composition question stem, extract the topic word set.

[0032] For a given composition topic (topic), we first need to clarify what kind of composition the topic (topic) requires us to write. Through the analysis of the stems of the composition of the college entrance examination, the stems of the questions can be divided into the following three categories:

[0033] ① Topic composition. In the question stem, it is clearly required what topic the essay should meet. For example, please write an essay on the topic of "voice".

[0034] ②Propositional and semi-propositional composition. The title of the composition is clearly required in the question stem (propositional composition), or a part of the topic is given, and candidates need to incomplete the topic (semi-propositional composition). For ...

specific Embodiment approach 2

[0060] Specific implementation mode two: the difference between this implementation mode and specific implementation mode one is that step one specifically includes:

[0061] Step one one, the word vector W in each sentence in the first training corpus i Input to the first GRU model to get the sentence vector representation of each sentence

[0062] Step 12. Represent the sentence vector of each sentence Input to the second GRU model to obtain the vector representation V of the entire text doc ;

[0063] Step 13. Express the vector of the entire text as V doc And the pre-built vocabulary is input to the decoder, and the softmax function is used to predict a probability value for each word in the vocabulary, which is used to represent the probability that each word contains text semantics;

[0064] Step 14: Select a word set consisting of all words exceeding a certain threshold from the predicted probability value as the topic word set W.

[0065] Specifically, this emb...

specific Embodiment approach 3

[0074] Specific implementation mode three: the difference between this implementation mode and specific implementation mode one or two is that step two specifically includes:

[0075] Step 21. Sentence extraction steps:

[0076] For each word w in the given set of all topic words W, find all sentences containing w in the second training corpus, and put all the sentences found by all words into the set S, for the set S For each sentence of , obtain the vector representation of the sentence; the method for obtaining the sentence vector representation is: obtain the word embedding of each word in the sentence, and then average these word embeddings in each dimension;

[0077] Find the vector representation of the topic word set.

[0078] Make cosine similarity between the vector representation of the topic word set and each sentence vector in the set S respectively.

[0079] Select the sentence with the highest similarity score in each type of sentence as the sentence to be exp...

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 relates to a cascading-type composition generating method. The technical purpose is to make up for the insufficiencies in the prior art that only composition scoring is studied, and there are no studies on a composition generating method yet, and it is hard to analyze a title of a composition through an existing subject analysis technology. One or more topic words are utilized to represent a main idea of a to-be-generated composition; after the topic words are obtained, composition generation is broken up into the topic word expansion, sentence extraction and the text organization; after the topic words are expanded, a sentence extracting module is utilized to look for sentences relevant to the topic words, finally a text organization module is utilized to rank the extractedsentences, so that the sentences become a coherent entirety. By means of the cascading-type composition generating method, phrases can also be mined from an extracted sentence set to make supplement for existing topic words. The cascading-type composition generating method is applicable to automatically generating compositions.

Description

technical field [0001] The invention relates to the technical field of theme analysis, in particular to a method for generating cascading compositions. Background technique [0002] In the prior art, most of the researches related to composition focus on automatic composition scoring, and there is no research on composition generation methods. The primary problem of composition generation is to analyze the theme of the composition. Existing theme analysis techniques are all based on a large number of text collections, and mainly extract the topic information or factual topic information on the surface of the article. The composition topic is generally short, and it is required to extend the topic to get the deep theme contained in the topic. Therefore, it is difficult to directly use existing topic analysis techniques to conduct topic analysis for composition topics. Contents of the invention [0003] The purpose of the present invention is to solve the shortcomings that...

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
IPC IPC(8): G06F17/27G06F17/30
CPCG06F16/35G06F40/211G06F40/35
Inventor 秦兵孙承杰冷海涛刘挺
Owner 语仓科技(北京)有限公司
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