Short text emotion factor extraction method and device based on deep learning

A deep learning, short text technology, applied in the field of machine translation, which can solve the problems of poor object recognition results and low natural language in emotional evaluation

Active Publication Date: 2017-02-01
GLOBAL TONE COMM TECH
View PDF5 Cites 63 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0009] What the present invention aims to solve is that the understanding of natural language in existing computer systems is still at a relatively low stage, and the technical problem of poor recognition results of emotional evaluation objects

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
  • Short text emotion factor extraction method and device based on deep learning
  • Short text emotion factor extraction method and device based on deep learning
  • Short text emotion factor extraction method and device based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0023] Embodiment 1, a method for extracting emotional elements from short texts based on deep learning, such as figure 1 As shown, including: using a bidirectional long-short-term memory cycle neural network to model sentences, and then constructing a classifier for each category to classify; for the input sentence, each word in the sentence is represented as a word vector, Input into the recurrent neural network as an input sequence; calculate each hidden state in the recurrent neural network in turn, and calculate the feature representation of the sentence; after obtaining the feature representation of the sentence, use a logical classifier to classify the sentence and identify the sentence The categories of sentiment elements reviewed in . In the model, the sentence is modeled directly using the state of the hidden layer.

[0024] A short text emotional element extraction method based on deep learning, based on the deep learning method of neural network, can automatically...

Embodiment 2

[0025] Embodiment two, a method for extracting emotional elements of short texts based on deep learning, such as Figure 5-7 Shown, on the basis of embodiment one. Further includes:

[0026] better, such as Figure 5 As shown, the sequential calculation of each hidden state in the cyclic neural network, specifically, the calculation method of the hidden layer node at the tth moment is as follows, Among them, h t f is the hidden node value of the forward recurrent neural network, h t b is the hidden node value of the backward cyclic neural network, and the hidden node value at the last moment is selected as the vector representation of the sentence, that is Where c is the required sentence vector representation, and the colon represents vector splicing. In the present invention, this strategy is denoted as brnn-final. This method is one of the most straightforward strategies to obtain the overall representation of the sentence. Use the last moment state to capture al...

Embodiment 3

[0086] Embodiment three, a device for extracting emotional elements from short texts based on deep learning, such as Figure 8 As shown, it includes: a modeling unit, which is used to model sentences with a bidirectional long-short-term memory recurrent neural network, and then constructs a classifier for each category to classify; an input unit is used for input sentences. Each word in the sentence is expressed as a word vector, which is input into the cyclic neural network as an input sequence; the calculation unit is used to calculate each hidden state in the cyclic neural network in turn, and calculates the feature representation of the sentence; the classification unit , after obtaining the feature representation of the sentence, a logical classifier is used to classify the sentence and identify the category of the emotional element commented in the sentence.

[0087] A short text emotional element extraction device based on deep learning, based on the deep learning metho...

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 discloses a short text emotion factor extraction method and device based on deep learning and relates to the technical field of machine translation, aiming at solving the technical problems that the understanding on natural languages of an existing computer system is still at a relatively low phase and an identification result of an emotion evaluation object is not good. According to the technical scheme, the short text emotion factor extraction method comprises: modeling sentences through adopting a recurrent neural network of bidirectional long-time and short-time memory; constructing a classifier for each type and classifying; with regard to input sentences, taking each word in the sentences as a word vector, taking the word vector as an input sequence and inputting the word vector into the recurrent neural network; calculating each hide status in the recurrent neural network in sequence and calculating character representation of the sentences; and after obtaining the character representation of the sentences, classifying the sentences by adopting a logic classifier and identifying types of emotion factor commented in the sentences.

Description

technical field [0001] The present invention relates to the technical field of machine translation, in particular to a method and device for extracting emotional elements from short texts based on deep learning. Background technique [0002] With the large-scale popularization of Web2.0, various resources are growing exponentially, and the channels for people to communicate and exchange through the Internet are becoming more and more smooth. The Internet has undoubtedly become an important information carrier in the contemporary era, providing a broad platform for Internet users to publish, exchange and share their own opinions. More and more users choose to use various communication platforms on the Internet to share their views, life experience and work experience. As a result, a large amount of comment information has been generated on various network platforms. These information contain certain subjective emotions expressed by the publishers of the information. If we c...

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): G06F17/27G06F17/30
CPCG06F16/35G06F40/279G06F40/30
Inventor 程国艮巢文涵周庆
Owner GLOBAL TONE COMM TECH
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