Text emotion classification method based on the joint deep learning model

A technology of emotion classification and deep learning, which is applied in text database clustering/classification, unstructured text data retrieval, character and pattern recognition, etc. It can solve problems such as data sparseness and dimensionality disaster, and achieve improved classification effect and improved The effect of the classification effect

Inactive Publication Date: 2017-04-26
HARBIN INST OF TECH
View PDF4 Cites 172 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to solve the existing shallow classification methods such as SVM, which will bring prob...

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
  • Text emotion classification method based on the joint deep learning model
  • Text emotion classification method based on the joint deep learning model
  • Text emotion classification method based on the joint deep learning model

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0019] Specific implementation mode one: as figure 1 and figure 2 As shown, a text sentiment classification method based on a joint deep learning model includes the following steps:

[0020] Step 1: process each word in the text data, use the word2vec tool to train each word in the processed text data, and obtain a word vector dictionary, wherein each word corresponds to a word vector; the word2vec tool is Google Lyric Vector Tool;

[0021] Step 2: The original input is a word sequence (w 1 ,...,w s-1 ,w s ,...,w max ), get the matrix M of each sentence, M=(x 1 ,...,x s-1 ,x s ,...,x max )∈R d×max ; The LSTM layer trains the fixed-length matrix M into a fixed-dimensional vector (embedding) input layer to improve, and generates d-dimensional h word vectors with contextual semantic relations (the word vectors will follow the training process during the training process) and change); where the w 1 ,...,w s-1 ,w s ,...,w max is the word label corresponding to each ...

specific Embodiment approach 2

[0027] Specific embodiment two: the difference between this embodiment and specific embodiment one is that it is characterized in that: the specific process of processing each word in the text data in the step one is:

[0028] Use the word segmentation program (stutter word segmentation) to segment the sentences in the text data, remove special characters, and retain Chinese characters, English, punctuation and emoticons (special characters such as phonetic symbols, special symbols, etc.).

[0029] Other steps and parameters are the same as those in Embodiment 1.

specific Embodiment approach 3

[0030] Specific embodiment three: the difference between this embodiment and specific embodiment one or two is: the original input is word sequence (w 1 ,...,w s-1 ,w s ,...,w max )Specifically:

[0031] All the words in the text data form a word dictionary, give each word an id label, replace all the words in the text data with the id label, and get the word sequence (w 1 ,...,w s-1 ,w s ,...,w max ), as the input of the joint deep learning model.

[0032] Other steps and parameters are the same as those in Embodiment 1 or Embodiment 2.

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 provides a text emotion classification method based on the joint deep learning model which relates to the text emotion classification method. The method is designed with the object of solving the problems with the dimension disaster and sparse data incurred from the existing support vector machine and other shallow layer classification methods. The method comprises: 1) processing each word in the text data; using the word2vec tool to train each processed word in the text data so as to obtain a word vector dictionary; 2) obtaining the matrix M of each sentence; training the matrix M by the LSTM layer and converting it into vector with fixed dimensions; improving the input layer; generating d-dimensional h word vectors with context semantic relations; 3) using a CNN as a trainable characteristic detector to extract characteristics from the d-dimensional h word vectors with context semantic relations; and 4) connecting the extracted characteristics in order; outputting to obtain the probability of each classification wherein the classification with the maximal probability value is the predicated classification. The invention is applied to the natural language processing field.

Description

technical field [0001] The invention relates to a text sentiment classification method based on a joint deep learning model. Background technique [0002] Text sentiment classification, as the name suggests, is to determine whether an evaluation text expresses a commendatory or derogatory sentiment, where the text can be at the document level, sentence level or element level. Sentiment classification can be viewed as both a classification problem and a regression problem. If it is a regression problem, just do a mapping to the sentiment category after calculating the sentiment score of the text. When viewed as a classification problem, there are a large number of text classification techniques available, but compared with general text classification, sentiment classification has its own unique problems that need special treatment, and its unique characteristics are available. Sentiment analysis text can be at document level, sentence level or element level, and the present...

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): G06K9/62G06F17/30
CPCG06F16/35G06F18/217G06F18/2415
Inventor 徐冰杨沐昀杨艳赵铁军郑德权朱聪慧曹海龙
Owner HARBIN INST OF 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