.png){kind=icon}
返回
Generic framework for multilingual short text categorization using convolutional neural network
.png){kind=icon}
PRE
AI
DOI：10.1007/s11042-020-10314-9.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Online social media is a powerful source of information that can influence users' decisions. Due to the huge volume of data generated by such media, many researches have been done to automate text categorization. However, finding useful information to satisfy user's needs is not an easy task. There are many challenges to overcome especially in short text categorization that in addition to being a time-consuming and costly process, short messages have misspellings, typos, irony words and lack of context. To solve these challenges, this article proposes GM-ShorT, a Generic framework for Multilingual Short Text Categorization based on Convolutional Neural Network (CNN). For this, GM-ShorT collects online social media data. Such data were used as input to CNN that is combined with a word embedding mechanism to categorize short text messages. We explored several architectures for CNN and show that GM-ShorT can be used in multilingual Short text categorization with an accuracy of 13.58% higher when compared to other classical approaches.
Keyword:
Convolutional neural network
Text categorization
Multilingual text
Social media
期刊
IF:
3
论文数:
2.0W
被引数:
3.2W
机构
引用论文
Twitter permeability to financial events: an experiment towards a model for sensing irregularities
OAAI
OAAI分享收藏
A survey of multilingual human-tagged short message datasets for sentiment analysis tasks.png)
SOFT COMPUTING
SOFT COMPUTINGIF2.5
PREAI
分享收藏
查看更多内容