ICFICE 2018
Current Result Document :
| 한글제목(Korean Title) |
A Convolutional Neural Network Approach for Classification of Text based on the Texts Collected from Web |
| 영문제목(English Title) |
A Convolutional Neural Network Approach for Classification of Text based on the Texts Collected from Web |
| 저자(Author) |
Satyabrata Aich
Sabyasachi Chakraborty
Kiwon Choi
Kim Hee-Cheol
|
| 원문수록처(Citation) |
VOL 10 NO. 01 PP. 0262 ~ 0264 (2018. 06) |
한글내용
(Korean Abstract) |
|
영문내용
(English Abstract) |
In the last few years the popularity of the deep learning model based approach for the text classification has increased because it was succefully used in the field of computer vision and speech recognition.Although the traditional approaches such as bag of words and bag of ngram approaches provides good result ,however for large text data the model error is higher in those approaches compared to the deep learning based model.Although there are different deep learnming models are available such as long-short term memory (LSTM) ,recurrent neural network (RNN),and convolutional neural nertwork (CNN) we have used CNN because of its simplicity and also provides high accuracy in the field of text classification . In this paper we have collected the texts from web in the different fields and applied the CNN based approaches to classify the text belongs to different category by tuning the hyperparameters and changing the sequence of text vectors.We found accuracies ranging from 85-92% based upon the hyparameter tuning and shuffling of the sequence of the text vectors.
|
| 키워드(Keyword) |
Text classification
Convolutional neural network
Deep learning
LSTM
|
| 파일첨부 |
PDF 다운로드
|
