Deep Sentiment Analysis: A Case Study on Stemmed Turkish Twitter Data
Sentiment analysis using stemmed Twitter data from various languages is an emerging research topic. In this paper, we address three data augmentation techniques namely Shift, Shuffle, and Hybrid to increase the size of the training data; and then we use three key types of deep learning (DL) models n...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2021-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9395633/ |
| _version_ | 1819315661846872064 |
|---|---|
| author | Harisu Abdullahi Shehu Md. Haidar Sharif Md. Haris Uddin Sharif Ripon Datta Sezai Tokat Sahin Uyaver Huseyin Kusetogullari Rabie A. Ramadan |
| author_facet | Harisu Abdullahi Shehu Md. Haidar Sharif Md. Haris Uddin Sharif Ripon Datta Sezai Tokat Sahin Uyaver Huseyin Kusetogullari Rabie A. Ramadan |
| author_sort | Harisu Abdullahi Shehu |
| collection | DOAJ |
| description | Sentiment analysis using stemmed Twitter data from various languages is an emerging research topic. In this paper, we address three data augmentation techniques namely Shift, Shuffle, and Hybrid to increase the size of the training data; and then we use three key types of deep learning (DL) models namely recurrent neural network (RNN), convolution neural network (CNN), and hierarchical attention network (HAN) to classify the stemmed Turkish Twitter data for sentiment analysis. The performance of these DL models has been compared with the existing traditional machine learning (TML) models. The performance of TML models has been affected negatively by the stemmed data, but the performance of DL models has been improved greatly with the utilization of the augmentation techniques. Based on the simulation, experimental, and statistical results analysis deeming identical datasets, it has been concluded that the TML models outperform the DL models with respect to both training-time (<italic>TTM</italic>) and runtime (<italic>RTM</italic>) complexities of the algorithms; but the DL models outperform the TML models with respect to the most important performance factors as well as the average performance rankings. |
| first_indexed | 2024-12-24T10:03:40Z |
| format | Article |
| id | doaj.art-a414a5901aac42dd9d9c1d3e45ea8be3 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-24T10:03:40Z |
| publishDate | 2021-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-a414a5901aac42dd9d9c1d3e45ea8be32022-12-21T17:00:54ZengIEEEIEEE Access2169-35362021-01-019568365685410.1109/ACCESS.2021.30713939395633Deep Sentiment Analysis: A Case Study on Stemmed Turkish Twitter DataHarisu Abdullahi Shehu0https://orcid.org/0000-0002-9689-3290Md. Haidar Sharif1https://orcid.org/0000-0001-7235-6004Md. Haris Uddin Sharif2https://orcid.org/0000-0002-1169-8438Ripon Datta3https://orcid.org/0000-0003-4738-2918Sezai Tokat4https://orcid.org/0000-0003-0193-8220Sahin Uyaver5https://orcid.org/0000-0001-8776-3032Huseyin Kusetogullari6https://orcid.org/0000-0001-5762-6678Rabie A. Ramadan7https://orcid.org/0000-0002-0281-9381School of Engineering and Computer Science, Victoria University of Wellington, Wellington, New ZealandCollege of Computer Science and Engineering, University of Hail, Hail, Saudi ArabiaDepartment of International Graduate Services, University of the Cumberlands, Williamsburg, KY, USADepartment of International Graduate Services, University of the Cumberlands, Williamsburg, KY, USADepartment of Computer Engineering, Pamukkale University, Denizli, TurkeyDepartment of Energy Science and Technologies, Turkish-German University, Istanbul, TurkeyDepartment of Computer Science, Blekinge Institute of Technology, Karlskrona, SwedenComputer Engineering Department, College of Engineering, Cairo University, Cairo, EgyptSentiment analysis using stemmed Twitter data from various languages is an emerging research topic. In this paper, we address three data augmentation techniques namely Shift, Shuffle, and Hybrid to increase the size of the training data; and then we use three key types of deep learning (DL) models namely recurrent neural network (RNN), convolution neural network (CNN), and hierarchical attention network (HAN) to classify the stemmed Turkish Twitter data for sentiment analysis. The performance of these DL models has been compared with the existing traditional machine learning (TML) models. The performance of TML models has been affected negatively by the stemmed data, but the performance of DL models has been improved greatly with the utilization of the augmentation techniques. Based on the simulation, experimental, and statistical results analysis deeming identical datasets, it has been concluded that the TML models outperform the DL models with respect to both training-time (<italic>TTM</italic>) and runtime (<italic>RTM</italic>) complexities of the algorithms; but the DL models outperform the TML models with respect to the most important performance factors as well as the average performance rankings.https://ieeexplore.ieee.org/document/9395633/Data augmentationdeep learningmachine learningneural networkssentiment analysisTurkish |
| spellingShingle | Harisu Abdullahi Shehu Md. Haidar Sharif Md. Haris Uddin Sharif Ripon Datta Sezai Tokat Sahin Uyaver Huseyin Kusetogullari Rabie A. Ramadan Deep Sentiment Analysis: A Case Study on Stemmed Turkish Twitter Data IEEE Access Data augmentation deep learning machine learning neural networks sentiment analysis Turkish |
| title | Deep Sentiment Analysis: A Case Study on Stemmed Turkish Twitter Data |
| title_full | Deep Sentiment Analysis: A Case Study on Stemmed Turkish Twitter Data |
| title_fullStr | Deep Sentiment Analysis: A Case Study on Stemmed Turkish Twitter Data |
| title_full_unstemmed | Deep Sentiment Analysis: A Case Study on Stemmed Turkish Twitter Data |
| title_short | Deep Sentiment Analysis: A Case Study on Stemmed Turkish Twitter Data |
| title_sort | deep sentiment analysis a case study on stemmed turkish twitter data |
| topic | Data augmentation deep learning machine learning neural networks sentiment analysis Turkish |
| url | https://ieeexplore.ieee.org/document/9395633/ |
| work_keys_str_mv | AT harisuabdullahishehu deepsentimentanalysisacasestudyonstemmedturkishtwitterdata AT mdhaidarsharif deepsentimentanalysisacasestudyonstemmedturkishtwitterdata AT mdharisuddinsharif deepsentimentanalysisacasestudyonstemmedturkishtwitterdata AT ripondatta deepsentimentanalysisacasestudyonstemmedturkishtwitterdata AT sezaitokat deepsentimentanalysisacasestudyonstemmedturkishtwitterdata AT sahinuyaver deepsentimentanalysisacasestudyonstemmedturkishtwitterdata AT huseyinkusetogullari deepsentimentanalysisacasestudyonstemmedturkishtwitterdata AT rabiearamadan deepsentimentanalysisacasestudyonstemmedturkishtwitterdata |