TY  - JOUR
AU  - Khan, Muhammad Owais
AU  - Khan, Wahab
AU  - Wang, Yanan
AU  - Rehman, Aziz Ur
AU  - Khan, Muhammad Alamzeb
PY  - 2025
DA  - 2025/07/14
TI  - Enhancing Sentiment Analysis of Roman Urdu Using Augmentation Techniques and Deep Learning Models
JO  - ICCK Transactions on Advanced Computing and Systems
T2  - ICCK Transactions on Advanced Computing and Systems
JF  - ICCK Transactions on Advanced Computing and Systems
VL  - 1
IS  - 3
SP  - 164
EP  - 179
DO  - 10.62762/TACS.2025.190575
UR  - https://www.icck.org/article/abs/TACS.2025.190575
KW  - Roman Urdu
KW  - sentiment analysis
KW  - deep learning
KW  - data augmentation
KW  - text classification
KW  - GRU
KW  - LSTM
KW  - RNN
AB  - Roman Urdu sentiment analysis faces significant challenges due to transliteration inconsistencies, informal language usage, and the lack of labeled datasets. This study proposes a novel framework that addresses these challenges by combining advanced data preprocessing techniques and data augmentation strategies such as synonym replacement, back-translation, and random word insertion. These methods enhance dataset diversity, improving the model’s generalization ability. A rich Roman Urdu dataset was collected from diverse sources, including social media platforms (Facebook, Twitter, YouTube), blogs, forums, and e-commerce sites, to capture a wide range of user opinions. Three deep learning models, Recurrent Neural Network (RNN), Gated Recurrent Unit (GRU), and Long Short-Term Memory (LSTM), were evaluated for sentiment classification. The results show that the LSTM model outperforms the others with an accuracy of 94%, compared to 90% for RNN and 92% for GRU. The LSTM’s ability to capture long-term dependencies and contextual nuances in Roman Urdu text makes it the most effective model for this task, demonstrating a significant improvement over the traditional method.
SN  - 3068-7969
PB  - Institute of Central Computation and Knowledge
LA  - English
ER  - 

@article{Khan2025Enhancing,
  author = {Muhammad Owais Khan and Wahab Khan and Yanan Wang and Aziz Ur Rehman and Muhammad Alamzeb Khan},
  title = {Enhancing Sentiment Analysis of Roman Urdu Using Augmentation Techniques and Deep Learning Models},
  journal = {ICCK Transactions on Advanced Computing and Systems},
  year = {2025},
  volume = {1},
  number = {3},
  pages = {164-179},
  doi = {10.62762/TACS.2025.190575},
  url = {https://www.icck.org/article/abs/TACS.2025.190575},
  abstract = {Roman Urdu sentiment analysis faces significant challenges due to transliteration inconsistencies, informal language usage, and the lack of labeled datasets. This study proposes a novel framework that addresses these challenges by combining advanced data preprocessing techniques and data augmentation strategies such as synonym replacement, back-translation, and random word insertion. These methods enhance dataset diversity, improving the model’s generalization ability. A rich Roman Urdu dataset was collected from diverse sources, including social media platforms (Facebook, Twitter, YouTube), blogs, forums, and e-commerce sites, to capture a wide range of user opinions. Three deep learning models, Recurrent Neural Network (RNN), Gated Recurrent Unit (GRU), and Long Short-Term Memory (LSTM), were evaluated for sentiment classification. The results show that the LSTM model outperforms the others with an accuracy of 94\%, compared to 90\% for RNN and 92\% for GRU. The LSTM’s ability to capture long-term dependencies and contextual nuances in Roman Urdu text makes it the most effective model for this task, demonstrating a significant improvement over the traditional method.},
  keywords = {Roman Urdu, sentiment analysis, deep learning, data augmentation, text classification, GRU, LSTM, RNN},
  issn = {3068-7969},
  publisher = {Institute of Central Computation and Knowledge}
}