Sentiment analysis on Algerian and modern standard Arabic using deep learning and SVM.
In this work, we used the SVM and LSTM techniques to perform sentiment analysis on the Algerian and the modern standard Arabic, where we used two techniques of word embedding the Tf-IDF and the Word2Vec (CBOW model).
In this repository, you will find the Python source code used in our research and you will find also the Word2vec text corpus, our dataset and our saved models (the Word2Vec, SVM and LSTM model).
For the library and the environment that we used (Requirements) :
- Python 3.6
- TensorFlow
- Scikit learn
- Gensim
- Anaconda environment
If you used any of our resources, please refer our work using this bibtex :
@INPROCEEDINGS{9068897, author={A. {Abdelli} and F. {Guerrouf} and O. {Tibermacine} and B. {Abdelli}}, booktitle={2019 International Conference on Intelligent Systems and Advanced Computing Sciences (ISACS)}, title={Sentiment Analysis of Arabic Algerian Dialect Using a Supervised Method}, year={2019}, volume={}, number={}, pages={1-6},}
Or plain text:
A. Abdelli, F. Guerrouf, O. Tibermacine and B. Abdelli, "Sentiment Analysis of Arabic Algerian Dialect Using a Supervised Method," 2019 International Conference on Intelligent Systems and Advanced Computing Sciences (ISACS), Taza, Morocco, 2019, pp. 1-6.