Arabic_Dialect_Sentiment_Analysis

A ML/DL based system working on SVM/BiLSTM, able to analyze Arabic Tweets and Classifies it to different types of Arabic accents.

Additional describtion about the project below.

Tools, Libs used

• HTML

•CSS

•GIT

•Word2Vec

•Colab

•Python

•Gensim

•Flask

•ngrok

•keras

•TensorFlow

Data Preprocessing

• Arabert pre-Trained Model https://github.com/aub-mind/arabert

git clone https://github.com/aub-mind/arabert.git
!pip install farasapy
!pip install pyarabic

from arabert.preprocess import ArabertPreprocessor

• Word2vec word embeddings from gensim library

!pip3 install --upgrade gensim

from gensim.models import word2vec

Modeling

• Machine Learning Modeling

from sklearn.svm import SVC

• Deep Learning Modeling

from tensorflow.keras import Sequential
from tensorflow.keras.layers import Bidirectional

Deployment on Flask, ngrok service

install important dependecies for Flask

!pip install flask-ngrok
!pip install pyngrok

Linux command on colab to install Flask dependencies

!curl -s https://ngrok-agent.s3.amazonaws.com/ngrok.asc | sudo tee /etc/apt/trusted.gpg.d/ngrok.asc >/dev/null && echo "deb https://ngrok-agent.s3.amazonaws.com buster main" | sudo tee /etc/apt/sources.list.d/ngrok.list && sudo apt update && sudo apt install ngrok

install flask libraries

from flask import Flask, request, render_template
from flask_ngrok import run_with_ngrok

Author

👤 Remon Atef Boshra

• Github @Remon128

• LinkedIn https://www.linkedin.com/in/remon-boshra/

• Twitter @remonaboshra

Show your Support

Give a ⭐️ if you like the project !

Contributions, issues, and feature requests are welcome!