FADHLOUN-Y/5TH-PLACE-Partial-Solution-for-the-Zindi-AI4D-iCompass-Social-Media-Sentiment-Analysis-for-Tunisian-

This challenge aims to classify sentiment in the Tunisian Arabizi dialect # 5TH PLACE Partial Solution for the Zindi AI4D iCompass Social Media Sentiment Analysis for Tunisian Arabizi Redirect to Challenge Website ## Objective Of the Challenge The objective of this challenge is to, given a sentence, classify whether the sentence is of positive, negative, or neutral sentiment. For messages conveying both a positive and negative sentiment, whichever is the stronger sentiment should be chosen. Predict if the text would be considered positive, negative, or neutral (for an average user). ## Quick Introduction The dataset has three sets of labels - Negative,Positive,Neutral and was highly Imbalanced. **54 % Positive Samples** **42 % Negative Samples** **04 % Neutral Samples** ## Our Approach We encountered two main problems with the dataset which are annotation and imbalacing of text classes .. We started to believe even if we had the chance to add external data and train our language modeling model we won't achieve great results so as a result of that we proceeded with two different approaches text-augmentation (which is not included in this solution ) and turning the problem into a binary task keeping only positive and negative texts . Our models were a combination between transformers and reccurent neural networks ( **RobertaXLM - LSTM** ) and (**Bert Multilingual Cased - LSTM**) inspired from Icompass Paper Learning Word Representations for Tunisian Sentiment Analysis . ## Instructions to run the code Code could be run using google colab . #### Environment Setup You'll find a requirement file that you could install in your own virtual environment. #### Data Setup Download data from the competition website and save it to the ./data/ directory. #### Training Phase To start training the models run ./train.sh it will take about 4 hours to run the two architectures . #### Testing Phase To start testing the models run ./test.sh it will create test files with ID of text sample and predictions if you are using the twolabels classifier …