TTLab at AraSentEval: SARF (صرف) Sentiment Analysis via Root-based Fusion for Multi-Dialectal Arabic

TTLab at AraSentEval: SARF (صرف) Sentiment Analysis via Root-based Fusion for Multi-Dialectal Arabic Authors: Ali Abusaleh, Bhuvanesh Verma, Alexander Mehler Conference: The 7th Workshop on Open-Source Arabic Corpora and Processing Tools (OSACT7) with 5 Shared Tasks, ELRA, 2026 Abstract Arabic sentiment analysis is challenged by morphological complexity and lexical variation across Arabic dialects, compounded by subjectivity in how speakers and writers express sentiment. We present SARF (صرف), a multi-view framework that integrates surface-level context with stemmed and rooted morphological representations using a shared MARBERTv2 encoder. The proposed hybrid BERT-CNN-BiLSTM-Attention architecture captures both local sentiment patterns and long-range contextual dependencies. Our experiments demonstrate that while stemming or rooting alone may reduce performance, their joint integration through cross-morphological attention produces more robust representations across diverse Arabic dialects. The proposed system achieved a Macro-F1 score of 0.9263, ranking 2nd out of 15 participating teams in the AraSentEval 2026 Shared Task. If you find this repository useful, please consider giving it a ⭐ and citing our work. Setup create python enviroment (Python 3.12 is recommended) python3 -m venv .venv activate the new virtual env source .venv/bin/activate/ install required libraries pip install -r requirements.txt how to use? have your input data in input/train_files/sent_train_v2.csv run python notebook in bert_cnn_lstm_with_cross_attention_final.ipynb Citation If you use this code, dataset, or any part of this repository in your research, please cite our paper: @inproceedings{Abusaleh:et:al:2026:sarf, title = {TTLab at AraSentEval: SARF( صرف) Sentiment Analysis via Root-based Fusion for Multi-Dialectal Arabic}, author = {Abusaleh, Ali and Verma, Bhuvanesh and Mehler, Alexander}, booktitle = {The 7th Workshop on Open-Source Arabic Corpora and Processing Tools (OSACT7) with 5 Shared Tasks}, month = {May}, year = {2026}, pages = {262--268}, address = {Palma, Mallorca, Spain}, publisher = {European Language Resources Association (ELRA)}, editor = {Al-Khalifa, Hend and El-Haj, Mo and Ezzini, Saad}, doi = {10.63317/4wj6s3ys5osk}, abstract = {Arabic sentiment analysis is challenged by morphological complexity and lexical variation across Arabic dialects, compounded by subjectivity in how speakers and writers express sentiment. In this paper, we present our submission for the AraSentEval 2026 Shared Task on Arabic Dialect Sentiment Analysis. We propose SARF (صرف) a multi-view architectural framework that integrates surface-level context with stemmed and rooted morphological perspectives using a shared MARBERTv2 encoder. Our system employs a hybrid BERT-CNN-BiLSTM-Attention architecture to capture both local sentiment n-grams and global sequential dependencies. Experimental results show that while individual morphological normalization strategies (stemming or rooting) may degrade performance, their joint integration via cross-morphological attention provides robust features across diverse dialects. Our final system achieved a competitive macro-F1-score of 0.9263, ranking 2nd out of 15 participating teams.} }