GhentCDH/ArABSA-Digital-Text-Analysis

Here you can find the work of Lily Foula on Arabic ABSA. # Arabic Aspect-Based Sentiment Analysis 🎓 This repository hosts the work done in the framework of the internship on digital text analysis conducted by Lily Foula in summer 2025. THe goal of the internship was to develop a Jupyter notebook for **Aspect-Based sentiment analysis** in Arabic. This was completed in collaboration with Lisa INSERT INSERT, with the support of the Ghent Center for Digital Humanities and the Language and Translation Technology Team (LT3). We decided to tackle this task by focusing our efforts on the development of aspect-based sentiment analysis workflows in two steps: 1. Aspect extraction*. 2. Sentiment analysis on the aspect and sentence columns. *_We define an aspect as a unit in the sentence which can be both a named entity (a proper name) and a noun._ ❗🧠 These Notebooks are not meant to reinvent the wheel. We simply want to build an infrastructure for scholars looking to perform aspect-based sentiment analysis on their corpus, and we do this by making step-wise code examples which can be freely used and adapted for your own purposes! 🚀 Below we list the resources and Jupyter Notebooks we created for **aspect-based sentiment analysis** for Arabic. ## Annotated Training Dataset For training this notebook, we used the annotated dataset from "Semeval-2016 task 5: Aspect based sentiment analysis", which is an Arabic dataset for Hotel reviews. We did not personally create nor annotate this dataset. This is the sample dataset you will find uploaded here! References: Mohammad, A. S., Qwasmeh, O., Talafha, B., Al-Ayyoub, M., Jararweh, Y., & Benkhelifa, E. (2016, December). An enhanced framework for aspect-based sentiment analysis of Hotels' reviews: Arabic reviews case study. In 2016 11th International Conference for Internet Technology and Secured Transactions (ICITST) (pp. 98-103). IEEE. Al-Smadi, M., Talafha, B., Al-Ayyoub, M., & Jararweh, Y. (2019). Using long short-term memory deep neural networks for aspect-based sentiment analysis …