Comprehensive Survey on State-of-the-Art Methodologies, Resources, Applications, and Challenges of Automated Sentiment Analysis for Low-Resource Languages

This survey presents a comprehensive review of automated sentiment analysis methodologies tailored to low-resource languages—those lacking sufficient digital and linguistic resources for Natural Language Processing (NLP). Although significant progress has been made in sentiment analysis for high-resource languages, low-resource languages remain underrepresented due to limited annotated corpora, sentiment lexicons, and computational tools. The paper categorizes existing techniques into rule-based, lexicon-based, machine learning, deep learning, hybrid, and cross-lingual approaches, evaluating their applicability and limitations in low-resource settings. It also explores available linguistic resources, annotation strategies, tools, and real-world applications, particularly in multilingual and multicultural contexts. Key challenges, such as data scarcity, linguistic complexity, and cultural nuances, are discussed alongside promising future directions, including transfer learning, few-shot learning, and community-led data initiatives. This work aims to guide researchers and policymakers in developing inclusive and equitable NLP systems that serve linguistically diverse communities.