Natural Language Processing of Social Media Sentiment Toward Halal Logos in Indonesian SMEs

The increasing global demand for halal products, particularly in Muslim-majority countries like Indonesia, underscores the importance of halal certification as a competitive advantage for small and medium enterprises (SMEs). Despite this, the level of halal certification among Indonesian SMEs remains low, often due to limited awareness and misconceptions regarding its necessity. This study aims to analyze public sentiment toward halal logos by utilizing big data from social media, specifically Twitter, and assess its implications for halal literacy and certification readiness among SMEs in underdeveloped regions. A mixed-method approach was employed, combining Natural Language Processing (NLP) for sentiment analysis of 207 tweets using lexicon-based techniques, and a structured questionnaire administered to 167 SME owners across four districts in Madura to measure their halal literacy. The sentiment analysis revealed that 58.9% of tweets expressed positive sentiment, 22.7% were neutral, and 18.4% were negative toward halal logos, indicating a generally favorable public perception. Concurrently, the halal literacy score of SMEs averaged 91.98 out of 100, suggesting a strong foundational understanding among business owners. The findings imply that public discourse on halal certification is a motivating factor for SMEs to pursue formal halal labeling. This research contributes to the literature by integrating sentiment analysis and Islamic economic behavior, highlighting the role of machine learning in shaping market readiness. Future work should expand data sources beyond Twitter and explore longitudinal analysis to assess evolving sentiment trends. This study fills a critical gap in linking consumer sentiment with SME compliance in the halal industry using AI-driven approaches.