This paper explores the use of AI and social media analytics to detect sustainability trends in Saudi Arabia's evolving market, in line with Vision 2030. The study processes millions of social media posts, news articles, and blogs to understand sustainability trends across various sectors. The AI-driven methodology offers sector-specific and cross-sector insights, providing decision-makers with a snapshot of market shifts, and can be adapted to other regions.
Keywords
social media analytics · sustainability trends · AI · Saudi Arabia · Vision 2030
This paper explores how AI and social media analytics can identify and track trends in Saudi Arabia across sectors such as construction, food and beverage, tourism, technology, and entertainment. The study analyzed millions of social media posts each month, classifying discussions and calculating scores to track trends. The AI-driven methodology was able to predict the emergence and growth of trends by utilizing social media data.
The paper introduces SaudiCulture, a new benchmark for evaluating the cultural competence of LLMs within Saudi Arabia, covering five major geographical regions and diverse cultural domains. The benchmark includes questions of varying complexity and distinguishes between common and specialized regional knowledge. Evaluations of five LLMs (GPT-4, Llama 3.3, FANAR, Jais, and AceGPT) revealed performance declines on region-specific questions, highlighting the need for region-specific knowledge in LLM training.
A national survey in Saudi Arabia of 330 participants reveals that 93% are actively using Generative AI, primarily for text-based tasks, while awareness and understanding remain uneven. Participants recognize benefits like productivity but caution against risks such as privacy, misinformation, and ethical misuse. The study highlights the need for AI literacy, culturally aligned solutions, and stronger frameworks for responsible deployment in Saudi Arabia.
KAUST organized an Arabic Sentiment Analysis Challenge where participants developed ML models to classify tweets as positive, negative, or neutral. The competition used the ASAD dataset with 55K tweets for training, 20K for validation, and 20K for final evaluation. The full dataset of 100K labeled tweets has been released for public use.