Employing Network Analysis of Big Data on Platform (X) in Relation to Saudi Vision 2030 (An Analytical Longitudinal Study for the Period from 2016–2022)

This study aimed to explore how artificial intelligence techniques can be utilized in the analysis of big data in relation to Saudi Vision 2030. Using methodologies such as network analysis and sentiment analysis, the study sought to provide in-depth insights into how Saudi Vision 2030 is perceived by the Saudi public, and to identify the aspects that garner the highest levels of attention, interaction, and engagement. The study focused on the Arabic hashtag (رؤية_السعودية_2030#) and the English hashtag (#SaudiVision2030) on the X platform, covering a seven-year period from 2016 to 2022. A mixed-method approach was adopted, combining traditional statistical analysis with network analysis techniques and natural language processing — using artificial intelligence tools — to gain a deeper understanding of public sentiment and the dominant topics in discussions surrounding Saudi Vision 2030. Tweets were categorized by sentiment into positive, negative, and neutral in order to assess the overall public mood toward the Vision and its diverse and comprehensive developmental goals, which fundamentally emphasize sustainability. Textual network analysis helped explore relationships among user opinions and responses to the Vision. The findings revealed that positive sentiment towards Vision 2030 programs was predominant, accounting for 50.3% of tweets, compared to 37% neutral and 12.6% negative sentiment. This indicates widespread support within Saudi society for the Vision and the economic, social, and cultural transformations it has introduced, all of which contribute to sustainable development. The study recommended the use of Large Language Models (LLMs) for big data analysis on social media platforms instead of traditional classification models, due to their superior capabilities in interpreting text and language. It also advised the application of Knowledge Graphs to represent and extract information from textual data, as this approach provides a clearer and more accurate representation of network relationships.