A new approach to analyzing microblogging of tweets in social networks based on fuzzy semantic relationships

In the era of pervasive social networks like Twitter and Facebook, the abundance of user-generated content poses a significant challenge for information analysis. Most previous work has focused on the relevance of the information to a given query. Our research introduces an innovative approach, employing textual analysis to explore social information within interconnected semantic relationships on the Twitter microblogging network. This type of analysis is known for its incompleteness, imprecision, and heterogeneity. To remedy these shortcomings, we integrate the notions of possibilistic logic, which proves to be the most adequate since it provides a solid theoretical basis for modeling incomplete and imprecise information. We proposed a new approach based on possibility and necessity semantic relationships between tweets. To assess the robustness of our approach across various contexts, we conducted experiments on four databases of different sizes. The experimental study demonstrated the efficiency of our approach compared to the main existing methods. The results confirmed the effectiveness of our method, grounded in the theory of possibilities. It achieves precision 94%  recall 73%  accuracy 85%  and G-Mean 84%Ṫhis method can serve as a versatile example of textual recommendation across various fields