Abstract

Social media platforms like X and Facebook generate vast amounts of image content daily, necessitating automated methods for classification and analysis. Integrating Business Intelligence (BI) with Recurrent Neural Network (RNN) techniques presents a promising approach to extract valuable insights from this data. This study proposes a methodology for social media image content classification using a hybrid architecture combining Convolutional Neural Networks (CNNs) for feature extraction and RNNs for capturing temporal dependencies. The model is trained on labeled image datasets from X and Facebook, leveraging transfer learning and data augmentation techniques. The contribution lies in the fusion of BI and deep learning techniques, offering a scalable solution for real-time image content classification on social media platforms. This approach enables businesses to streamline marketing analysis, trend detection, and content moderation tasks efficiently. Experimental results demonstrate the effectiveness of the proposed methodology, achieving high accuracy in classifying diverse image content. The model''s performance is validated through comprehensive evaluation metrics, showcasing its robustness and applicability in real-world scenarios.