This paper presents a groundbreaking approach to traffic prediction through the integration of crowdsourced data within a big data analytics framework. The proposed methodology addresses the limitations of traditional traffic prediction models and machine learning approaches by leveraging the collective intelligence from diverse sources, including navigation apps, social media, and connected vehicles. A dynamic algorithm continuously updates the model parameters in real-time, adapting to evolving traffic patterns in urban environments. Comprehensive experiments, conducted on the UrbanTrafficFlow (UTF) and CityMobilityPatterns (CMP) datasets, demonstrate the superior predictive accuracy and adaptability of the proposed model compared to existing methods, including LSTM-based models, spatiotemporal graph convolutional networks (ST-GCN), and traditional statistical models. The integration of diverse data sources showcases the model's ability to enhance accuracy, and the impact of crowdsourced data integration is highlighted. This research contributes to the advancement of intelligent transportation systems and big data analytics, paving the way for more resilient and efficient urban mobility solutions in the era of smart cities.