Abstract
With the rapid expansion of social media users and the ever-increasing data exchange between them, the era of big data has arrived. Integration of big data generates enormous benefits, making it a hotspot for research. However, big data demonstrates the heterogeneity brought on by multiple data sources. Big data integration is constrained by multi-source heterogeneous data. Moreover, the rise in the volume of social media data is affecting the efficiency of data integration. This study is concerned with developing a novel framework for data integration system that can manage the heterogeneity of massive social media data. The framework is comprised of four layers: data source layer, application layer, resource layer, and visualization layer. The framework establishes correlations between data stored in distributed data sources. We used RESTful APIs to offer end-users with reliable and effective web-based access to data using unique queries. The framework was evaluated based on firsthand impressions of test users, who answered a standardized set of questions after testing real-world inputs.