Heterogeneity is one of the essential features of big data. Integrating heterogeneous data sources with problems such as data-value conflicts, data representation conflicts, data-unit conflicts, and data precision conflicts, and data movement from heterogeneous sources into a central database are challenging. The proposed framework uses extract, transform and load technologies to integrate and publish data from multiple sources. The system supports a wide range of database sources and targets to potentially transform the data to a unified knowledge base. The System adopts an Object-Oriented Design Method (OODM) and provides flexible and independent design and management of ETL processes in order to solve the data integration problems. The System has the ability to transform the contents of the databases, its representation as to their data fields, resolve data type incompatibilities, semantic incompatibilities to a unified view for a user