TY  - JOUR
AU  - Tasani, Mimi Haryani 
AU  - Yusof, Mohd Kamir 
PY  - 2026
TI  - A Systematic Literature Review on Data Integration: Issues, Data Types, Architecture
JF  - Journal of Computer Science
VL  - 22
IS  - 2
DO  - 10.3844/jcssp.2026.487.503
UR  - https://thescipub.com/abstract/jcssp.2026.487.503
AB  - Data integration is an important process that enables organizations to merge data from various sources, creating a consolidated viewpoint that supports better decision-making, increased effectiveness, and deeper insights. Although data integration has advantages, it also comes with difficulties like maintaining data quality, adhering to compliance regulations, resolving technical issues, and merging with older systems. Effective management of data quality, robust data governance practices, and the use of advanced integration tools like data filtering are necessary to successfully overcome these challenges and ensure the harmonization and standardization of data. Systematic literature review explains by taking a strategic approach, companies can transform data integration into a valuable resource, establishing a base for improved analytics, customer understanding, and more efficient operations that bolster long-term growth and competitiveness. This research explores into the various complex obstacles of merging data in the realm of big data, focusing on structured, semi-structured, and unstructured data categories. With the progress of the healthcare sector, IoT, and big data analytics, there is a growing need for strong integration methods to manage the increasing volume and variety of data for better accessibility and usability. The research uncovers primary concerns in data integration including data quality and consistency, diverse data formats and standards, scalability and performance, and governance and compliance. It stresses the significance of recognizing the distinct traits and obstacles of various data forms to enable efficient integration. Moreover, the article examines different data integration models pointing out their advantages and disadvantages in meeting current data management requirements. The research seeks to improve data processing in today's data-driven world by sharing a detailed analysis of current obstacles and solutions in data integration.