<![CDATA[Customer information is handled in enterprise organisations through customer relationship management systems, transactional databases, support systems, and marketing systems, where the meaning is coded using incompatible local schemas. Without capturing the context of the source of those records, where and when they were created, and which regulatory contexts the records are relevant to, such unified views of customers are complete in content but lacking in interpretive fidelity. This paper aims to compile the literature in the fields of context awareness, linked data, entity resolution, ontology-based data access, master data management, knowledge graphs, data lake architecture and data protection regulation to propose an architecture for the Semantic Customer 360 that aims to preserve explicit contexts, and not merely data consolidation. The architecture places a governed data-lake ingestion zone, a hybrid schema-matching module, a probabilistic and graph-based entity-resolution pipeline, and a knowledge-graph-structured semantic layer under the hood of customer-facing analytics and service consumption. The synthesis shows that several candidate comparison techniques can lead to over 10 times fewer comparisons than naive pairwise comparison, that a choice of master data management strategy has measurable impact on governance overhead, that ontology-based mediation can also enable access for queries to sensitive data without moving or copying physical data, and that perceived regulatory compliance is measurable and correlated with customers' trust. The architecture proposed along with comparative analysis it provides can help enterprise data architects to gauge unstructured approaches to evaluating matching algorithms, integration strategies and governance mechanisms that can help them harmonize disparate customer data while keeping them easily traceable and audit-friendly in the present landscape of data protection requirements.]]>
