Representing Relational Data as RDF Triples

Any table T1:

T1.A	T1.B	T1.C
value_A1	value_B1	value_C1
value_A2	value_B2	value_C2
value_A3	value_B3	value_C3

Can be viewed as a table with an extra key:

Key	T1.A	T1.B	T1.C
1	value_A1	value_B1	value_C1
2	value_A2	value_B2	value_C2
3	value_A3	value_B3	value_C3

(Note: This step is unnecessary if the table already has a primary key)

Which can be viewed as a set of tables (one per column):

Key	T1.A
1	value_A1
2	value_A2
3	value_A3

Key	T1.B
1	value_B1
2	value_B2
3	value_B3

Key	T1.C
1	value_C1
2	value_C2
3	value_C3

With column names moved to data:

Key	Column	Value
1	T1.A	value_A1
2	T1.A	value_A2
3	T1.A	value_A3

Key	Column	Value
1	T1.B	value_B1
2	T1.B	value_B2
3	T1.B	value_B3

Key	Column	Value
1	T1.C	value_C1
2	T1.C	value_C2
3	T1.C	value_C3

Which can be viewed as a single table of triples:

Key	Column	Value
1	T1.A	value_A1
2	T1.A	value_A2
3	T1.A	value_A3
1	T1.B	value_B1
2	T1.B	value_B2
3	T1.B	value_B3
1	T1.C	value_C1
2	T1.C	value_C2
3	T1.C	value_C3

Correspondence to RDF Triples:

Key = Subject
Column = Predicate/Property
Value = Value