To enable the dynamic creation of 'supplier risk profiles', we envisage a database scheme that facilitates querying of supplier-related award information from contract records.
The proposed database captures essential attributes of contracts, allowing for extraction and flattening of data to create multiple contract award records for analysis.
Data cleansing and linking, such as extracting geographical details from addresses, enhances the quality of information drawn from tender and award XML sources.
Our project faces challenges including data heterogeneity and the dilemma of algorithmic choices amidst the complexity of multilingual and multi-task data environments.
Collection
[
|
...
]