Validation of FAIRiCUBE use cases
Validation of use cases (UCs) in FAIRiCUBE focuses on assessing whether each case is clearly defined, aligned with project goals, and practically applicable. It involves checking the completeness and clarity of UC specifications, evaluating whether the right data and methods are used, and conducting user assessments to ensure the outputs meet real-world needs. Fit-for-purpose checks are applied to confirm that the results are suitable for the intended context. This process also includes documenting findings and communicating them effectively to stakeholders. Importantly, UC validation is closely connected to the validation of processing workflows and platform functionality, ensuring consistency across technical and thematic components.
Use Case 1
| UC implementation step | Check type | Reference |
|---|---|---|
| UC specifications | Clear Goal defined | FAIRiCUBE Digital library – UC1 – Research questions |
| Required datasets identified | FAIRiCUBE Digital library – UC1 – Data and ingestion | |
| Required ML/AI approaches identified | FAIRiCUBE Digital library – UC1 – Processing steps and ML applications | |
| Workflow designed | FAIRiCUBE Website – UC1 workflow | |
| Visualisation of outputs designed | UC1 Scrollytelling | |
| User assessment / fitness-for-purpose | Support the users' work | Stakeholder engagement |
| Service orientation | ||
| Reliability | ||
| Applicability | ||
| Data systems stability, reliability, and interoperability | ||
| Data pre-processing and ingestion | List of characteristics | Github - Pre-processing: quality check |
| Descriptive Statistics Calculation | ||
| Spatial Validation | ||
| Anomaly detection | ||
| Error Labelling and Data Incorporation | Github - Processing: gap filling | |
| Reporting and Logging | Benchmark logs included in the STAC metadata of a/p resources | |
| Processing and Machine Learning | Algorithm implementation validation | Github - processing: MaxEnt example |
| Machine learning validation | Github - processing: MaxEnt model evaluation | |
| Benchmarking | Github - processing: MaxEnt benchmarks | |
| Comprehensive documentation | FAIRiCUBE Digital library – UC1 | |
| Data sharing | Information | FAIRiCUBE Data Catalog - UC1 - Suitability maps |
| Data | Datasets archived in Zenodo | |
| Processing | FAIRiCUBE Data Catalog - UC1 - Model SHAP | |
| Portrayal | Zenodo – Poster for the GDDS event | |
| AI ethics assessment | Ethics (Trustworthy AI) | FAIRiCUBE Hub – Validation of AI Ethics |
| GDPR applicability | Not Applicable |
Use Case 2
| UC implementation step | Check type | Reference |
|---|---|---|
| UC specifications | Clear Goal defined | FAIRiCUBE Digital library – UC2 – Overview |
| Required datasets identified | FAIRiCUBE Digital library – UC2 – Data | |
| Required ML/AI approaches identified | FAIRiCUBE Digital library – UC2 – Approach | |
| Workflow designed | FAIRiCUBE Website – UC2 workflow | |
| Visualisation of outputs designed | UC2 Scrollytelling | |
| User assessment / fitness-for-purpose | Support the users' work | missing reference |
| Service orientation | ||
| Reliability | ||
| Applicability | ||
| Data systems stability, reliability, and interoperability | ||
| Data pre-processing and ingestion | List of characteristics | Github - Data preparation scripts |
| Descriptive Statistics Calculation | ||
| Spatial Validation | ||
| Anomaly detection | ||
| Error Labelling and Data Incorporation | ||
| Reporting and Logging | ||
| Processing and Machine Learning | Algorithm implementation validation | Validation scripts |
| Machine learning validation | ML validation results | |
| Benchmarking | Github - Measurer | |
| Comprehensive documentation | Documentation | |
| Data sharing | Information | STAC metadata |
| Data | Data outputs | |
| Processing | Processing resources metadata | |
| Portrayal | Poster / Other materials | |
| AI ethics assessment | Ethics (Trustworthy AI) | FAIRiCUBE Hub – Validation of AI Ethics |
| GDPR applicability | Not Applicable |
Use Case 3
| UC implementation step | Check type | Reference |
|---|---|---|
| UC specifications | Clear Goal defined | FAIRiCUBE Digital library – UC3 – Research questions |
| Required datasets identified | FAIRiCUBE Digital library – UC3 – Data and ingestion | |
| Required ML/AI approaches identified | FAIRiCUBE Digital library – UC3 – Processing steps and ML applications | |
| Workflow designed | FAIRiCUBE Website – UC3 workflow | |
| Visualisation of outputs designed | UC3 Scrollytelling | |
| User assessment / fitness-for-purpose | Support the users' work | Github - UrbanDrosophila Ecology Project |
| Service orientation | ||
| Reliability | ||
| Applicability | ||
| Data systems stability, reliability, and interoperability | ||
| Data pre-processing and ingestion | List of characteristics | Github links |
| Descriptive Statistics Calculation | Github links | |
| Spatial Validation | Github links | |
| Anomaly detection | Github links | |
| Error Labelling and Data Incorporation | Github links | |
| Reporting and Logging | Github links | |
| Processing and Machine Learning | Algorithm implementation validation | Github links |
| Machine learning validation | Github links | |
| Benchmarking | Github links | |
| Comprehensive documentation | Github links | |
| Data sharing | Information | FAIRiCUBE Data Catalog |
| Data | Github – Landscape genomics | |
| Processing | FAIRiCUBE Data Catalog | |
| Portrayal | Zenodo – Poster for the GDDS event | |
| AI ethics assessment | Ethics (Trustworthy AI) | FAIRiCUBE Hub – Validation of AI Ethics |
| GDPR applicability | Not Applicable |
Use Case 4
| UC implementation step | Check type | Reference |
|---|---|---|
| UC specifications | Clear Goal defined | FAIRiCUBE Digital library – UC4 – Research questions |
| Required datasets identified | FAIRiCUBE Digital library – UC4 – Data and ingestion | |
| Required ML/AI approaches identified | FAIRiCUBE Digital library – UC4 – Processing steps and ML applications | |
| Workflow designed | FAIRiCUBE Digital library – UC4 – Workflow | |
| Visualisation of outputs designed | UC4 Scrollytelling | |
| User assessment / fitness-for-purpose | Support the users' work | missing reference |
| Service orientation | ||
| Reliability | ||
| Applicability | ||
| Data systems stability, reliability, and interoperability | ||
| Data pre-processing and ingestion | List of characteristics | Github - Energy estimation notebooks |
| Descriptive Statistics Calculation | ||
| Spatial Validation | ||
| Anomaly detection | ||
| Error Labelling and Data Incorporation | ||
| Reporting and Logging | ||
| Processing and Machine Learning | Algorithm implementation validation | Github - Notebooks |
| Machine learning validation | Github - Energy estimation notebooks | |
| Benchmarking | Github - Performance measures | |
| Comprehensive documentation | Github - Energy estimation notebooks (self-documented) | |
| Data sharing | Information | FAIRiCUBE catalog STAC API |
| Data | Zenodo | |
| Processing | FAIRiCUBE Data Catalog – EPISCOPE-based heuristic | |
| Portrayal | SBE25 Proceedings | |
| AI ethics assessment | Ethics (Trustworthy AI) | FAIRiCUBE Hub – Validation of AI Ethics |
| GDPR applicability | Not Applicable |
Use Case 5
| UC implementation step | Check type | Reference |
|---|---|---|
| UC specifications | Clear Goal defined | FAIRiCUBE Digital library – UC5 – Research questions |
| Required datasets identified | FAIRiCUBE Digital library – UC5 – Data retrieval | |
| Required ML/AI approaches identified | FAIRiCUBE Digital library – UC5 – Processing Steps and ML Applications | |
| Workflow designed | FAIRiCUBE Website – UC5 workflow | |
| Visualisation of outputs designed | UC5 Scrollytelling | |
| User assessment / fitness-for-purpose | Support the users' work | GithHub – User friendly script |
| Service orientation | ||
| Reliability | ||
| Applicability | ||
| Data systems stability, reliability, and interoperability | ||
| Data pre-processing and ingestion | List of characteristics | Github – Data pre-processing and ingestion |
| Descriptive Statistics Calculation | ||
| Spatial Validation | ||
| Anomaly detection | ||
| Error Labelling and Data Incorporation | ||
| Reporting and Logging | ||
| Processing and Machine Learning | Algorithm implementation validation | Github – Model output statistics |
| Github – Model workflow | ||
| Machine learning validation | Github – Model validation | |
| Benchmarking | Github – Model benchmarking | |
| Comprehensive documentation | FAIRiCUBE Digital Library – UC5 – Processing steps and ML Applications | |
| Data sharing | Information | FAIRiCUBE STAC Data Catalog |
| Data | Input data in Github Repository | |
| Processing | FAIRiCUBE STAC a/p resources Catalog | |
| Portrayal | Zenodo – Poster for the GDDS event | |
| AI ethics assessment | Ethics (Trustworthy AI) | FAIRiCUBE Hub – Validation of AI Ethics |
| GDPR applicability | Not applicable |
