Automated Data Quality Management For Topology Layers In Geospatial Databases: AN AI-DRIVEN APPROACH

chapter 3

What’s Currently Misaligned

1. Implementation vs. Conceptual Focus Your chapter presents the methodology as if you’re building and testing a prototype system, including:
• Training AI models (Random Forest, CNN, etc.)
• Data preprocessing and simulation
• Performance evaluation using metrics like precision, recall, and F1-score

However, your research aim is to design a conceptual AI-driven framework, not to implement or deploy a working model. The methodology should focus on design, structure, component analysis, and expert validation, not technical system performance.

What Needs to Be Done

1. Shift the Methodology to Framework Development Your Chapter 3 should reflect a Design Science Research (DSR) or conceptual design methodology, with stages such as:
• Literature and domain problem analysis
• Identification of topological data quality challenges
• Mapping suitable AI techniques (conceptually, not tested)
• Designing a layered conceptual framework
• Validation via expert surveys or interviews
2. Remove or Reframe Technical Implementation Sections discussing synthetic datasets, model training, or metrics like precision/recall should be removed or reframed. If you wish to keep the mention of AI models, refer to them as proposed or suitable techniques for inclusion in the conceptual framework — not ones you’ve trained or tested.
3. Include Ethical Approval Preparation Since your validation involves expert surveys or interviews, you are required to obtain ethical approval before data collection. You should include a section on research ethics, outlining how:
• Participants will be informed and consented
• Data will be stored securely and anonymized
• Feedback will be used for framework evaluation only

Additional Notes

• Rephrase “prototype” to “conceptual framework” throughout the chapter.
• Include a clear section outlining how expert feedback will be collected and analyzed qualitatively (e.g., thematic analysis, framework refinement).
• Consider including a diagram or table mapping AI techniques to potential framework components (e.g., error detection, correction logic, spatial reasoning).

Suggested Chapter 3 Structure (Revised)

1. Introduction
2. Research Design (Design Science approach for framework development)
3. Data Sources and Literature Review Basis
4. Framework Development Process
5. Validation Approach via Expert Survey/Interview
6. Ethical Considerations
7. Conclusion

i will share the the 3 chapters