AI Cybersecurity Framework Selector
Budget: $250 – $750 USD
I’m building a small-scale, university-level prototype that recommends the most suitable cybersecurity framework for a payment-gateway provider after it supplies basic company data through a simple GUI. The tool must consider only cybersecurity frameworks—no governance or risk models—with the initial comparison set restricted to CIS Controls v8, PCI DSS, NIST 800-53, NIST 800-61, and MITRE ATT&CK.
Scope of the build
• Data layer: curate or simulate representative “company data”—size, transaction volume, geography, existing controls—and prepare it as the training / inference set.
• Model: tune an appropriate NLP or rules-augmented model so that, given the company profile, it returns the single best-fit framework plus a short justification. Python with scikit-learn, spaCy, or a lightweight transformer is fine; I’m open to your suggestions.
• Interface: a minimal GUI (Streamlit or similar) where users paste or upload company parameters and receive the framework recommendation instantly.
• Validation: design measurable accuracy tests—cross-validation, confusion matrix, precision / recall—so I can cite quantitative performance in my paper.
• Documentation: comment the code thoroughly and supply a brief methodology report I can reference in my dissertation.
Acceptance criteria
1. GUI loads locally, accepts company data, and outputs exactly one of the five frameworks with rationale.
2. Model reaches an agreed-upon accuracy baseline (you’ll help define this during data design).
3. All source code, datasets, and a short report are delivered within 2-2.5 weeks.
Timeline
I need the full package—code, GUI, validation results, and write-up—within a month, so please outline a clear milestone plan when you respond
The Ai's response has to be divided into 2 a business view and a technical view as its targeted for board level on why that framework is best suited for the company and the frameworks have to be in regards to payment gateways.
Scope of the build
• Data layer: curate or simulate representative “company data”—size, transaction volume, geography, existing controls—and prepare it as the training / inference set.
• Model: tune an appropriate NLP or rules-augmented model so that, given the company profile, it returns the single best-fit framework plus a short justification. Python with scikit-learn, spaCy, or a lightweight transformer is fine; I’m open to your suggestions.
• Interface: a minimal GUI (Streamlit or similar) where users paste or upload company parameters and receive the framework recommendation instantly.
• Validation: design measurable accuracy tests—cross-validation, confusion matrix, precision / recall—so I can cite quantitative performance in my paper.
• Documentation: comment the code thoroughly and supply a brief methodology report I can reference in my dissertation.
Acceptance criteria
1. GUI loads locally, accepts company data, and outputs exactly one of the five frameworks with rationale.
2. Model reaches an agreed-upon accuracy baseline (you’ll help define this during data design).
3. All source code, datasets, and a short report are delivered within 2-2.5 weeks.
Timeline
I need the full package—code, GUI, validation results, and write-up—within a month, so please outline a clear milestone plan when you respond
The Ai's response has to be divided into 2 a business view and a technical view as its targeted for board level on why that framework is best suited for the company and the frameworks have to be in regards to payment gateways.