1/42
Vocabulary flashcards covering the architecture, regulatory context (PDPL 151/2020), integration interfaces, and algorithmic workflows for Egypt's sovereign data-protection (Product A) and AI risk-tier classification (Product B) platform.
PDPL 151/2020
Egypt's data protection law, referenced alongside the 2025 regulations and the PDPC, governing consent, RoPA, data-subject requests, breach notices, transfers, and DPO accountability.
Product A
An API-first data-protection workflow service covering local identifier discovery, controller/processor RoPAs, consent, requests, breaches, and cross-border transfer workflows.
Product B
An explainable AI risk tier classification service that evaluates five fuzzy factors to produce risk tiers, strongest rules, and bilingual Arabic/English explanations.
Contract C-01
The versioned integration contract used to send Product B AI risk tier results to the Assurance platform.
Contract C-X1
The platform integration contract through which Product A delivers evidence events and hashes to the Assurance service and ledger.
Contract C-19
The platform interface used to transmit PDPL rule packs.
Contract C-X2
A two-way integration interface that delivers review tasks to the DPO and receives signed decisions.
Contract C-X3
A two-way platform integration interface for the tenant and DPO directory.
Contract C-10
An incoming integration interface that provides Arabic entity services to the platform.
Contract C-14
An incoming integration interface delivering the offline package used by Tier 1 local scanners.
RoPA Manager
A literature benchmark combining hybrid RAG and a local LLM to extract internal document fields in a simulated Vietnamese-law setting.
Mumtathil
A literature benchmark for Arabic Saudi privacy-policy classification that operates exclusively on published policy text.
Tier 1 (Architecture)
The local connector layer where client databases and ERPs are scanned locally using validators and an offline Arabic package, exporting only metadata, counts, and salted hashes while keeping raw personal-data values local.
Tier 2 (Architecture)
The tenant-isolated core environment running Product A RoPA, rule, and deadline services, as well as Product B fuzzy classification and bilingual explanation engines.
Tier 3 (Architecture)
The architecture layer that handles event delivery to Assurance, the evidence ledger, and the DPO Console.
PDPC Workbook
The official Egyptian Excel export format containing 25 controller columns, 12 processor columns, and 7 Identification entries.
Prohibited Gate
A hard validation rule in Product B where any declared prohibited AI practice immediately triggers a prohibited tier classification that numeric scores cannot override.
Fuzzy Stage 1 (Product B)
An inference stage that processes scale, reversibility, and data sensitivity across 27 Mamdani rules to calculate an impact score.
Fuzzy Stage 2 (Product B)
An inference stage that evaluates impact, sector sensitivity, and decision autonomy using 27 rules to determine the AI risk tier.
Centroid Defuzzification
The defuzzification method used in Product B to map fuzzy inference outputs into minimal, limited, or high risk tiers.
Boundary Flag
A mechanism in Product B that flags an assessment for human review whenever the computed score falls near the tier thresholds.
PDPL 151/2020
Egypt's Personal Data Protection Law governing consent, RoPA, data-subject requests, breach notices, cross-border transfers, and DPO accountability, supplemented by 2025 executive regulations.
PDPC
The Egyptian Personal Data Protection Center, which defines compliance frameworks and standardized Excel workbook formats for RoPA reporting.
Product A
An API-first data-protection workflow service handling local identifier discovery, controller/processor RoPAs, consent, data-subject requests (DSRs), breach notices, and transfer workflows.
Product B
An API-first explainable AI classification service that evaluates systems across five fuzzy factors to output a risk tier, strongest rules, and bilingual Arabic/English explanations.
C-01
The platform integration interface used to deliver Product B AI risk classification results to the Assurance service.
C-X1
The platform integration interface used to queue and transmit Product A compliance evidence events and hashes to the evidence ledger.
C-19
The integration interface through which standardized PDPL compliance rule packs are distributed across the platform.
C-X2
A two-way platform interface for managing Data Protection Officer (DPO) review tasks and returning signed decisions.
C-X3
A two-way platform interface used to synchronize the tenant and Data Protection Officer (DPO) directory.
C-10
The inbound platform interface providing Arabic entity identification and recognition services to local validators.
C-14
The inbound platform interface that delivers offline packages to Tier 1 local client database and ERP scanners.
RoPA Manager
A literature baseline system combining hybrid RAG and a local LLM to extract internal document fields under a simulated Vietnamese-law setting.
Mumtathil
A literature baseline system designed for Arabic privacy-policy classification operating exclusively on published Saudi policy text.
Tier 1 (Architecture)
The on-premise connector layer where client databases and ERPs are scanned locally, ensuring raw personal data stays local while exporting only metadata, counts, and salted hashes.
Tier 2 (Architecture)
The tenant-isolated platform core hosting Product A RoPA, rule, and deadline services, alongside Product B fuzzy classification and bilingual explanation engines.
Tier 3 (Architecture)
The platform orchestration layer managing event delivery to Assurance, the immutable evidence ledger, and the DPO Console.
PDPC Workbook Format
The official Egyptian reporting template comprising 25 controller columns, 12 processor columns, and 7 Identification sheet entries.
Prohibited Gate
A mandatory hard-rule check in Product B where any declared prohibited AI practice immediately classifies the system as prohibited, overriding any numeric fuzzy evaluation.
Fuzzy Stage 1 (Product B)
The first inference stage in Product B that maps scale, reversibility, and data sensitivity into an impact score using 27 Mamdani rules.
Fuzzy Stage 2 (Product B)
The second inference stage in Product B that combines impact, sector sensitivity, and decision autonomy into an overall risk score using 27 Mamdani rules.
Centroid Defuzzification
The mathematical defuzzification method used in Product B to map continuous fuzzy risk distributions into discrete minimal, limited, or high-risk tiers.
Boundary Flag
An automated alert triggered when an AI system's calculated score lies near risk tier boundaries, routing the assessment to human experts for calibration.