Visual Intelligence Layer

Interactive knowledge graph visualizations mapping how the SBPI Auto-Research Engine connects ontology design, competitive intelligence, and compounding IP value.

150 Nodes 608 Edges 8 Clusters 3 Structural Gaps Modularity: 0.516

System Knowledge Graph

Entity relationships extracted from the full Auto-Research Engine report. Node size encodes betweenness centrality (gateway influence). Dashed red lines show structural gaps between disconnected clusters.

Prediction Accuracy
Ontology Dimensions
Streaming Confidence
Knowledge Graphs
Revenue Infrastructure
Graph Relationships
Query Architecture
Investor Signals

Full Entity Graph — 150 Nodes, 608 Edges

mod 0.516 | 8 clusters | focused diversity
Hover nodes for details. Gateway nodes (high betweenness) bridge multiple conceptual domains. Dashed lines indicate structural gaps between disconnected clusters.

Key structural gap: Streaming Confidence (company rankings, momentum signals) is disconnected from both Revenue Infrastructure (business model, cost structure) and Query Architecture (SPARQL pipeline, SHACL validation). The prediction engine produces signals, but those signals don't yet flow into the revenue model or the query infrastructure automatically.

System Architecture Flow

The two data pipelines that power the auto-research engine — ecosystem graph enrichment and semantic layer intelligence — visualized as a directed flow graph.

Pipeline Architecture — Dual Flow System

10 components | 2 flows | OTK lineage
Top flow: Ecosystem Graph (session → entity extraction → InfraNodus). Bottom flow: Semantic Layer (JSON → RDF → SHACL → Oxigraph → SPARQL → Insights). Both converge at the knowledge graph.

Component Dependencies

Production Stack Dependency Graph

11 tools | data flow direction
Arrows show data flow direction. Central position indicates higher connectivity. The ontology (OWL 2) is the structural hub — everything references it.

SBPI Dimension Analysis

Five scoring dimensions with expert-calibrated weights. The radar chart shows how each company's profile diverges from the balanced ideal.

Dimension Weight Distribution

5 dimensions | weighted 15-25%
Distribution Power carries the highest weight (25%) because mobile-first app store presence determines audience reach. Monetization Infrastructure carries the lowest (15%) because platform giants subsidize the vertical.

Tier Distribution

Company Distribution Across Performance Tiers

22 companies | 5 tiers | W11-2026
The micro-drama vertical is bottom-heavy: no company reaches Dominant tier. ReelShort (84.0) is closest. Most companies cluster in Emerging and Niche tiers, indicating a young market with room for structural disruption.

Momentum Signal Map

W12-2026 predictions plotted by momentum magnitude (x-axis) and confidence level (y-axis). Bubble size encodes absolute momentum strength.

Prediction Signal Scatter — W12-2026

9 signals | 60-85% confidence | 4 detection types
Upper-right quadrant: strong bullish signals with high confidence (JioHotstar). Lower-left: bearish signals (Amazon, Netflix). The asymmetry — more bullish than bearish signals — reflects the vertical's growth phase.

Confidence Formula Breakdown

Confidence Score Construction

base 0.60 | 3 adjustments | cap 0.95
Each bar shows how the confidence score builds from base (0.60) through adjustments. JioHotstar hits 0.85 through strong movement (+0.10) and very strong movement (+0.10) plus consistent magnitude (+0.05).

Scaling Trajectory

Logarithmic growth path from current 1,672 triples toward the billion-node thesis. Each milestone unlocks new capability tiers.

Triple Count Growth Trajectory

6 milestones | log scale | 2026-2028
Current position at 1,672 triples (marked). Each order of magnitude unlocks qualitatively different capabilities. The curve reflects the dual-source growth: weekly SBPI data + client engagement entity extraction.

Self-Improving Cycle

Autopoietic Improvement Loop

5 stages | continuous | compounding
Each engagement enriches the knowledge graph, which improves the ontology, which makes the next engagement faster and more precise. The loop is self-sustaining once the ontology reaches sufficient density.

IP Engine Variables

Four variables that determine knowledge graph value. Visualized as a quadrant map showing current position and improvement trajectory.

IP Engine Variable Space

N_d | α | E_c | λ
Node Density (N_d) and Ontological Alpha (α) are the moat variables — they compound with each engagement. Extraction Efficiency (E_c) improves with automation. Decay Rate (λ) is domain-specific and managed through weekly refresh cycles.

Revenue Flywheel

Dual Revenue Model — Service + IP Accumulation

4 phases | service → platform
Every engagement produces both immediate revenue (deliverable) and long-term IP (knowledge graph triples). The cost per triple drops from $33 to $0.02 as automation scales — a 1,650x efficiency gain.