Skip to content

Graph-Native Infrastructure for Context and Accountable AI Systems

fix: Resolve centrality result misread in DecisionQuery

KaifAhmad17a24273f41b2
diff --git a/semantica/context/decision_query.py b/semantica/context/decision_query.py
index b9bd3272..51a523ac 100644
--- a/semantica/context/decision_query.py
+++ b/semantica/context/decision_query.py
@@ -783,7 +783,7 @@ class DecisionQuery:
# Calculate degree centrality
centrality = self.kg_components["centrality_calculator"].calculate_degree_centrality(subgraph)
- boost = centrality.get(decision_id, 0.0)
+ boost = centrality.get('centrality', {}).get(decision_id, 0.0)
# Cache the result
self._cache[cache_key] = boost
@@ -882,16 +882,25 @@ class DecisionQuery:
# Calculate centrality measures
if "centrality_calculator" in self.kg_components:
subgraph = self._get_decision_subgraph(decision_id, max_depth)
- centrality_measures = self.kg_components["centrality_calculator"].calculate_all_centrality(subgraph)
- analysis["centrality_measures"] = centrality_measures.get(decision_id, {})
+ centrality_results = self.kg_components["centrality_calculator"].calculate_all_centrality(subgraph)
+
+ # Extract centrality measures for this decision from nested structure
+ decision_measures = {}
+ centrality_measures = centrality_results.get('centrality_measures', {})
+
+ for measure_type, measure_data in centrality_measures.items():
+ if isinstance(measure_data, dict) and 'centrality' in measure_data:
+ decision_measures[measure_type] = measure_data['centrality'].get(decision_id, 0.0)
+
+ analysis["centrality_measures"] = decision_measures
# Calculate overall influence score
measures = analysis["centrality_measures"]
analysis["influence_score"] = (
- 0.3 * measures.get("degree_centrality", 0.0) +
- 0.3 * measures.get("betweenness_centrality", 0.0) +
- 0.2 * measures.get("closeness_centrality", 0.0) +
- 0.2 * measures.get("eigenvector_centrality", 0.0)
+ 0.3 * measures.get("degree", 0.0) +
+ 0.3 * measures.get("betweenness", 0.0) +
+ 0.2 * measures.get("closeness", 0.0) +
+ 0.2 * measures.get("eigenvector", 0.0)
)
# Community detection