Spaces:
Sleeping
Sleeping
File size: 12,715 Bytes
3934b26 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 | """Equilibrium detection and analysis for opinion dynamics"""
from typing import List, Dict
import statistics
from collections import Counter
from .models import (
RoundResult,
EquilibriumState,
PersonaOpinion,
OpinionPosition,
)
from .network import InfluenceNetwork
class EquilibriumDetector:
"""
Analyzes opinion dynamics results to detect equilibrium and extract insights.
Provides:
- Equilibrium detection
- Consensus metrics
- Opinion cluster analysis
- Opinion leader identification
- Evolution timeline analysis
"""
def __init__(self, convergence_threshold: float = 0.1):
self.convergence_threshold = convergence_threshold
def analyze_equilibrium(
self,
results: List[RoundResult],
influence_network: InfluenceNetwork,
max_rounds: int,
) -> EquilibriumState:
"""
Analyze final equilibrium state.
Args:
results: Complete history of opinion dynamics
influence_network: Influence network used
max_rounds: Maximum rounds configured
Returns:
EquilibriumState with comprehensive analysis
"""
if not results:
raise ValueError("No results to analyze")
final_round = results[-1]
final_opinions = final_round.opinions
# Check if equilibrium was reached
reached_equilibrium = (
len(results) < max_rounds
and final_round.total_change < self.convergence_threshold
)
# Calculate consensus metrics
consensus_strength = self._calculate_consensus_strength(final_opinions)
majority_position, majority_pct = self._find_majority_position(
final_opinions
)
# Position distribution
position_dist = self._calculate_position_distribution(final_opinions)
# Identify opinion clusters
opinion_clusters = self._identify_opinion_clusters(
final_opinions, results
)
# Identify opinion leaders
opinion_leaders = self._identify_opinion_leaders(
final_opinions, results, influence_network
)
return EquilibriumState(
reached_equilibrium=reached_equilibrium,
total_rounds=len(results),
final_opinions=final_opinions,
consensus_strength=consensus_strength,
majority_position=majority_position,
majority_percentage=majority_pct,
position_distribution=position_dist,
opinion_clusters=opinion_clusters,
opinion_leaders=opinion_leaders,
evolution_timeline=results,
)
def _calculate_consensus_strength(
self, opinions: List[PersonaOpinion]
) -> float:
"""
Calculate how much consensus exists (0-1).
1 = Everyone agrees perfectly
0 = Maximum disagreement
"""
if len(opinions) <= 1:
return 1.0
# Calculate variance in positions
positions = [op.position_score for op in opinions]
variance = statistics.variance(positions)
# Max variance is when half are at -3, half at +3
max_variance = 9.0 # (3 - (-3))^2 / 4
# Normalize: high variance = low consensus
consensus = 1.0 - (variance / max_variance)
return max(0.0, min(1.0, consensus))
def _find_majority_position(
self, opinions: List[PersonaOpinion]
) -> tuple:
"""Find the majority position and percentage"""
position_counts = Counter(op.position.value for op in opinions)
majority_pos, majority_count = position_counts.most_common(1)[0]
majority_pct = (majority_count / len(opinions)) * 100
majority_position = OpinionPosition(majority_pos)
return majority_position, majority_pct
def _calculate_position_distribution(
self, opinions: List[PersonaOpinion]
) -> Dict[str, int]:
"""Count personas at each position"""
distribution = {}
for opinion in opinions:
pos = opinion.position.value
distribution[pos] = distribution.get(pos, 0) + 1
return distribution
def _identify_opinion_clusters(
self, final_opinions: List[PersonaOpinion], results: List[RoundResult]
) -> List[Dict[str, any]]:
"""
Identify stable opinion clusters.
Clusters are groups of personas with:
- Similar final positions
- Similar evolution patterns
"""
# Group by final position (within 1 point)
clusters = []
used_personas = set()
sorted_opinions = sorted(
final_opinions, key=lambda x: x.position_score
)
for opinion in sorted_opinions:
if opinion.persona_id in used_personas:
continue
# Find similar personas
cluster_members = [opinion]
used_personas.add(opinion.persona_id)
for other in sorted_opinions:
if other.persona_id in used_personas:
continue
if abs(other.position_score - opinion.position_score) <= 1.0:
cluster_members.append(other)
used_personas.add(other.persona_id)
if cluster_members:
avg_position = statistics.mean(
[m.position_score for m in cluster_members]
)
clusters.append({
"size": len(cluster_members),
"members": [m.persona_name for m in cluster_members],
"member_ids": [m.persona_id for m in cluster_members],
"average_position": avg_position,
"position_label": OpinionPosition.from_score(avg_position).value,
"stability": self._calculate_cluster_stability(
cluster_members, results
),
})
return sorted(clusters, key=lambda x: x["size"], reverse=True)
def _calculate_cluster_stability(
self, members: List[PersonaOpinion], results: List[RoundResult]
) -> float:
"""
Calculate how stable this cluster is (0-1).
High stability = members stayed together throughout
"""
if len(results) <= 1:
return 1.0
member_ids = {m.persona_id for m in members}
# Track how many rounds members stayed close
stable_rounds = 0
for result in results[1:]: # Skip first round
# Get positions for members in this round
member_positions = [
op.position_score
for op in result.opinions
if op.persona_id in member_ids
]
if len(member_positions) <= 1:
stable_rounds += 1
continue
# Check variance within cluster
variance = statistics.variance(member_positions)
if variance < 1.0: # Stayed close
stable_rounds += 1
return stable_rounds / (len(results) - 1)
def _identify_opinion_leaders(
self,
final_opinions: List[PersonaOpinion],
results: List[RoundResult],
influence_network: InfluenceNetwork,
) -> List[Dict[str, any]]:
"""
Identify opinion leaders - personas who influenced the outcome most.
Criteria:
1. High influence centrality in network
2. How many people moved toward their position
3. Consistency (didn't change position much)
"""
leaders = []
network_metrics = influence_network.calculate_network_metrics()
centrality = network_metrics.centrality_scores
for opinion in final_opinions:
persona_id = opinion.persona_id
# 1. Network centrality
centrality_score = centrality.get(persona_id, 0.0)
# 2. Influence impact: how many moved toward them
impact_score = self._calculate_influence_impact(
persona_id, results, opinion.position_score
)
# 3. Consistency: didn't change much
consistency_score = self._calculate_consistency(persona_id, results)
# Combined leadership score
leadership_score = (
centrality_score * 0.4
+ impact_score * 0.4
+ consistency_score * 0.2
)
leaders.append({
"persona_id": persona_id,
"persona_name": opinion.persona_name,
"leadership_score": leadership_score,
"centrality": centrality_score,
"influence_impact": impact_score,
"consistency": consistency_score,
"final_position": opinion.position.value,
})
# Sort by leadership score
leaders = sorted(
leaders, key=lambda x: x["leadership_score"], reverse=True
)
return leaders[:5] # Top 5 leaders
def _calculate_influence_impact(
self, persona_id: str, results: List[RoundResult], final_position: float
) -> float:
"""
Calculate how much others moved toward this persona's position.
Higher score = more people converged to their view
"""
if len(results) <= 1:
return 0.0
# Get this persona's position in round 1
initial_position = None
for op in results[0].opinions:
if op.persona_id == persona_id:
initial_position = op.position_score
break
if initial_position is None:
return 0.0
# Count how many others moved toward their position
convergence_count = 0
total_others = 0
for other_id in [op.persona_id for op in results[0].opinions]:
if other_id == persona_id:
continue
total_others += 1
# Get initial and final position for this other persona
other_initial = None
other_final = None
for op in results[0].opinions:
if op.persona_id == other_id:
other_initial = op.position_score
for op in results[-1].opinions:
if op.persona_id == other_id:
other_final = op.position_score
if other_initial is None or other_final is None:
continue
# Did they move toward this persona?
initial_distance = abs(other_initial - initial_position)
final_distance = abs(other_final - final_position)
if final_distance < initial_distance:
convergence_count += 1
return convergence_count / total_others if total_others > 0 else 0.0
def _calculate_consistency(
self, persona_id: str, results: List[RoundResult]
) -> float:
"""
Calculate how consistent this persona was (0-1).
Higher score = didn't change position much
"""
positions = []
for result in results:
for op in result.opinions:
if op.persona_id == persona_id:
positions.append(op.position_score)
if len(positions) <= 1:
return 1.0
# Calculate variance
variance = statistics.variance(positions)
# Normalize (max variance is 9 if going from -3 to +3)
consistency = 1.0 - (variance / 9.0)
return max(0.0, min(1.0, consistency))
def get_convergence_summary(
self, equilibrium: EquilibriumState
) -> str:
"""Generate human-readable summary of convergence"""
summary = []
if equilibrium.reached_equilibrium:
summary.append(
f"✓ Equilibrium reached after {equilibrium.total_rounds} rounds"
)
else:
summary.append(
f"⚠ No equilibrium reached in {equilibrium.total_rounds} rounds"
)
summary.append(
f"📊 Consensus strength: {equilibrium.consensus_strength:.0%}"
)
summary.append(
f"🎯 Majority position: {equilibrium.majority_position.value} "
f"({equilibrium.majority_percentage:.0f}%)"
)
summary.append(
f"👥 Opinion clusters: {len(equilibrium.opinion_clusters)}"
)
if equilibrium.opinion_leaders:
leader = equilibrium.opinion_leaders[0]
summary.append(
f"⭐ Top opinion leader: {leader['persona_name']}"
)
return "\n".join(summary)
|