Code#1

import numpy as np
import networkx as nx
import matplotlib.pyplot as plt

# ------------------------------
# 1. Define a Dummy Medical Dataset
# ------------------------------

entities = [
    'Fever', 'Cough', 'Flu', 'Headache', 'Sore Throat',
    'Body Ache', 'Cold', 'COVID', 'Pneumonia', 'Bronchitis'
]

embeddings = {
    'Fever':       np.array([0.9, 0.3]),
    'Cough':       np.array([0.8, 0.4]),
    'Flu':         np.array([0.85, 0.35]),
    'Headache':    np.array([0.3, 0.9]),
    'Sore Throat': np.array([0.6, 0.7]),
    'Body Ache':   np.array([0.75, 0.45]),
    'Cold':        np.array([0.8, 0.5]),
    'COVID':       np.array([0.82, 0.48]),
    'Pneumonia':   np.array([0.4, 0.6]),
    'Bronchitis':  np.array([0.42, 0.58])
}

# ------------------------------
# 2. Define Helper Functions for Similarity and Fuzzy Membership
# ------------------------------

def cosine_similarity(vec1, vec2):
    return np.dot(vec1, vec2) / (np.linalg.norm(vec1) * np.linalg.norm(vec2))

def sigmoid(x):
    return 1 / (1 + np.exp(-x))

def fuzzy_membership(similarity):
    return sigmoid(similarity)

# ------------------------------
# 3. Construct the Fuzzy Graph
# ------------------------------

G = nx.Graph()

for entity in entities:
    G.add_node(entity)

threshold = 0.55

for i in range(len(entities)):
    for j in range(i+1, len(entities)):
        sim = cosine_similarity(embeddings[entities[i]], embeddings[entities[j]])
        mu_val = fuzzy_membership(sim)
        if mu_val > threshold:
            G.add_edge(entities[i], entities[j], weight=mu_val)

# ------------------------------
# 4. Augmenting With a Bayesian Network Slice
# ------------------------------

alpha = 0.9
bayesian_network = {}
for u, v, data in G.edges(data=True):
    mu_val = data['weight']
    bayesian_network[(u, v)] = alpha * mu_val
    bayesian_network[(v, u)] = alpha * mu_val

# ------------------------------
# 5. Visualization of the Fuzzy Graph
# ------------------------------

pos = nx.spring_layout(G, seed=42)

node_color = 'skyblue'
node_size = 800

edges = G.edges(data=True)
edge_weights = [data['weight'] for (u, v, data) in edges]

cmap = plt.cm.plasma
norm = plt.Normalize(min(edge_weights), max(edge_weights))
edge_colors = [cmap(norm(weight)) for weight in edge_weights]

plt.figure(figsize=(10, 7))
nx.draw_networkx_nodes(G, pos, node_color=node_color, node_size=node_size)
nx.draw_networkx_labels(G, pos, font_size=10, font_weight='bold')
nx.draw_networkx_edges(G, pos, edge_color=edge_colors, width=2)

sm = plt.cm.ScalarMappable(cmap=cmap, norm=norm)
sm.set_array([])

plt.colorbar(sm, label='Fuzzy Membership Weight', ax=plt.gca())

plt.title('Fuzzy Graph of Dummy Medical Entities')
plt.axis('off')
plt.show()

# ------------------------------
# 6. Display the Bayesian Network Probabilities
# ------------------------------

print("Bayesian Network Conditional Probabilities (P(child=True|parent=True)):")
for edge, prob in bayesian_network.items():
    print(f"P({edge[1]}=True | {edge[0]}=True) = {prob:.3f}")