Name : KOWSALYA M

Register No: 212222230069

Experiment 1

DATE:01/09/2024

Implementation of Bayesian Networks

Aim :

To create a bayesian Network for the given dataset in Python

Algorithm:

Step 1:Import necessary libraries: pandas, networkx, matplotlib.pyplot, Bbn, Edge, EdgeType, BbnNode, Variable, EvidenceBuilder, InferenceController
Step 2:Set pandas options to display more columns
Step 3:Read in weather data from a CSV file using pandas
Step 4:Remove records where the target variable RainTomorrow has missing values
Step 5:Fill in missing values in other columns with the column mean
Step 6:Create bands for variables that will be used in the model (Humidity9amCat, Humidity3pmCat, and WindGustSpeedCat)
Step 7:Define a function to calculate probability distributions, which go into the Bayesian Belief Network (BBN)
Step 8:Create BbnNode objects for Humidity9amCat, Humidity3pmCat, WindGustSpeedCat, and RainTomorrow, using the probs() function to calculate their probabilities
Step 9:Create a Bbn object and add the BbnNode objects to it, along with edges between the nodes
Step 10:Convert the BBN to a join tree using the InferenceController
Step 11:Set node positions for the graph
Step 12:Set options for the graph appearance
Step 13:Generate the graph using networkx
Step 14:Update margins and display the graph using matplotlib.pyplot

Program:

import networkx as nx
import pandas as pd
import matplotlib.pyplot as plt
from pybbn.graph.dag import Bbn
from pybbn.graph.dag import Edge,EdgeType
from pybbn.graph.jointree import EvidenceBuilder
from pybbn.graph.node import BbnNode
from pybbn.graph.variable import Variable
from pybbn.pptc.inferencecontroller import InferenceController
pd.options.display.max_columns=50

df=pd.read_csv('weatherAUS.csv',encoding='utf-8')
df=df[pd.isnull(df['RainTomorrow'])==False]
df = df.drop(columns='Date')

numeric_columns = df.select_dtypes(include=['number']).columns
df.loc[:, numeric_columns] = df[numeric_columns].fillna(df[numeric_columns].mean())

df['WindGustSpeedCat']=df['WindGustSpeed'].apply(lambda x: '0.<=40'   if x<=40 else '1.40-50' if 40<x<=50 else '2.>50')
df['Humidity9amCat']=df['Humidity9am'].apply(lambda x: '1.>60' if x>60 else '0.<=60')
df['Humidity3pmCat']=df['Humidity3pm'].apply(lambda x: '1.>60' if x>60 else '0.<=60')

print(df)

def probs(data, child, parent1=None, parent2=None):
    if parent1==None:
        # Calculate probabilities
        prob=pd.crosstab(data[child], 'Empty', margins=False, normalize='columns').sort_index().to_numpy().reshape(-1).tolist()
    elif parent1!=None:
            # Check if child node has 1 parent or 2 parents
            if parent2==None:
                # Caclucate probabilities
                prob=pd.crosstab(data[parent1],data[child], margins=False, normalize='index').sort_index().to_numpy().reshape(-1).tolist()
            else:
                # Caclucate probabilities
                prob=pd.crosstab([data[parent1],data[parent2]],data[child], margins=False, normalize='index').sort_index().to_numpy().reshape(-1).tolist()
    else: print("Error in Probability Frequency Calculations")
    return prob

H9am = BbnNode(Variable(0, 'H9am', ['<=60', '>60']), probs(df, child='Humidity9amCat'))
H3pm = BbnNode(Variable(1, 'H3pm', ['<=60', '>60']), probs(df, child='Humidity3pmCat', parent1='Humidity9amCat'))
W = BbnNode(Variable(2, 'W', ['<=40', '40-50', '>50']), probs(df, child='WindGustSpeedCat'))
RT = BbnNode(Variable(3, 'RT', ['No', 'Yes']), probs(df, child='RainTomorrow', parent1='Humidity3pmCat', parent2='WindGustSpeedCat'))

bbn = Bbn() \
    .add_node(H9am) \
    .add_node(H3pm) \
    .add_node(W) \
    .add_node(RT) \
    .add_edge(Edge(H9am, H3pm, EdgeType.DIRECTED)) \
    .add_edge(Edge(H3pm, RT, EdgeType.DIRECTED)) \
    .add_edge(Edge(W, RT, EdgeType.DIRECTED))

join_tree = InferenceController.apply(bbn)

pos = {0: (-1, 2), 1: (-1, 0.5), 2: (1, 0.5), 3: (0, -1)}

options = {
    "font_size": 16,
    "node_size": 4000,
    "node_color": "pink",
    "edgecolors": "blue",
    "edge_color": "green",
    "linewidths": 5,
    "width": 5,}

n, d = bbn.to_nx_graph()
nx.draw(n, with_labels=True, labels=d, pos=pos, **options)

ax = plt.gca()
ax.margins(0.10)
plt.axis("off")
plt.show()
print(probs(df, child='Humidity9amCat'))
print(probs(df, child='Humidity3pmCat', parent1='Humidity9amCat'))
print(probs(df, child='WindGustSpeedCat'))
print(probs(df, child='RainTomorrow', parent1='Humidity3pmCat', parent2='WindGustSpeedCat'))

Output:

Variable bands:

Graph:

Result:

Thus a Bayesian Network is generated using Python