In [ ]: # This Python 3 environment comes with many helpful analytics libraries inst

# It is defined by the kaggle/python Docker image: https://github.com/kaggle

# For example, here's several helpful packages to load

import numpy as np # linear algebra

import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)

# Input data files are available in the read-only "../input/" directory

# For example, running this (by clicking run or pressing Shift+Enter) will l

import os
for dirname, _, filenames in os.walk('/kaggle/input'):
for filename in filenames:
print(os.path.join(dirname, filename))

# You can write up to 20GB to the current directory (/kaggle/working/) that

# You can also write temporary files to /kaggle/temp/, but they won't be sav

Student Performance Analysis Using

Machine Learning
This notebook implements various machine learning techniques on the Student
Performance dataset from Kaggle.

Included Techniques:

Simple & Multiple Linear Regression

Polynomial, Lasso & Ridge Regression
Naïve Bayes, Logistic Regression
Decision Tree, SVM, K-NN Classifier
Artificial Neural Network
K-Means and Hierarchical Clustering

Evaluation Metrics:

R² Score (Regression)
Confusion Matrix, F1 Score (Classification)
Silhouette Score (Clustering)

In [1]: import pandas as pd

import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns

from sklearn.model_selection import train_test_split

from sklearn.preprocessing import StandardScaler, PolynomialFeatures
from sklearn.metrics import mean_squared_error, r2_score, confusion_matrix,
In [4]: df = pd.read_csv('/kaggle/input/students-performance-in-exams/StudentsPerfor
df = pd.get_dummies(df, drop_first=True)

X = df.drop(['math score'], axis=1)

y = df['math score']

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, ran

scaler = StandardScaler()
X_train = scaler.fit_transform(X_train)
X_test = scaler.transform(X_test)

📈 Regression Models
1️⃣ Simple Linear Regression
In [5]: from sklearn.linear_model import LinearRegression

model = LinearRegression()
model.fit(X_train[:, [0]], y_train)
y_pred = model.predict(X_test[:, [0]])

print("R² Score:", r2_score(y_test, y_pred))

R² Score: 0.6804469009921283

2️⃣ Multiple Linear Regression

In [6]: model = LinearRegression()
model.fit(X_train, y_train)
y_pred = model.predict(X_test)

print("R² Score:", r2_score(y_test, y_pred))

R² Score: 0.8804332983749564

3️⃣ Polynomial Regression

In [7]: poly = PolynomialFeatures(degree=2)
X_poly = poly.fit_transform(X_train)

model = LinearRegression()
model.fit(X_poly, y_train)
y_pred = model.predict(poly.transform(X_test))

print("R² Score:", r2_score(y_test, y_pred))

R² Score: 0.8650480765142721
 4️⃣ Ridge and Lasso Regression
In [8]: from sklearn.linear_model import Ridge, Lasso

ridge = Ridge(alpha=1.0).fit(X_train, y_train)

lasso = Lasso(alpha=0.1).fit(X_train, y_train)

print("Ridge R²:", r2_score(y_test, ridge.predict(X_test)))

print("Lasso R²:", r2_score(y_test, lasso.predict(X_test)))

Ridge R²: 0.8805453685953484

🤖 Classification Models
Lasso R²: 0.8822147639745545

🔁 Convert Target to Binary (Pass/Fail)

In [9]: df['target'] = ['pass' if s >= 50 else 'fail' for s in df['math score']]

X = df.drop(['math score', 'target'], axis=1)

y = df['target']

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, ran

X_train = scaler.fit_transform(X_train)
X_test = scaler.transform(X_test)

5️⃣ Logistic Regression

In [10]: from sklearn.linear_model import LogisticRegression

log_model = LogisticRegression()
log_model.fit(X_train, y_train)
log_preds = log_model.predict(X_test)

print("Confusion Matrix:\n", confusion_matrix(y_test, log_preds))

print("Classification Report:\n", classification_report(y_test, log_preds))
 Confusion Matrix:
[[ 20 14]
[ 7 159]]
Classification Report:
precision recall f1-score support

fail 0.74 0.59 0.66 34

pass 0.92 0.96 0.94 166

accuracy 0.90 200

macro avg 0.83 0.77 0.80 200
weighted avg 0.89 0.90 0.89 200

6️⃣ Naïve Bayes Classifier

In [11]: from sklearn.naive_bayes import GaussianNB

nb_model = GaussianNB()
nb_model.fit(X_train, y_train)
nb_preds = nb_model.predict(X_test)

print("Confusion Matrix:\n", confusion_matrix(y_test, nb_preds))

print("Classification Report:\n", classification_report(y_test, nb_preds))

Confusion Matrix:
[[ 23 11]
[ 18 148]]
Classification Report:
precision recall f1-score support

fail 0.56 0.68 0.61 34

pass 0.93 0.89 0.91 166

accuracy 0.85 200

macro avg 0.75 0.78 0.76 200
weighted avg 0.87 0.85 0.86 200

7️⃣ K-Nearest Neighbors (K-NN)

In [12]: from sklearn.neighbors import KNeighborsClassifier

knn_model = KNeighborsClassifier(n_neighbors=5)
knn_model.fit(X_train, y_train)
knn_preds = knn_model.predict(X_test)

print("Confusion Matrix:\n", confusion_matrix(y_test, knn_preds))

print("Classification Report:\n", classification_report(y_test, knn_preds))
 Confusion Matrix:
[[ 9 25]
[ 4 162]]
Classification Report:
precision recall f1-score support

fail 0.69 0.26 0.38 34

pass 0.87 0.98 0.92 166

accuracy 0.85 200

macro avg 0.78 0.62 0.65 200
weighted avg 0.84 0.85 0.83 200

8️⃣ Decision Tree Classification

In [13]: from sklearn.tree import DecisionTreeClassifier

tree_model = DecisionTreeClassifier(random_state=0)
tree_model.fit(X_train, y_train)
tree_preds = tree_model.predict(X_test)

print("Confusion Matrix:\n", confusion_matrix(y_test, tree_preds))

print("Classification Report:\n", classification_report(y_test, tree_preds))

Confusion Matrix:
[[ 20 14]
[ 11 155]]
Classification Report:
precision recall f1-score support

fail 0.65 0.59 0.62 34

pass 0.92 0.93 0.93 166

accuracy 0.88 200

macro avg 0.78 0.76 0.77 200
weighted avg 0.87 0.88 0.87 200

9️⃣ Support Vector Machine (SVM)

Classification
In [14]: from sklearn.svm import SVC

svm_model = SVC()
svm_model.fit(X_train, y_train)
svm_preds = svm_model.predict(X_test)

print("Confusion Matrix:\n", confusion_matrix(y_test, svm_preds))

print("Classification Report:\n", classification_report(y_test, svm_preds))
 Confusion Matrix:
[[ 16 18]
[ 2 164]]
Classification Report:
precision recall f1-score support

fail 0.89 0.47 0.62 34

pass 0.90 0.99 0.94 166

accuracy 0.90 200

macro avg 0.89 0.73 0.78 200
weighted avg 0.90 0.90 0.89 200

🔟 Artificial Neural Network (ANN)

In [15]: from sklearn.neural_network import MLPClassifier

ann_model = MLPClassifier(hidden_layer_sizes=(10,), max_iter=300, random_sta

ann_model.fit(X_train, y_train)
ann_preds = ann_model.predict(X_test)

print("Confusion Matrix:\n", confusion_matrix(y_test, ann_preds))

print("Classification Report:\n", classification_report(y_test, ann_preds))

Confusion Matrix:
[[ 19 15]
[ 8 158]]
Classification Report:
precision recall f1-score support

fail 0.70 0.56 0.62 34

pass 0.91 0.95 0.93 166

accuracy 0.89 200

macro avg 0.81 0.76 0.78 200
weighted avg 0.88 0.89 0.88 200

/usr/local/lib/python3.11/dist-packages/sklearn/neural_network/_multilayer_p
erceptron.py:686: ConvergenceWarning: Stochastic Optimizer: Maximum iteratio
ns (300) reached and the optimization hasn't converged yet.

📊 Clustering Models
warnings.warn(

1️⃣1️⃣ K-Means Clustering

In [16]: from sklearn.cluster import KMeans

X = df.drop(['math score', 'target'], axis=1)

X_scaled = scaler.fit_transform(X)
 kmeans = KMeans(n_clusters=2, random_state=0).fit(X_scaled)
print("KMeans Silhouette Score:", silhouette_score(X_scaled, kmeans.labels_)

KMeans Silhouette Score: 0.10611407723279058

/usr/local/lib/python3.11/dist-packages/sklearn/cluster/_kmeans.py:870: Futu
reWarning: The default value of `n_init` will change from 10 to 'auto' in 1.
4. Set the value of `n_init` explicitly to suppress the warning
warnings.warn(

1️⃣2️⃣ Hierarchical Clustering

In [17]: from sklearn.cluster import AgglomerativeClustering

hclust = AgglomerativeClustering(n_clusters=2).fit(X_scaled)
print("Hierarchical Clustering Silhouette Score:", silhouette_score(X_scaled

Hierarchical Clustering Silhouette Score: 0.15767520836587193

📌 Conclusion
This notebook demonstrates the application of various machine learning
algorithms to predict student performance. We implemented multiple regression
models, classification algorithms, and clustering techniques to analyze the
dataset and evaluate the effectiveness of each model.

Thank you for reviewing this analysis!

Submitted by:
Raunak Kumar Singh

University Name:
Atmaram Sanatan Dharma College

Course Name:
B.sc Computer Science Hons

Date of Submission:
15-04-2025

This notebook was converted with convert.ploomber.io