Machine Learning Fundamental 5

📘 Pertemuan 5 – Supervised Learning: Classification

🎯 Tujuan Pembelajaran

Setelah mengikuti pertemuan ini mahasiswa mampu:

1. Memahami konsep classification dalam machine learning
2. Memahami algoritma Logistic Regression
3. Memahami konsep K-Nearest Neighbors (KNN)
4. Melakukan training model klasifikasi
5. Mengevaluasi model menggunakan confusion matrix dan classification metrics

---

1️⃣ Apa itu Classification?

Classification adalah metode supervised learning yang digunakan untuk memprediksi kelas atau kategori.

Contoh kasus:

Kasus	Target
Spam detection	Spam / Not Spam
Diagnosis penyakit	Positif / Negatif
Kelulusan mahasiswa	Lulus / Tidak
Fraud detection	Fraud / Normal

---

2️⃣ Dataset Contoh

Kita akan menggunakan dataset Titanic untuk memprediksi apakah penumpang selamat atau tidak.

import pandas as pd
import seaborn as sns

df = sns.load_dataset("titanic")
df.head()

Target

survived
0 = tidak selamat
1 = selamat

3️⃣ Data Preparation

Memilih fitur yang akan digunakan

df = df[['survived','pclass','sex','age','fare']]
df = df.dropna()

Encoding data kategorikal:

from sklearn.preprocessing import LabelEncoder

le = LabelEncoder()
df['sex'] = le.fit_transform(df['sex'])

4️⃣ Train Test Split

from sklearn.model_selection import train_test_split

X = df.drop("survived", axis=1)
y = df["survived"]

X_train, X_test, y_train, y_test = train_test_split(
    X,
    y,
    test_size=0.2,
    random_state=42
)

5️⃣ Logistic Regression

Logistic Regression digunakan untuk binary classification.
Persamaan sigmoid:
P(y=1) = \frac{1}{1 + e^{-z}}

Implementasi:

from sklearn.linear_model import LogisticRegression

model = LogisticRegression()
model.fit(X_train, y_train)

y_pred = model.predict(X_test)

6️⃣ K-Nearest Neighbors (KNN)

KNN bekerja dengan mencari tetangga terdekat.
Langkah algoritma:

1. Pilih nilai K
2. Hitung jarak ke data lain
3. Ambil K tetangga terdekat
4. Gunakan voting mayoritas

Implementasi:

from sklearn.neighbors import KNeighborsClassifier

knn = KNeighborsClassifier(n_neighbors=5)

knn.fit(X_train, y_train)

y_pred_knn = knn.predict(X_test)

7️⃣ Evaluasi Model

Accuracy

from sklearn.metrics import accuracy_score

accuracy = accuracy_score(y_test, y_pred)
print("Accuracy:", accuracy)

Confusion Matrix

Confusion matrix menunjukkan performa klasifikasi.

	Predicted No	Predicted Yes
Actual No	TN	FP
Actual Yes	FN	TP

from sklearn.metrics import confusion_matrix

cm = confusion_matrix(y_test, y_pred)
print(cm)

Classification Report

from sklearn.metrics import classification_report

print(classification_report(y_test, y_pred))

Metric yang ditampilkan:

• Precision
• Recall
• F1 Score
• Accuracy

8️⃣ Visualisasi Confusion Matrix

import seaborn as sns
import matplotlib.pyplot as plt

sns.heatmap(cm, annot=True, fmt="d", cmap="Blues")
plt.xlabel("Predicted")
plt.ylabel("Actual")
plt.title("Confusion Matrix")
plt.show()

🧪 Praktikum Lengkap

import pandas as pd
import seaborn as sns

from sklearn.model_selection import train_test_split
from sklearn.preprocessing import LabelEncoder
from sklearn.linear_model import LogisticRegression
from sklearn.metrics import accuracy_score

df = sns.load_dataset("titanic")

df = df[['survived','pclass','sex','age','fare']]
df = df.dropna()

le = LabelEncoder()
df['sex'] = le.fit_transform(df['sex'])

X = df.drop("survived", axis=1)
y = df["survived"]

X_train, X_test, y_train, y_test = train_test_split(
    X,
    y,
    test_size=0.2,
    random_state=42
)

model = LogisticRegression()

model.fit(X_train, y_train)

y_pred = model.predict(X_test)

print("Accuracy:", accuracy_score(y_test, y_pred))

📝 Tugas Praktikum

1. Gunakan dataset Titanic.
2. Lakukan:
   • preprocessing
   • logistic regression
   • KNN
3. Bandingkan akurasi kedua model.

📝 Tugas Mandiri

1. Ambil dataset klasifikasi dari Kaggle.
2. Lakukan:
   • EDA
   • preprocessing
   • classification model
   • evaluasi model
3. Gunakan template laporan berikut:
   https://github.com/AzharRizkiZ/Template-DS-ML⁠
4. Upload proyek ke GitHub repository masing-masing.

🎓 Target Kompetensi

Mahasiswa mampu:

1. Memahami konsep classification
2. Mengimplementasikan logistic regression
3. Menggunakan KNN
4. Mengevaluasi model klasifikasi