Slide 1

Slide 1 text

Machine Learning Muhammad Fikry Rizal Curriculum Developer Machine Learning Unsupervised Learning: Hidden Clustering and Pattern Discovery

Slide 2

Slide 2 text

Hi, I’m M. Fikry Rizal 👋 Latest Work Experiences: ● AI/ML Curriculum Developer, Dicoding 2024 - present ● Data Engineer Intern, Torche Education 2022 - 2023 Education: ● UIN Syarif Hidayatullah Jakarta 2020 - 2024 Bachelor Degree, Physics ● Bangkit Academy 2023 2023 Machine Learning About Me Muhammad Fikry Rizal https://github.com /mfikryrz Machine Learning

Slide 3

Slide 3 text

Machine Learning Machine Learning Overview 1. Hi, Machine Learning! 2. Machine Learning Workflow 3. Supervised Learning: Klasifikasi 4. Supervised Learning: Regresi 5. Unsupervised Learning - Clustering 6. Feature Engineering 7. Hyperparameter Tuning 8. Overfitting & Underfitting

Slide 4

Slide 4 text

Unsupervised Learning Concept Machine Learning

Slide 5

Slide 5 text

Unsupervised Learning Concept Machine Learning

Slide 6

Slide 6 text

Machine Learning Types of Clustering Methods Hierarchical Clustering Non-hierarchical Clustering NHC

Slide 7

Slide 7 text

Machine Learning KMeans Clustering: The Core of Unsupervised Learning

Slide 8

Slide 8 text

Machine Learning KMeans Clustering: The Core of Unsupervised Learning

Slide 9

Slide 9 text

Hands-on Coding: Let's Build It Machine Learning

Slide 10

Slide 10 text

Quiz #1 DevCoach 186 Machine Learning Format jawaban: #quiz1-username-jawaban

Slide 11

Slide 11 text

Quiz #1 DevCoach 186 Ketika bekerja dengan algoritma clustering, seperti KMeans, apa saja pendekatan yang dapat digunakan untuk menentukan jumlah cluster secara optimal? Pilih semua opsi yang benar.) Machine Learning a). Elbow Method b). Menggunakan Gap Statistic

Slide 12

Slide 12 text

Machine Learning Feature Engineering & Hyperparameter Tuning

Slide 13

Slide 13 text

Machine Learning Feature Engineering: Enhancing Data for Better Models

Slide 14

Slide 14 text

Machine Learning Feature Engineering: Feature Selection

Slide 15

Slide 15 text

Machine Learning Feature Engineering: Feature Selection

Slide 16

Slide 16 text

Machine Learning Feature Engineering: Feature Extraction

Slide 17

Slide 17 text

Machine Learning Feature Engineering: Feature Extraction

Slide 18

Slide 18 text

Machine Learning Feature Engineering: Feature Extraction

Slide 19

Slide 19 text

Machine Learning Hyperparameter Tuning: Grid Search vs Random Search

Slide 20

Slide 20 text

Machine Learning Tuning Is Done, Now Letʼs Validate the Model Like a Pro ● Does hyperparameter tuning improve model performance? Compare performance metrics before and after tuning to ensure that the tuning process provides a significant improvement. ● Is the model overfitting or underfitting? Ensure that the model performs well on test data and not just on training data. ● Are there other metrics that should be considered? Sometimes, accuracy alone is not enough. Use additional metrics such as precision, recall, or F1-score to gain a more comprehensive understanding of the model's performance. ● A thorough evaluation ensures that the model produced after tuning Not only performs well on the training dataset but can also generalize effectively to new data.

Slide 21

Slide 21 text

Quiz #2 DevCoach 186 Machine Learning Format jawaban: #quiz2-username-jawaban

Slide 22

Slide 22 text

Quiz #2 DevCoach 186 Machine Learning a). Melakukan standarisasi pada outlier. b). Menghapus data outlier. c). Mengganti nilai outlier dengan nilai Q1 dan Q3. Berikut ini pilihlah opsi yang BUKAN teknik digunakan untuk menangani outliers dalam feature engineering?

Slide 23

Slide 23 text

Machine Learning Underfitting to Overfitting: Finding the Sweet Spot in Model Training

Slide 24

Slide 24 text

Machine Learning Underfitting to Overfitting: Finding the Sweet Spot in Model Training

Slide 25

Slide 25 text

Machine Learning Methods to Identify Overfitting and Underfitting in Machine Learning

Slide 26

Slide 26 text

Machine Learning Techniques to Overcome Overfitting: Regularization

Slide 27

Slide 27 text

Machine Learning Techniques to Overcome Underfitting: Using More Complex Models

Slide 28

Slide 28 text

Hands-on Coding: Let's Build It Machine Learning

Slide 29

Slide 29 text

Quiz #3 DevCoach 186 Machine Learning Format jawaban: #quiz3-username-jawaban

Slide 30

Slide 30 text

Quiz #3 DevCoach 186 Anda menggunakan learning curves untuk menganalisis model machine learning Anda. Jika learning curves menunjukkan bahwa error training stabil rendah, tetapi error validation tinggi dan menurun, apa indikasi utama dari model Anda? Machine Learning a). Model mengalami underfitting. b). Model mengalami overfitting.

Slide 31

Slide 31 text

Thank You Machine Learning

Slide 32

Slide 32 text

Feedback! dicoding.id/devcoachfeedback Machine Learning