Enroll Course: https://www.udemy.com/course/ittensive-machine-learning-clustering/
Are you looking to dive deep into the world of unsupervised machine learning? The “Машинное обучение: кластеризация и аномалии на Python” course from ITtensive on Udemy is an excellent choice for anyone seeking to understand and implement clustering and anomaly detection techniques using Python.
This course, the second in their unsupervised machine learning series, uses a practical approach, focusing on a real-world hackathon challenge from Yandex.Nedvizhimosti to predict the listing exposure period. This hands-on experience makes the learning process engaging and relevant.
The course is meticulously structured into four parts, ensuring a comprehensive learning journey:
**Part 1: Data Workflow Fundamentals**
This section lays the groundwork by covering the entire data workflow, from defining problem types and setting objectives to working with machine learning models to minimize prediction errors. It also delves into the fundamental principles of building ML models, essential metrics, and introduces basic models like linear regression and ensemble methods.
**Part 2: Basic Clustering Models**
Here, you’ll explore the core of clustering. You’ll learn about internal and external clustering metrics, master K-Means and FOREL algorithms, and practice their application. The course also covers agglomerative clustering and the concepts behind Mahalanobis distance and Gaussian Mixture Models (GMM). A practical assignment involves building a simple clustering model for raw data.
**Part 3: Advanced Clustering Techniques**
This part takes your clustering skills to the next level. You’ll differentiate between DBSCAN, HDBSCAN, and OPTICS, understand proximity propagation, and explore growing neural gas and Self-Organizing Maps (SOM). Advanced topics like the Kirchhoff matrix, spectral clustering, and building ensemble clustering models are also covered.
**Part 4: Anomaly Detection Mastery**
The final section focuses on anomaly detection. You’ll learn about the pAUC metric, apply the Smirnov-Grubbs test, and practice ellipsoidal approximation. Key algorithms like LOF, ABOD, COPOD, and Isolation Forests (including extended versions) are explained and implemented. The course culminates in building a solution for the 2020 Hackathon challenge.
**Recommendation:**
This course is highly recommended for its practical approach, comprehensive coverage of both basic and advanced techniques, and real-world application. The instructors provide a structured learning path that builds knowledge progressively. While the course is in Russian, the detailed curriculum and the focus on Python implementation make it accessible for those with some understanding of the language or a willingness to use translation tools. If you’re serious about mastering clustering and anomaly detection in Python, this ITtensive course on Udemy is a valuable investment.
**Important Note:** To access ITtensive courses on Udemy, you need to email support@ittensive.com with the course title or the group of courses you wish to enroll in.
Enroll Course: https://www.udemy.com/course/ittensive-machine-learning-clustering/