Машинное обучение: кластеризация и аномалии на Python
Learn unsupervised machine learning! Discover data clusters & identify anomalies. Master powerful techniques for data analysis.
Course Content
Машинное обучение: кластеризация и аномалии на Python is an advanced Udemy course taught by Центр digital-профессий ITtensive. It runs 7h 56m, is taught in Russian, and includes a certificate of completion. Learners rate it 4.9 out of 5 from 4 reviews, with 84 students enrolled. It is a paid course, last updated January 2022.
Unlock the power of unsupervised machine learning with this comprehensive course! This is the second installment in our series, building upon foundational knowledge to master data analysis techniques.
Access to this course and the entire "Unsupervised Machine Learning" series on Udemy is available – simply email [email protected] with the course or series name you'd like to access.
This course focuses on clustering and anomaly detection, demonstrated through a real-world Yandex.Realty hackathon challenge focused on predicting property listing duration. You'll gain practical skills applicable to various data science projects.
The course is structured into four key parts:
Part 1: Data Preparation & Foundational Models: We'll cover the entire data science workflow, from problem definition to model selection. Learn fundamental machine learning models like linear regression and ensemble methods. Understand essential metrics for evaluating model performance and minimizing prediction error. This section provides a solid base for advanced techniques.
Part 2: Clustering Fundamentals: Dive into core clustering algorithms. Explore K-Means and FOREL clustering, understanding both internal and external clustering metrics. Learn the principles of agglomerative clustering and apply it to real-world datasets. Discover the Mahalanobis distance and the workings of Gaussian Mixture Models (GMM). You'll build a simple clustering model from scratch as a practical exercise.
Part 3: Advanced Clustering Techniques: Explore cutting-edge clustering methods including DBSCAN, HDBSCAN, and OPTICS. Understand the principles of density-based clustering and the expanding neural gas model. Experiment with Self-Organizing Maps (SOM) and the Kirkhoff matrix. Apply spectral clustering and combine multiple clustering models into an ensemble.
Part 4: Anomaly Detection & Advanced Methods: Master anomaly detection techniques and the pAUC metric. Utilize the Smirnov-Grubbs test and explore ellipsoidal approximation. Compare and contrast Local Outlier Factor (LOF) and ABOD. Learn to train and apply the COPOD model, as well as iForest and Isolation Forest for robust anomaly detection. Conclude by applying your skills to solve the 2020 Yandex.Realty hackathon challenge.
This course is ideal for data scientists, machine learning engineers, and anyone looking to develop practical skills in unsupervised learning, data clustering, and anomaly detection. Gain valuable experience with industry-standard techniques and build a portfolio-worthy project.
Want AI in your product? Hire the team behind TutorialSearch.io
LLM features, RAG, agents and automation wired into a real product — built by the devs who shipped this site.