Discovering Unsupervised Learning Methods in Python
This training course introduces unsupervised learning methods in Python. It helps you understand clustering algorithms, dimensionality reduction, and the identification of hidden structures in data. Through practical examples
Create your free account
How the course works
Each lesson is a structured text, illustrated with examples and diagrams, followed by a quiz that is automatically graded. You progress at your own pace, and your notes stay attached to the lesson in which you took them.
-
Acquire a solid understanding of the fundamental principles of unsupervised learning
-
Understand how the main clustering algorithms and dimensionality reduction techniques work
-
Learn to apply unsupervised methods in Python to analyze datasets
-
Learn to identify groups, trends and hidden structures in data
-
Develop the ability to interpret results and choose methods suited to data analysis challenges
-
Data analyst: collecting, processing and analysing data to identify trends and support decision-making.
-
Data scientist: designing and applying statistical methods and machine learning algorithms to extract relevant insights.
-
Machine learning engineer: developing and optimising machine learning models tailored to the needs of organisations.
-
Artificial Intelligence Analyst: studying data and using algorithmic methods to solve complex problems.
-
Data Science Consultant: helping companies choose and apply data analysis methods.
-
Data Segmentation Specialist: identifying homogeneous groups using clustering algorithms.
-
Anomaly Detection Analyst: spotting unusual behaviour or atypical data in various datasets.
-
Data-Oriented Python Developer: creating scripts and tools to prepare, analyse and exploit data.
-
Statistical Study Officer: exploring datasets and interpreting results to answer specific questions.
Accessibility

You have a disability and wish to take a training course with us? This email address is dedicated to you: contact@datagoz.com . For any other situation, don't hesitate to contact us to assess how to ensure a learning experience.
How do I finance my training?
Certificate of completion: at the end of this course, you will receive a DATAGOZ certificate of completion, issued in your name and verifiable online.
The DATAGOZ team wishes you a great training!
6 chapters, 54 lessons
- 2. Unsupervised Learningopen
- 2. Determine the optimal number of clustersopen
- 2. Cluster points in two dimensionsopen
- 2. Analyze the clustering resultsopen
- 2. Measure the quality of the clusters obtainedopen
- 2. Choose the number of clusters for the seedsopen
- 2. Evaluate the clusters formed from the seedsopen
- 2. Prepare the variables to get better clustering resultsopen
- 2. Standardize the fish data before clusteringopen
- 2. Identify clusters among the fishopen
- 2. Group stocks into clusters with K-meansopen
- 2. Spot stocks with similar trendsopen
- 3. Visualize hierarchical structures in the datasubscribers
- 3. Determine the number of merges requiredsubscribers
- 3. Group the seed data using hierarchical classificationsubscribers
- 3. Analyze the hierarchical relationships between the actionssubscribers
- 3. Assign labels to the groups resulting from hierarchical clusteringsubscribers
- 3. Identify the closest data groupssubscribers
- 3. Compare hierarchical clusterings according to linkage methodssubscribers
- 3. Study the clusters at different levels of the hierarchysubscribers
- 3. Extract the labels of the groups formedsubscribers
- 3. Project the data into two dimensions with t-SNEsubscribers
- 3. Visualize the grain data with t-SNEsubscribers
- 3. Map the similarities between stocks with t-SNEsubscribers
- 4. Visualize how PCA workssubscribers
- 4. Spot the natural correlations between the datasubscribers
- 4. Decorrelate the seed measurements using PCAsubscribers
- 4. Identify the principal componentssubscribers
- 4. Determine the intrinsic dimensionality of the datasubscribers
- 4. Analyze the role of the first principal componentsubscribers
- 4. Study the variance explained by the PCA componentssubscribers
- 4. Estimate the intrinsic dimensionality of the data on fishsubscribers
- 4. Reduce the number of dimensions with PCAsubscribers
- 4. Simplify the fish measurements using dimensionality reductionsubscribers
- 4. Represent words with TF-IDF scoressubscribers
- 4. Cluster Wikipedia articles, part 1subscribers
- 4. Continue clustering Wikipedia articles, part 2subscribers
- 5. Discover non-negative matrix factorization (NMF)subscribers
- 5. Identify data made of non-negative valuessubscribers
- 5. Apply NMF to Wikipedia articlessubscribers
- 5. Extract features from Wikipedia articles with NMFsubscribers
- 5. Reconstruct data from NMF componentssubscribers
- 5. Break down data into interpretable elements with NMFsubscribers
- 5. Identify document topics with NMFsubscribers
- 5. Explore the LED digits datasetsubscribers
- 5. Extract image components with NMFsubscribers
- 5. Understand why PCA cannot identify distinct partssubscribers
- 5. Build recommender systems with NMFsubscribers
- 5. Find articles similar to the one about Cristiano Ronaldosubscribers
- 5. Recommend music artists, part onesubscribers
- 5. Improve music artist recommendations, part twosubscribers
- 5. Review interpretable feature discovery methodssubscribers
Everything that comes with the lessons
How long, depending on your availability
What you get at the end
Once the course is completed, the certificate is issued in your name. Its code allows a recruiter to verify it in just a few seconds.
This is not a diploma or a professional certification; it is verifiable proof that you have followed and successfully completed this course.
Who writes this course
A clear, hands-on teaching approach, with concrete examples to help you learn effectively.
Before you start
Start this course today
13 lessons open, no credit card required. You'll see the level and tone before deciding.