Unsupervised Machine Learning Free Course for Intermediate
Introduction
Unsupervised Machine Learning is an intermediate course for individuals with an intermediate understanding of machine learning. It has seven modules, including a final project that will test your knowledge at the end of the course. This course offers a basic introduction to different types of machine learning. Within this course, students will explore ways to find insights from data sets and filter out data that is not targeted or labeled. Students will learn about clustering and dimension reduction algorithms for unsupervised learning and extract the algorithm suitable for the data.
What Will You Gain From This Course?
Following the completion of this course, students will be:
- Able to understand different types of problems suitable for unsupervised learning approaches.
- Have a better understanding of how dimensionality makes clustering difficult with many features.
- Able to explain and apply common clustering techniques.
- Able to use dimensionality-reduction algorithms.
- Able to utilize clustering points as needed.
- Able to compare the performance of per-cluster models.
- Have a better understanding of relevant metrics for cluster-characterizing.
Skills Acquired:
- Cluster Analysis
- Dimensionality Reduction
- Unsupervised Learning
- Principal Component Analysis (PCA)
- K Means Clustering
- Machine learning
- Clustering
- Dimension Reduction Algorithms
- Data Analysis
- Algorithms
Who Can Benefit From This Course?
This course is designed for:
- Machine learning enthusiast eager to use new concepts to solve problems.
- Aspiring data scientists looking for ways to have a hands-on experience with machine learning.
- Junior data scientists with theory-based knowledge but are struggling with details.
- Individuals with a basic understanding but not enough information about business settings.
- Data analyst struggling with quality issues.
- Student with basic understanding but not enough work experience.
Course Content
7 Modules – 38 Videos - 7 Readings – 9 Quizzes – 13 App Items - 1 Peer Review – Certificate of Completion
Introduction to Unsupervised Learning and K Means
This is the first module of the course and spans around three hours. This module offers an introduction to unsupervised learning and how it can be used in real life. For students still struggling with theory and concepts, this module offers a much better knowledge of all the theories behind K means and the way to use it in real life.
Apart from unsupervised learning, the participant will also learn about the practical application of clustering observations using k-means. This section also includes a case study for unsupervised learning. The quizzes within this section offer an opportunity to assess knowledge.
- 11 - Videos
- 2 - Readings
- 3 - Quizzes
- 3 - App Items
Distance Metrics and Computational Hurdles
The second module of this course will take more than three hours to complete. This module offers information about distance metrics and computational hurdles. The student will get to learn about different concepts, including Distance Metrics, i.e., Euclidean distance, Manhattan Distance, Cosine distance, and Jaccard Distance.
Moreover, participants will also get to explore the limitations and curse of Dimensionality. This module contains two quizzes and two app items, i.e., demo lab and practice lab. The quiz part offers an evaluation opportunity, whereas the app items offer a chance to practice all the newly learned skills.
- 6 - Videos
- 1 - Reading
- 2 - Quizzes
- 2 - App Items
Selecting Cluster Algorithm
This is the third module of the course, and it will take nearly four hours to complete. In this module, the student will learn about computational hurdles that the analyst has to face while clustering algorithms. It also offers an in-depth understanding of different clustering implementations and ways to overcome them.
By the end of this module, the learner will master the recapitulation of common clustering algorithms and ways to compare them. Following the completion of this module, the student will become proficient enough to choose and utilize clustering techniques that best suit the data.
- 11 - Videos
- 1 - Reading
- 3 - Quizzes
- 3 - App Items
Dimensionality Reduction
The fourth module of this course is about Dimensionality Reduction. It spans four hours and offers an in-depth understanding of dimensionality reduction and Principal Component Analysis. It starts with fundamental concepts and then explores more advanced topics, such as practical application.
This module's goal is to equip the student with enough knowledge to apply dimensionality reduction and Principal Component Analysis to big data, imaging, and pre-processing data. The section also includes two quizzes for self-assessment.
- 5 - Videos
- 1 - Reading
- 2 - Quizzes
- 4 - App Items
Non-Linear and Distance Dimensionality Reduction
The fifth module of this course spans two hours and mainly introduces dimensionality reduction techniques like Kernelal Principal Component Analysis and multidimensional scaling. The student will learn about these techniques and ways to utilize them.
Both techniques are considered relatively better than Principal Component Analysis in many applications. By the end of this module, students also learn about Kernel PCA, Multidimensional Scaling, its use, requirements, and real-life applications.
- 2 - Videos
- 1 - Reading
- 2 - Quizzes
- 3 - App Items
Metrix Factorization
The sixth module of this course is all about metric factorization and requires nearly three hours to complete. The module starts with a brief introduction to matrix factorization and then talks about ways to use matrix factorization for big data, text mining, and pre-processing data.
The module also includes two quizzes for self-assessment. By the end of this course, the student will be able to use matrix factorization as a class of collaborative filtering algorithms within the recommender systems.
- 3 - Videos
- 1 - Reading
- 2 - Quizzes
- 3 - App Items
Final Project
This is the final part of the course and spans just one hour. Now that the student has all the essential tools in the toolkit for highlighting unsupervised learning abilities, they are expected to use them. Students can choose any sample and apply all the techniques learned throughout the course.
By the end of this course, there is a peer review as well. Students can get to interact with the teacher, fellow students, and peers with better understanding. Within this interactive session, the user can leave suggestions about the course and ways to make it better.
- 1 - Peer Review
Description
This is not a beginner-level course. Instead, this is a course that requires some basic understanding of programming; before enrolling in this training program, the student is expected to have familiarity with Python development environment, Data Cleaning, Exploratory Data Analysis, Calculus, Linear Algebra, Probability, and Statistics.
All these fundamental concepts are taught in data analysis. However, students with limited knowledge are encouraged to take a fundamental course in data analysis. This course mainly focuses on Machine Learning, specifically Unsupervised Learning. However, there are other machine learning types as well. If the student wishes to explore machine learning more, there are other available courses.
This course has seven modules. Each module focuses on different concepts and offers an opportunity to put the newly learned concepts to use. The first module offers a basic introduction of introduction to Unsupervised Learning and K Means.
The second module explores concepts like Distance Metrics and Computational Hurdles. The third and fourth modules equip the student with skills like selection of Cluster Algorithm and Dimensionality Reduction. At the same time, the fifth and sixth modules are all about Non-Linear, distance-based Dimensionality Reduction, and matrix factorization. Finally, the student will have to create a final project that will be assessed based on the newly learned skills.
Meet the Instructor
Don Tapscott - Adjunct Professor – INSEAD
Don Tapscott is a leading authority on the impact of technology on business and society. His expertise is evident in his over 16 books on the subject, including the influential "Wikinomics: How Mass Collaboration Changes Everything," which is taught at leading universities worldwide and translated into over 25 languages.
Tapscott's passion for technology extends to the next generation. He co-authored a book with his son, Alex, a prominent figure in blockchain and cryptocurrency, exploring how to navigate the coming wave of technological disruption.
Don Tapscott's impressive credentials include co-founding the Blockchain Research Institute and being a member of the Order of Canada. He currently leverages his knowledge as an Adjunct Professor at INSEAD and Chancellor of Trent University in Ontario.