Course summary

This hands-on 1-day, instructor-led, live, online training course covers industry-standard and less common, more specialised clustering algorithms and customer segmentation methods along with their computational implementations in Python programming language. It also serves as an in-depth introduction to most recent and cutting-edge Big Data cluster analysis tools and methods which provide greater computational efficiency and scalability compared to typical clustering methods. Through a series of coding tutorials, brief lectures and short practical exercises, the course demonstrates how to apply, optimise, visualise and evaluate clustering methods in academic, industrial and business settings e.g. in social science, biological and medical sciences, genetics, customer/product segmentation and recommendation systems.

All clustering algorithms presented during this course will be implemented in Python programming language either through the custom-made code or with functions and methods available in Python libraries e.g. NumPy, pandas, SciPy, Scikit-Learn, and Statsmodels.

This course also includes a two-week follow-up period during which you are encouraged to ask questions directly to the tutor (via email) e.g. to advise you on methods or specific Python functions which may be suitable in your analysis.

Who is this course for?

This instructor-led, live, online, short course is suitable for post-doctoral researchers, Master’s or PhD students, industry data analysts or enterprise data/ML scientists and ML engineers, who are currently using the Python programming language (preferably at intermediate level) and would like to expand their skills to include theoretical understanding and practical implementations of industry-standard and modern, cutting-edge, scalable clustering and customer segmentation algorithms in Python.

What is included in the course fee?

At Mind Project, we pride ourselves for delivering great quality, goal-directed, interactive and cost-effective training. Our remote, hybrid, and on-premises tutor-led courses are designed to maximise your opportunity to learn complex topics in shortest possible time while being able to discuss and interact with our tutor and other fellow attendees. Therefore, the registration fee for this course includes a number of benefits that many other training providers don’t offer:

• a remotely delivered live training with our expert tutor over 1 full day (from 9:30 until 16:30) – you will be able to ask questions, interact and discuss the topic with other attendees from the comfort of your home/office,
• you will have access to course online materials for 1 year via our Mind Project Learning Platform e.g. course video recordings, exercises and quizzes, Python code scripts used during tutorials and live sessions, example datasets, optional and mandatory reading (e.g. blog articles, academic papers, industry reports), and other external recommended resources e.g. online books,
• two-week post-course follow-up period during which you will be able to ask questions to our tutor related to the material presented during the course and receive our recommendations of Python methods/functions suitable for your analysis and data,
• exercises and tasks completed by you during the live sessions will help you in better material retention and will enhance your learning progress,
• small group size allows easy interaction and stimulating environment for successful learning,
• you will receive a certificate of course completion for attending this course.

Programme outline

During this instructor-led live course you will:

• Implement industry standard partitional, linkage-based, density-based or spectral clustering and customer segmentation methods such as k-means and hierarchical clustering approaches, as well as less common, but more specialised algorithms such as affinity propagation, mean shift, and DBSCAN, to identify meaningful clusters in datasets from social science, biology/medical science/genetics, and business/finance fields,
• Learn about recent, cutting-edge cluster analysis techniques applicable to Big Data e.g. parallel implementations of density-based lightning connection clustering (LAPO-DBSCAN) and improved k-means (Wang, 2022) which rectify typical issues of common clustering approaches,
• Understand data processing requirements, computational efficiency and specific use cases for each presented method,
• Implement more complex and better optimised variants of typical methods by selecting their hyperparameters e.g. using different distance metrics, cluster linkage approaches and underlying mathematical algorithms,
• Evaluate and compare the clustering solutions returned by various methods and their variants through a number of metrics such Davies-Bouldin and Calinski-Harabasz scores, Dunn’s validation index, Silhouette Index, Jaccard Index, (Adjusted) Rand Index, Fowlkes-Mallows Index or (Adjusted) Mutual Information estimates,
• Visualise the obtained clusters in 2D and 3D plots, calculate profiling attributes for each cluster and interpret them,
• Discuss advanced methods of learning deep representations by reviewing Deep Learning clustering approaches of different types e.g. sequential multistep (e.g. Fast Spectral and Deep Sparse Subspace Clustering), joint (e.g. Task-Specific and Graph-Regularised Networks and Deep Clustering Networks) or closed-loop multistep (e.g. Deep Embedding Clustering – DEC) methods.