Introduction to Machine Learning and Data Mining | DBDA.X408
Machine learning (ML) is the foundation for many artificial intelligence (AI), and ML algorithms that underlie online shopping recommendations, credit card fraud detection, relevant social media content delivery, rideshare trip pricing, and traffic navigation.
In this course you’ll explore essential ML concepts, tools, and methodology, such as classical and modern algorithms that drive real-world applications such as search engines, image analysis, biometrics, industrial automation, and market segmentation. You’ll work with practical data-driven applications and gain a practical background for creating new products and improving existing ones.
Starting with an introduction to the mathematics underlying ML, we’ll leverage open source Python-based libraries, including Pandas, NumPy, and Sklearn. You’ll improve your intuitive understanding of the underlying algorithms, such as regression, classification, and clustering, as well as related Python-based code samples. You’ll work in a small team or by yourself on a project to present during the final week of class.
At the conclusion of the course, you should be able to:
- Identify and formulate ML problems
- Understand and implement algorithms to solve ML problems
- Explain the implementation, working, and practical benefit of many ML topics
- Analyze the performance of given or implemented ML solutions on practical datasets
- Defining ML using simple problems and intuitive solutions for supervised learning and Bayesian classifiers
- Probability density
- Linear classifiers—common straightforward classifiers with practical applications
- Cross-validation in data-poor situations
- Principal component analysis—correlation matrices, eigenvalues, and eigenvectors
- Unsupervised Learning: Using accumulated buying histories from a customer database to evaluate the quality of clustering results
- Neural networks and deep learning: Without using complex mathematics, learn how neural networks are trained (Tensorflow and Keras)
- Natural language processing: How computer algorithms glean meaning and sentiment from written text and respond intelligently
For best results in this class, the following topics are highly recommended, some of which are covered in the suggested prerequisite course (listed below):
- Familiarity with Google Colaboratory and Jupyter Notebooks
- Reasonably good programming/debugging skills beyond the basic or beginner level
- Familiarity with Python programming, Numpy, and Pandas
- Comfortable with basic knowledge of algebra, calculus, probability and statistics
Course Availability Notification
Please use this form to be notified when this course is open for enrollment.