-
-
- Online Learning
- Custom Programs
-
-
-
- Academic Services
- Course and Program Information
-
-
-
Berkeley Global
Learn why the open-source programming language Python has been extensively adopted by the machine-learning community and industry. Python allows its users to create products that parse, reduce, simplify and categorize data, and then extract actionable intelligence from that data. In this course, you’ll use Python to understand machine-learning concepts, terms and methodology, and then build applications to gain an intuitive understanding of the mathematics underlying the program. Example real-world applications include search engines, image analysis, bioinformatics, industrial automation, speech recognition and more.
Prerequisites:
Required
- Knowledge of statistics as covered in a first semester undergraduate course. Need to fulfill this prereq? Take a course in:
- Introduction to Statistics STAT X10
- Introduction to Statistics STAT X10
- Ability to program in at least one high-level programming language such as Python, Java, Ruby, JavaScript, C or C++. Need to fulfill this prereq? Take a course in:
- Introduction to Computers and Programming COMPSCI X444.4
- Programming Python COMPSCI X434
- First Course in Java EL ENG X429.9
- JavaScript and jQuery: An Introduction COMPSCI X452.1
- Introduction to C Language Programming EL ENG X24
Course Outline
Expand or collapse section
Course Objectives
- Identify and formulate machine-learning problems.
- Understand and implement algorithms in Python to solve simple machine-learning problems.
- Analyze the performance of given or implemented machine-learning solutions on practical datasets.
What You Learn
- Machine learning and data mining
- Histograms classifiers
- Probability density functions
- Bayesian classifiers
- Class-conditional density, priors and posteriors
- Python packages
- Multidimensional data
- Covariance matrix
- Principal component analysis
- Elements of higher-dimensional visualization
- Linear classifiers
- Perceptron, Fisher's linear discriminant, least-mean-square
- Classifier performance evaluation
- Accuracy, sensitivity, specificity, positive predictive value
- Receiver operating characteristic
- Training, testing and validation
- Multinormal probability density function
- Weighted means and covariance matrix
- Mixture distributions
- Unsupervised learning
- Expectation maximization algorithm
- K-means algorithm, k-nearest neighbor algorithm
- Markov models
- Dynamic programming
- Viterbi algorithm
- Ensemble learning
- Outlier and anomaly detection
- Neural networks
- Activation functions
- Backpropagation algorithm
- Auto-encoder
How You Learn
- Lecture and presentation of main topics
- Worked examples that illustrate concepts
- In-class quiz
- Team activities
- Group project
- Online discussion and forum
Is This Course Right for You?
This course is geared to professionals, students and job-seekers interested in learning the fundamentals of machine learning and data mining and who want to build, evaluate or showcase machine learning applications in Python. Also, anyone interested in understanding the foundational mathematical concepts underlying modern data science should enroll.
Loading...