Course
Discontinued
No
Course Code
CSIS 3290
Descriptive
Fundamentals of Machine Learning in Data Science
Department
Computing Studies & Information Systems
Faculty
Commerce & Business Administration
Credits
3.00
Start Date
End Term
202030
PLAR
No
Semester Length
15 Weeks
Max Class Size
35
Contact Hours
Lecture: 2 Hours per week
Laboratory: 2 Hours per week
Total: 4 Hours per week
Method(s) Of Instruction
Lecture
Lab
Learning Activities
Lecture, seminars, demonstrations, and hands-on exercises/projects in the lab
Course Description
In this course, students will learn to apply machine learning concepts to analyze data and make predictions. Students will learn how to collect and wrangle data, to explore data using statistics and visualizations, to transform data for further modeling, to model data using machine learning algorithms to predict data patterns, and to evaluate these model-based predictions. Students will be expected to have prior experience with fundamentals of programming.
Course Content
- Programming language review for data analytics
- Basic syntax, variables, control flow, loops, install and import libraries for data processing such as SciPy, NumPy, Pandas, Sci-Kit Learn, TensorFlow or other similar libraries and packages.
- Data And Features: using libraries such as NumPy and Pandas
- Represent data using lists, arrays for structured data
- Work with data frames using packages such as Pandas to represent diverse data
- Use Control Flow for filtering data and performing filtered computations
- Manipulate Data using functions and packages to process the data and perform computations
- Understand, determine and represent Features
- Perform Data Wrangling
- Exploring Data: - using libraries such as Matplotlib
- Visualize Data by creating plots using tools such as Matplotlib
- Perform high dimensionality visualizations
- Transforming Data: using libraries such as Sci-Kit learn
- Create Data Transformers and apply dimensionality reducing techniques as PCA
- Data Modeling: using libraries such as Keras, TensorFlow and scikit-learn
- Use machine learning techniques such as clustering, supervised learning, K-nearest neighbours, Regression to model the data
- Evaluating Data: Evaluate modeled data using evaluation techniques
- Create and apply confusion matrices
- Perform cross-validation using scoring metrics
- Implement and apply power tuning and pipelining to evaluate the data
Learning Outcomes
At the end of this course, the successful student will be able to:
- Install and use appropriate tools and libraries needed for Data Science
- Understand and process data and features
- Collect and Wrangle Data for further processing
- Explore Data using statistics and visualizations
- Transform Data to a structure suitable for data modeling
- Model Data using machine learning algorithms
- Evaluate model-based predictions
Means of Assessment
|
Labs |
0-5% |
|
Project(s) |
15-25% |
|
Midterm Examination |
30-40% |
|
Final Examination* |
30-40% |
|
Total |
100% |
Some of these assessments may involve group work.
* Practical hands-on computer exam
In order to pass the course, students must, in addition to receiving an overall course grade of 50%, also achieve a grade of at least 50% on the combined weighted examination components (including quizzes, tests, exams).
Textbook Materials
Custom courseware, class notes provided by the instructor, and online resources or other textbooks as approved by the department
Prerequisites
Equivalencies