Data Science Specialization

The Data Analytics Journey uses the analytics life cycle to conceptualize the processes, tools, and techniques for implementing data analysis, data engineering, and decision process engineering. Students gain fluency in gathering requirements, asking business questions, establishing evaluation metrics, identifying communication models, and aligning the analytics project outcomes to business goals. This course presents an overview of the various tracks offered in the program and the career options in these specializations. There is no prerequisite for this course.

Data Management builds proficiency in using both relational and non-relational databases. Topics include selection of a data storage architecture, data types, data structures, normalization and denormalization, and querying databases. Structured Query Language (SQL) topics such as Data Definition Language (DDL) and Data Manipulation Language (DML) are covered, including joins, aggregations, and transactions. Non-relational approaches to organizing and querying data are contrasted with relational approaches to build competency in adapting data storage architectures to business needs.

Analytics Programming builds algorithmic thinking using both the Python and R programming languages. This course builds from the foundations of programming. Learners use libraries and packages to perform common analytics tasks, including acquiring, organizing, and manipulating data sets. The course also presents methods for applying statistical functions and graphical user interfaces to perform basic analysis and to present findings. There is no pre-requisite for this course.

Data Preparation and Exploration applies analytical programming skills to the early steps of the data analytics life cycle. This course covers cleaning data to ensure the structure, accuracy, and quality of the data; interpretation of descriptive and inferential statistics as well as visualizations of data; and wrangling data to prepare it for further analysis. The course introduces hypothesis testing, focusing on application for parametric tests, and addresses communication skills and tools to explain an analyst’s findings to others within an organization. The following courses are prerequisites: The Data Analytics Journey, Data Management, and Analytics Programming.

Statistical Data Mining focuses on concepts in data preparation and supervised and unsupervised machine learning techniques. The course helps students gain basic knowledge in statistics, data preparation, regression, and dimensional reduction. Learners implement supervised models—specifically classification and prediction data mining models—to identify relationships among variables that are not apparent with more surface-level techniques. The course also explains when, how, and why to use unsupervised models to best meet organizational needs. The following course is a prerequisite: Data Preparation and Exploration.

Data Storytelling for Diverse Audiences focuses on communicating observations and patterns to diverse stakeholders, a key aspect of the data analytics life cycle. This course helps learners gain communication and storytelling skills in order to motivate change and answer business problems. It also covers data visualizations, audio representations, interactive dashboards, interpersonal communication, and presentation skills.

Deployment is the practice of operationalizing data analysis within a business environment. Given an analysis, learners determine the business functional and non-functional requirements for wider use and implement pipelines and functions to deploy analyses at scale. Topics including security, scalability, usability, and availability are discussed. Prerequisites for this course are Analytical Programming, Data Management, Data Preparation, and Statistical Data Mining.

Advanced Analytics extends analytics techniques from machine learning to artificial intelligence more broadly, including topics in neural networks, deep learning, and natural language processing. The course covers approaches to developing these models including PyTorch and TensorFlow. Students learn to apply a combination of techniques to solve complex business challenges including computer vision and sentiment analysis.

Optimization is a large class of business problems requiring the iterative algorithmic maximization or minimization or one or more variables. Students in this course will select and use a variety of optimization approaches to address various business needs. The course covers classes of optimization problems at a foundational level (continuous/discrete, linear/nonlinear, and bounded/unbounded) and the solving of linear optimization problems in both Python and R through the use of gradient and non-gradient-based algorithms. Analytics Programming is a prerequisite. 

Machine Learning is the broad discipline of developing algorithms and statistical models to predict, classify, or cluster data and that iteratively improve over time. Machine Learning focuses on building, training, running, and testing supervised and unsupervised models and quantifying the accuracy and precision of those models to determine which may best be used in a particular business situation. Supervised methods covered include k-nearest neighbors, logistic regression, decision trees, and support vector machines. Unsupervised models covered include k-means clustering, hierarchical clustering, and t-distributed stochastic neighbor embedding (t-SNE). Ensemble methods are also presented. Prerequisites are Analytics Programming and Statistical Data Mining.

The Data Science Capstone integrates the learning in the MSDA core and the three courses within the specialization. The student evaluates various needs and opportunities in an organization or marketplace; identifies the business requirements; translates the business requirements into technical requirements; and creates a comprehensive project plan to solve the problem in a way that satisfies the customer or business needs. Projects within this specialization include the design and construction of machine learning approaches, optimization, and/or advanced analytics techniques as the project requires.