Data Science, MS
The College of Computer and Information Science (CCIS) and the Department of Electrical and Computer Engineering (ECE) jointly offer a new interdisciplinary Master of Science program in data science. This program is designed to give students a comprehensive framework for processing, modeling, analyzing, and reasoning about data. Students will engage in an extensive core intended to develop depth in computational modeling, data collection and integration, data storage and retrieval, data processing, modeling and analytics, and visualization. Students will also be given a variety of elective areas in CCIS, the College of Engineering (COE), and throughout the campus to explore key contextual areas or more complex technical applications. Successful program graduates will be well positioned to attain data scientist and data engineer positions in a fast-growing field or to progress into doctoral degrees in related disciplines.
The Master of Science in Data Science is comprised of eight courses; five core courses and three electives. The core courses are designed and developed by the CCIS and ECE faculty. Elective courses consist of graduate courses offered in CCIS, COE, and other partner colleges.
The Master of Science in Data Science curriculum requires five core courses that represent the essential mathematical/statistical and technical knowledge for deep data analysis. These courses examine foundational programming concepts and languages, integration, collection, storage, retrieval, large-scale computing, mathematical concepts in statistics, linear algebra, and optimization, as well as visual and computational analysis, machine learning, and visualization. The courses are tailored toward technically or mathematically trained students.
The five core courses include:
- Two core courses in algorithms and data processing
- Two core courses in machine learning and data mining
- One core course in information visualization
Three elective courses are drawn from a selection of courses across Northeastern.
Students who complete the MS degree will be able to:
- Collect data from numerous sources (databases, files, XML, JSON, CSV, and Web APIs) and integrate them into a form in which the data is fit for analysis
- Use R and Python to explore data, produce summary statistics, perform statistical analyses; use standard data mining and machine-learning models for effective analysis
- Select, plan, and implement storage, search, and retrieval components of large-scale structure and unstructured repositories
- Retrieve data for analysis, which requires knowledge of standard retrieval mechanisms such as SQL and XPath, but also retrieval of unstructured information such as text, image, and a variety of alternate formats
- Match the methodological principles and limitations of machine learning and data mining methods to specific applied problems and communicate the applicability and the advantages/disadvantages of the methods in the specific problem to nondata experts
- Carry out the full data analysis workflow, including unsupervised class discovery, supervised class comparison, and supervised class prediction; Summarize, interpret, and communicate the analysis of results
- Organize visualization of data for analysis, understanding, and communication; choose appropriate visualization method for a given data type using effective design and human perception principle
- Develop methods for modeling, analyzing, and reasoning about data arising in one or more application domains such as social science, health informatics, web and social media, climate informatics, urban informatics, geographical information systems, business analytics, bioinformatics, complex networks, public health, and game design
- Manage, process, analyze, and visualize data at scale. This outcome allows students to handle data where the conventional information technology fail.
Each incoming masters student, regardless of his or her background, takes two placement exams administered one week prior to the beginning of the semester. The two exams cover fundamentals of computer science and programming skills and basic statistics, probability, and linear algebra. If the student does not get a B or above in a part of the placement exam, then the student must take the corresponding introductory course.
- Introduction to Programming for Data Science (DS 5010) The introductory course on fundamentals of programming and data structures covers data structures (lists, arrays, trees, hash tables, etc.), program design, programming practices, testing, debugging, maintainability, data collection techniques, and data cleaning and preprocessing. This course will have a class project where the students will use the concepts they learn to collect data from the web, clean, and preprocess and ready for analysis.
- Introduction to Linear Algebra and Probability for Data Science (DS 5020) The introductory course on basics of statistics, probability, and linear algebra covers random variables, frequency distributions, measures of central tendency, measures of dispersion, moments of a distribution, discrete and continuous probability distributions, chain rule, Bayes' rule, correlation theory, basic sampling, matrix operations, trace of a matrix, norms, linear independence and ranks, inverse of a matrix, orthogonal matrices, range and null space of a matrix, the determinant of a matrix, positive semidefinite matrices, eigenvalues and eigenvectors.
Required Course Work
A grade of B or higher is required in the following courses:
|Complete 4 semester hours from the following:||4|
|Fundamentals of Computer Engineering|
|Data Management and Processing|
|DS 5110||Introduction to Data Management and Processing||4|
|Machine Learning and Data Mining|
|DS 5220||Supervised Machine Learning and Learning Theory||4|
|DS 5230||Unsupervised Machine Learning and Data Mining||4|
|Presentation and Visualization|
|Complete 12 semester hours from the following:||12|
|College of Computer and Information Science|
|Foundations of Artificial Intelligence|
|Natural Language Processing|
|Empirical Research Methods|
|Special Topics in Artificial Intelligence|
|Special Topics in Database Management|
|College of Engineering|
|Special Topics in Civil Engineering|
|Detection and Estimation Theory|
|Advanced Computer Vision|
|Advanced Machine Learning|
|Data Mining for Engineering Applications|
|Data Mining in Engineering|
|Statistical Methods in Engineering|
|College of Social Sciences and Humanities|
|Dynamic Modeling for Environmental Decision Making|
|Big Data for Cities|
|Geographic Information Systems for Urban and Regional Policy|
|Urban Theory and Science|
|Advanced Spatial Analysis of Urban Systems|
|Perspectives on Social Science Inquiry|
|D'Amore-McKim School of Business|
|Business Analytics Fundamentals|
|Predictive Analytics for Managers|
|Introduction to Big Data and Digital Marketing Analytics|
|College of Science|
|Statistics for Bioinformatics|
|Complex Networks and Applications|
|Network Science Data|
|Bouvé College of Health Sciences|
|Epidemiology and Population Health|
|Introduction to Epidemiology|
|Biostatistics in Public Health|
|College of Arts, Media and Design|
|Game Design and Analysis|
Note: Students that take 3-credit-hour elective courses (i.e., Bouvé, CSSH courses) will register for an accompanying data science project course in the same semester ( (DS 8982)). In order to earn this additional credit, students will be expected to work with faculty to design an additional project in line with the curricular aims of their chosen elective and the data science core learning outcomes.
Program Credit/GPA Requirements
32 total semester hours required
Minimum 3.000 GPA required