400
Machine and Statistical Learning is a collection of mathematical and statistical techniques used to detect, classify, and infer patterns in large and/or complex data sets. Examples of machine and statistical learning algorithms are all around us: speech recognition on your phone, text prediction in internet searches, medical school placement algorithms, and the prediction of what you may want to watch next on your video stream. This course gives an overview of many concepts, techniques, and algorithms in modern machine and statistical learning including both supervised and unsupervised learning. Topics include linear regression, classification, cross validation, dimension reduction, nonlinear regression, tree-based methods, support vector machines, principal component analysis, artificial neural networks, and clustering. The course will give students the ideas and intuition behind modern machine and statistical learning methods as well as a more formal understanding of how, why, and when they work. The underlying theme in the course is application of the algorithms to real data sets.
This course will explore current techniques in computation for data science, including (but not limited to) parallel and distributed coding algorithms and modern programming languages.
This course will present students with a substantial experience in data analysis. Students will investigate and analyze data from a variety of sources, working both as individuals and in project teams. This serves as a capstone experience for the DS major.
Special Topics courses include ad-hoc courses on various selected topics that are not part of the regular curriculum, however they may still fulfill certain curricular requirements. Special topics courses are offered at the discretion of each department and will be published as part of the semester course schedule - view available sections for more information. Questions about special topics classes can be directed to the instructor or department chair.
The senior thesis is designed to encourage creative thinking and to stimulate individual research. A student may undertake a thesis in an area in which s/he has the necessary background. Ordinarily a thesis topic is chosen in the student's major or minor. It is also possible to choose an interdisciplinary topic. Interested students should decide upon a thesis topic as early as possible in the junior year so that adequate attention may be given to the project. In order to be eligible to apply to write a thesis, a student must have achieved a cumulative grade point average of at least 3.25 based upon all courses attempted at Carroll College. The thesis committee consists of a director and two readers. The thesis director is a full-time Carroll College faculty member from the student's major discipline or approved by the department chair of the student's major. At least one reader must be from outside the student's major. The thesis director and the appropriate department chair must approve all readers. The thesis committee should assist and mentor the student during the entire project. For any projects involving human participants, each student and his or her director must follow the guidelines published by the Institutional Review Board (IRB). Students must submit a copy of their IRB approval letter with their thesis application. As part of the IRB approval process, each student and his or her director must also complete training by the National Cancer Institute Protection of Human Participants. The thesis is typically to be completed for three (3) credits in the discipline that best matches the content of the thesis. Departments with a designated thesis research/writing course may award credits differently with approval of the Curriculum Committee. If the thesis credits exceed the full-time tuition credit limit for students, the charge for additional credits will be waived. Applications and further information are available in the Registrar's Office.