Frequently Asked Questions
Individuals must have a knowledge base in calculus (2 semesters), plus multivariable calculus, linear algebra, and statistics (at least 1 semester each). Also required is programming experience in a general purpose programming language (e.g., C, C++, Java, Python), including basic algorithms and data structures equivalent to the first two semesters of beginning computer science courses either as part of the undergraduate degree or subsequent work experience. Experience with mathematical software environments such as Matlab, R or the equivalent is a big plus. Work experience will need to be documented on an applicant’s CV and supported by letters of recommendation if an applicant plans to use it in lieu of coursework.
You should have experience equivalent to 2 semesters of beginning computer science. This is more than just programming. Some sample topics are listed below. We do not expect most applicants have even seen all of these, but your experience should be at a level where you could learn these concepts fairly easily. If you have not had 2 semesters of formal course work, it would greatly help your application to describe your computing experience in your Statement of Purpose, for example by describing the most complicated computing problem you have solved. Many of our applicants have picked up some of the necessary experience through work or by taking some make-up courses (other than self-paced courses), so lacking formal experience in everything does not rule out admission. It will help if a letter of recommendation addresses this.
For those wishing for some extra formal training before entering our program, we recommend the equivalent of our Computer Science classes CSci 1933 and 4041 (with CSci 1133 only for those with almost no computing experience with a general purpose programming language). In addition, our CSci 2021 and 4061 are good preparation for those interested in large-scale data applications that depend on parallel, distributed, or cloud computing. In any case, otherwise strong candidates can be admitted on condition that they take one or more of these courses in their first semester.
Here is a non-exclusive sample of topics it is helpful to know, or may need to succeed in our classes (the last few topics more critical for Big Data applications):
- divide and conquer problem solving paradigm.
- deques in object oriented fashion using linked lists
- symbolic algebra manipulation
- simple numerical and iterative methods
- sorting and searching: implementation & theoretical analysis of complexity of different methods
- concepts of hash tables or dictionaries, queues, stacks, [binary] trees, heaps
- implement a discrete event simulation of a set of waiting queues
- use simple interactive graphics program interface: e.g., make graphical objects that interact with a user.
- elementary graph algorithms: graph search (BFS,DFS). graph insert & delete.
- insert, retrieve & delete in various data structures: lists, trees, hash tables.
- dynamic programming
- memoization (what is it, when does it help)
- asynchronous processes and communication, locks and semaphores.
- data and algorithm abstraction and modularity.
Why must I provide a separate list of courses fulfilling the prerequisites when they are already listed on my transcript?
Our experience has shown that the greatest hindrance to success in our program has been a student's lack of previous preparation in computing and/or mathematics. By asking applicants to list the relevant courses already listed on their transcript, it is made clear to both the applicant and the program faculty how the background of the student fits with our program. Even if the preparation in one prerequiste topic area is weak, applicants who are strong in other areas may still be admitted, possibly with some conditions for make-up work. Applicants are encouraged to elaborate on their experience in the prerequisite areas in their SOP.
The GRE is a mandatory component for a complete application for the M.S, but not for the Certificate. If you took the GRE so long ago that the official scores are no longer available, please include an unoffical copy of your old scores with your application and an explanation. Any application requires evidence that you can succeed in rigorous graduate-level academic coursework, and the GRE is an important, but not the only, piece of evidence for this. Hence any application for the M.S. without GRE scores will be considered incomplete.
TOEFL scores are mandatory for most international applicants applying to the Data Science masters program.
Exceptions may be made for those applicants that meet the following requirements:
Applicants who have completed 24 quarter credits or 16 semester credits (within the past 24 months)
Are in residence as a full-time student at a recognized institution of higher learning in the United States (or other English-speaking country) before entering the University of Minnesota may be exempted from this requirement.
Applicants who have been working full-time in the United States in a Data Science related field for at least two years can inquire about an exemption.
If you have questions regarding this exception please contact us.
Yes,the program can be completed in two years. For more information on courses and curriculum please visit the curriculum page. The Post-Baccalaureate Certificate can be completed in one year. Students may opt to enroll as part-time students, taking longer to complete their degree.
9 hours of graduate coursework equates to roughly 40 hours of work each week. It is highly recommended that students take 6 credit hours if they plan to work while attending the program.
Most course requirements for this degree can be met using courses offered over the University's instructional video system, UNITE (www.unite.umn.edu), without coming to campus.
Occasional visits to campus may be necessary for certain courses including the capstone project.
I am not yet in Data Science, but have already taken all the courses needed for a track. Can I use them as part of my Data Science program?
A limited number of credits for courses taken here at the University of Minnesota after obtaining your bachelor's degree can be used toward your Data Science degree, whether or not admitted to the Graduate School, as long as you register for graduate credit.
Required courses taken here as an undergraduate cannot be used directly, but once enrolled in the Data Science program you can replace any required course already taken with a more advanced course in the same track, or a related elective, with DGS approval.
Courses taken as a matriculated U of MN graduate student may transfer with faculty review and approval, up to a limit.
The Data Science Program has no internal funding sources, inasmuch, students are expected to provide their own funding.There are rare cases, after the student has established a record at the U of M, when MS students do get a teaching assistantship through a participating department or a research assistantship through an individual professor. More information must be obtained from the individual departments regarding funding opportunities after being accepted to the program.
Yes, credits may be transferred into the program.The University of Minnesota requires that 60% of the MS coursework be taken at the University as an admitted, registered graduate student. Given that the program is 31 credit hours, no more that 12 credits may be transferred. Additionally, a student may have no more than 8 semester credits in common between two MS degrees. If you already have an MS degree from another university, only 8 credits may be transferred and counted toward your degree. You may not transfer in courses from an undergraduate degree.
Tuition and fees for the current academic year can be found here:"http://onestop.umn.edu/finances/costs_and_tuition/index.html". Data Science students pay tuition at the general "Graduate and Professional" rate found here. Students enrolled entirely through UNITE pay tuition at the resident rate plus a UNITE fee regardless of where they reside.
Because Fall 2015 is the inagural year for the Data Science program we do not have historic numbers yet. This information will be made available as the program grows over time.
Yes, this program is listed as a STEM program, with a CIP code of 11.0401.