Page 205 - Contributed Paper Session (CPS) - Volume 6
P. 205

CPS1883 Christina A. et al.
            about high integrity systems in the context of mission critical systems, which,
            in case of a failure, can cause immediate, essential problems for the operation
            of  an  organization,  such  as  ERP,  CRM,  etc.  This  means  that  the  computer
            scientists with a Master’s degree from HIS often hold a position, which involves
            both responsibility and a lot of data analysis.
                Within  the  HIS  program,  two  courses  concerning  statistics  and  data
            analysis are mandatory for all students: The students have to take the basic
            course Introductory Data Analysis and the semi-advanced course Data Mining
            Methods. In the first course, Introductory Data Analysis, it is assumed that the
            students already have passed a statistics course at B.Sc. level, i.e. the course
            starts with a review of basic concepts and then moves on to topics within
            inferential statistics and linear and logistic regression. The course Data Mining
            Methods introduces further methods for analyzing data, such as decision trees,
            neural networks and support vector machines, stressing the importance of
            data preparatory issues.
                After these two  courses, the students can choose to study the elective
            course Multivariate Data Analysis, which is a high-level course, directly based
            on the knowledge the students should have gained in the first two courses.
            The Multivariate Data Analysis course prepares for application of advanced
            multivariate statistical methods to practical problems and serves as a platform
            for those students who consider their master’s thesis to be within data analysis.
            Among  the  contents  included  in  the  course,  we  find  topics  as  principal
            component analysis (PCA), partial least squares regression (PLS), discriminant
            analysis, canonical correlation and cluster analysis. The objective of this course
            is to provide the students with both a sufficient theoretical foundation and in
            addition to this to give them a thorough knowledge of how to apply theory to
            some  real-world  situations  by  using  statistical  software,  such  as  R,  for
            analyzing large and complex data sets. The software R has been chosen, since
            it is used in many different industrial applications in companies in Germany.
            Furthermore, the software is easy to use, but still accommodates advanced
            statistical methods. Finally, another reason to use R is that it is free software,
            i.e. we do not have to deal with license costs and conditions.   The examination
            of the course is a computer-based, written exam, i.e. the exam takes place in
            a computer room. This enables us to test the students’ ability to apply the
            multivariate methods to real-world problems.
                The program HIS is taught as a traditional M.Sc. program and in general,
            the  courses  take  place  face-to-face  at  the  campus.  However,  in  order  to
            facilitate for the students to organize their busy schedule, the university seeks
            to offer online courses as an option for time and location independent studies.
            As  a  part  of  these  efforts,  the  course  Multivariate  Data  Analysis  has  been
            virtualized and is now offered as an online course.


                                                               194 | I S I   W S C   2 0 1 9
   200   201   202   203   204   205   206   207   208   209   210