Data Mining using ANN (Informatics)

Type: For the student's choice

Department: discrete analysis and intelligent system

Curriculum

SemesterCreditsReporting
83Setoff

Laboratory works

SemesterAmount of hoursGroupTeacher(s)
814PMi-43

Practical

SemesterAmount of hoursGroupTeacher(s)
828PMi-43

Course description

Aim. Learn the concept of data mining and algorithms of cluster analysis.

Summary. The course considers the procedure of cluster analysis and algorithms of partitional & hierarchical clustering. Student will have to deal with terms like clustering, data set, feature, cluster, centroid and dendrogram. To perform an efficient data mining it is important to look at feature types and an appropriate proximity measures. After that clustering algorithms, namely K-means and UPGMA, are considered. Finally to assess the quality of clustering result student will get familiar with cluster validity measures.

Target. Master the concept of cluster analysis. Get acquainted with terms data mining, cluster, centroid, dendrogram, data set, feature. Learn feature types and an appropriate proximity measures. Master the concept of partitional and hierarchical clustering. Learn to apply K-means and UPGMA algorithms on practice; use cluster validity measures to assess quality of clustering.

After completion of this course a student should

  • know: procedure of cluster analysis; feature types and proximity measures; the concept of hierarchical and partitional clustering; clustering algorithms, namely K-means and UPGMA; cluster validity measures;
  • be able to: perform feature extraction for clustering algorithms; use an appropriate proximity measures; apply K-means and UPGMA algorithms; measure the quality of clustering.