only need question 2.a

1. Frequent Pattern Mining - use for part (c) an approach different from the one used for (a) a. Find the frequents sets of courses by assuming minimum support is 30% b. Rotate/transpose the table c. Find sets of students mostly taking courses together (min_sup=1/3) d. Find two association rules between courses and three association rules between students assume minimum confidence is 50% in both cases. 2. Clustering a. Use k-means algorithms to cluster students into 3 clusters by considering courses as their features and use as distance ( 8 - number of common 1 's) b. Use DB-Scan to cluster students by using same distance function and consider epsilon as 4 and two as number of neighbors for a core object. c. Compare the results from (a) and (b) by elaborating on the effectiveness of each approach