HW 02 - 20.09. Frequent Itemsets
1. Task 2 from book, page 404.
2. Task 7 from book, page 406.
3. Task 6 from book, page 406.
4. Use some software from Links and run frequent itemset finding Apply it on toy data sets, like the ones above (task1, 3).
5. Get the anonymised supermarket dataset http://fimi.ua.ac.be/data/retail.dat from http://fimi.ua.ac.be/data/ Analyse it with some implementation from the existing software list on Links.
- What is the "reasonable" support for you to "understand" the results?
- How long does the analysis algorithm run on different support thresholds?
- Report some interesting association rules (frequent and with high confidence)
6. Bonus (3p) Make a performance comparison between 3 different implementations or algorithms using 3 different data sets