Description The main objective of this project is to find frequent itemsets by implementing two efficient algorithms:

A-Priori and PCY. The goal is to find frequent pairs of elements.

You do not need to find triples and larger itemsets.

Programming Language :Java

Dataset The retail dataset contains anonymized retail market basket data (88K baskets) from an anonymous retail store.

The preprocessing step to map text labels into integers has already been done. Use Sublime Text, TextPad or Notepad++ or other software to open the file.

Experiments Perform the scalability study for finding frequent pairs of elements by dividing the dataset into different chunks and measure the time performance. Provide the line chart. Provide results for the following support thresholds: 1%, 5%, 10%. For example, if your chunk is 10% of the dataset, you have around 8,800 baskets. Therefore, if your support threshold is 5%, you should count the pairs that appear in at least 440 baskets.

Implement Multistage (3 Passes) version of PCY, using one extra hash table. (add the results to the line chart) Implement Multihash version of PCY, using one extra hash table. (add the results to the line chart)

DATASETTT : a very large file of numbers