Publication Date


Degree Type

Master's Project

Degree Name

Master of Science (MS)


Computer Science


Association rule mining helps us to identify the association between items from a large transactional data set. It has always been a time consuming process because of repeatedly scanning of the data set. Apriori Algorithm [1] and FP-Tree Algorithm [2] are the two methods to find out the association of items in a large transactional item set. Both the above algorithm works differently (Apriori follows Bottom-Up Approach & FP-Tree follows Top-Down Approach) in order to get the association. Associations of items generated from the above two algorithms can be represented in geometry. The geometrical form of associations is called Simplical Complex. By exploring the FP-Tree method and using the bit pattern of records we can quickly indentify the possible longest associations on transactional data. The proposed Bitmap method in using the FP-Tree method is a new approach which helps quickly by using Human-Aide to find out the longest association. This method quickly finds out the longest association of items by looking the bit pattern of items in a large item set. By aligning the similar bits of records and arranging the attributes in order of highest frequency first are the underline logics of the above algorithm to get the longest association of items.