Bees Swarm Optimization for Association Rule Mining on CUDA

Bio inspired algorithms plays an important role in solving complex and huge problems in computer science. It helps us in finding the optimal desired way to solve extremely dynamic, difficult and robust problems. These algorithm help us to cope with the technological need of a new era. Association Rule mining (ARM) is well studied and famous optimization problem which finds useful rules from given transactional databases. Many algorithms are already proposed in past which shows their efficiency when dealing with different sizes of data-sets. Unfortunately, their efficiency is not enough for handling large scale data-sets. Bio inspired bees swarm intelligence algorithm for association rule mining is more efficient. These kinds of problems need more powerful processors and are time expensive. For such issues solution can be provided by graphics processing units (GPUs). They are massively multi-threaded processors. In this case GPUs can be used to increase performance of the computation. Bees swarm intelligence algorithm for association rule mining can be designed using GPUs in multi-threaded environment which will be efficient for given data-sets.

Autorentext

Alankar Shelar has completed his Master of Engineering in Computer Science. From last 7 yrs. he is working as Assistant Professor in Textile and Engineering Institute, Ichalkaranji. He is well known among students to teach different programming languages. His interested areas are Swarm Intelligence, Parallel Computing, Data Mining, Machine Learning