PROJECT TITLE :

A Cooperative Coevolution Framework for Parallel Learning to Rank

ABSTRACT:

We have a tendency to propose CCRank, the first parallel framework for learning to rank primarily based on evolutionary algorithms (EA), reaching to significantly improve learning potency whereas maintaining accuracy. CCRank is based on cooperative coevolution (CC), a divide-and-conquer framework that has demonstrated high promise in operate optimization for problems with giant search house and complex structures. Moreover, CC naturally allows parallelization of sub-solutions to the decomposed sub-problems, that can substantially boost learning potency. With CCRank, we tend to investigate parallel CC in the context of learning to rank. We tend to implement CCRank with three EA-based mostly learning to rank algorithms for demonstration. Extensive experiments on benchmark datasets in comparison with the state-of-the-art algorithms show the performance gains of CCRank in potency and accuracy.