An alternative design of the PageRank algorithm for web page optimization using an improved Max–Min ant system

Abstract

The rapid growth and the dynamic nature of the web have continued to bring chal
lenges to the PageRank algorithm. The PageRank algorithm remains the best ranking 
criteria so far. Many researchers have worked to accelerate PageRank computation, 
but these two challenges still remain an issue. We propose a model architecture that 
optimizes the dynamic nature of ant colony optimization to prune the cones of graph 
nodes and retrieve relevant nodes for users’ display. The potential benefit of our pro
posed design indicates an improvement over the traditional PageRank in case of its 
computation cost and efficiency.