Learning a ranking function for information retrieval using HybridABC
Abstract
In this paper we propose a ranking algorithm, HybridABC that is built on swarm based
algorithm. In our proposed HybridABC algorithm we merged Artificial Bee Colony (ABC)
algorithm with Differential Evolution (DE) algorithm. The ABC is a swarm-based metaheuristic
algorithm inspired by the intelligent foraging pattern of bees and Differential
Evolution is a population-based stochastic search technique. The proposed implementation of
ABC has been tested using the LETOR dataset, which is a standard benchmark dataset for
evaluating ranking functions. Our results display that our proposed HybridABC can compete
and in many cases more efficient than other state-of-the-art algorithm proposed in ranking
web pages based on Genetic Algorithm (GA).