Flower pollination algorithm for data generation and analytics - a diagnostic analysis

• Main Authors: Julius Beneoluchi Odili, A. Noraziah, Asegunloluwa Eunice Babalola
• Format: Article
• Language: English
• Published: Elsevier 2020-07-01
• Series: Scientific African
• Subjects: African buffalo optimization; Data analytics; Flower pollination algorithm; Shekel foxhole function; Speedy Genetic Algorithm; Sphere function
• Online Access: http://www.sciencedirect.com/science/article/pii/S2468227620301782

The effectiveness of optimization in scientific and engineering applications has made optimization a popular area of scientific investigation in data generation and analytics leading to the design of several optimization algorithms. In view of the huge number of optimization algorithms in literature, there is the need for a thorough diagnostic evaluation so as to bring out the strengths and weaknesses of each technique: that way, assist researchers in making informed choices whenever they are confronted with an optimization problem. This paper aims to fill the gap in literature of the Flower Pollination Algorithm in terms of diagnostic assessment of the impact of the number of iteration and search agents in solving the popular benchmark Sphere function and the unpopular but complex multimodal Dejong 5 function, otherwise called Shekel Foxhole function. After a number of empirical evaluations, the study finds out that the Flower Pollination Algorithm is not only a fast technique but also obtained good results when the appropriate iteration and flower population is used.