with the improvement of profoundly efficient chemoinformatics data collection technology, classification of chemical data emerges as a vital topic in chemoinformatics. Towards building highly accurate predictive models for chemical data, here we introduce two feature selection algorithms. The first algorithm based on Chicken swarm optimization (FS-CSO) and the second algorithm based on Particle swarm optimization (FS-PSO). The proposed algorithms were applied to four datasets and FS-CSO proves advance over FS-PSO. Also, the two algorithms compared against two previous algorithms, BPSO-BP and BPSO-PSO, that used in feature selection for molecular classification and FS-CSO proves advance over them as well