Abstract: Bacterial Foraging Optimizer (BFO) is a very recent swarm intelligence technique inspired by the foraging behavior of Escherichia coli (E. coli). The key step in BFO is the chemotaxis movement of bacteria, which models a trial of solutions of the optimization problems. Based on our previous work, we proposed a modified BFO (MBFO), where a linear decreasing chemotaxis step mechanism is incorporated into run and swim step of chemotatix cycle of original BFO. To illustrate the efficiency of the proposed algorithm, a constrained Markowitz model with transaction fee and short sales were taken as a test example. On the basis of the numerical results, we can conclude that the proposed method can provide the more flexible and accurate results than those obtained by original BFO and PSO.