Jia Liu

Jia Liu

PhD student

Academic and research departments

Department of Computer Science.


Research interests

My publications


Ben Niu, Jia Liu, Jing Liu, Chen Yang (2016). Brain Storm Optimization for Portfolio Optimization
View abstract View full publication
In this paper, Brain storm optimization (BSO) algorithm is employed to solve portfolio optimization (PO) problem with transaction fee and no short sales. In addition, simplified BSO (SBSO) and BSO in objective space (BSO-OS) are also utilized in the same model. The potential portfolio proportion is regarded as ideas that individual generated. In the experimental study, three cases with different risk aversion factors are considered. Simulation results demonstrate that both original BSO and modified BSOs obviously outperform PSO and BFO in PO problem. In particular, BSO-OS, which not only saves the computation time but also finds the optimal value, shows an extraordinary performance in all set of test data.
Hong Wang, Jia Liu, Wenjie Yi, Ben Niu, Jaejong Baek (2017). An Improved Brain Storm Optimization with Learning Strategy
View abstract View full publication
Brain Storm Optimization (BSO) algorithm is a brand-new and promising swarm intelligence algorithm by mimicking human being’s behavior of brainstorming. This paper presents an improved BSO, i.e., BSO with learning strategy (BSOLS). It utilizes a novel learning strategy whereby the first half individuals with better fitness values maintain their superiority by keeping away from the worst ones while other individuals with worse fitness values improve their performances by learning from the excellent ones. The improved algorithm is tested on 10 classical benchmark functions. Comparative experimental results illustrate that the proposed algorithm performs significantly better than the original BSO and standard particle swarm optimization algorithm.
Lijing Tan, Jia Liu, Wenjie Yi (2016). Group Discussion Mechanism Based Particle Swarm Optimization
View abstract View full publication
Inspired by the group discussion behavior of students in class, a new group topology is designed and incorporated into original particle swarm optimization (PSO). And thus, a novel modified PSO, called group discussion mechanism based particle swarm optimization (GDPSO), is proposed. Using a group discussion mechanism, GDPSO divides a swarm into several groups for local search, in which some smaller teams with a dynamic change topology are included. Particles with the best fitness value in each group will be selected to learn from each other for global search. To evaluate the performance of GDPSO, four benchmark functions are selected as test functions. In the simulation studies, the performance of GDPSO is compared with some variants of PSOs, including the standard PSO (SPSO), PSO-Ring and PSO-Square. The results confirm the effectiveness of GDPSO in some of the benchmarks.
Hong Wang, Lulu Zuo, Jia Liu, Chen Yang, Ya Li, Jaejong Baek (2017). A Comparison of Heuristic Algorithms for Bus Dispatch
View abstract View full publication
Bus dispatch (BD) system plays an essential role to ensure the efficiency of public transportation, which has been frequently addressed by the heuristic algorithms. In this paper, five well-exploited heuristic algorithms, i.e. Genetic algorithm (GA), Particle Swarm Optimization (PSO), Artificial Bee Colony algorithm (ABC), Bacterial Foraging Optimization (BFO) and Differential Evolution algorithm (DE), are employed and compared for solving the problem of BD. The comparison results indicate that DE is the best method in dealing with the problem of BD in terms of mean, minimum, and maximum, while BFO obtains the minor lower value of standard deviation and achieves the similar convergence speed in comparison to DE. The performance of PSO seems to outperform the remaining two algorithms (i.e. ABC and GA) in most cases. However, among five algorithms, GA achieves the worst results in terms of the weight estimated objective (i.e. number of departures and average waiting time).
Lianbo Ma, Xu Li, Jia Liu, Yang Gao (2016). A Novel Multi-objective Bionic Algorithm Based on Plant Root System Growth Mechanism
View abstract View full publication
This paper proposes and develops a novel multi-objective optimization scheme called MORSGO based on iterative adaptation of plant root growth behaviors. In MORSGO, the basic local and global search operators are designed deliberately based on auxin-regulated tropism of the natural root system, including branching, regrowing of different types of roots. The fast non-dominated sorting approach is employed to get priority of non-dominated solutions obtained during the search process, and the diversity over archived individuals is maintained by using dynamical crowded distance estimation strategy. Accordingly, Pareto-optimal solutions obtained by MORSGO have merits of better diversity and lower computation cost. The proposed MORSGO is evaluated on a set of bio-objective and tri-objective test functions taken from the ZDT benchmarks in terms of two commonly used metrics IGD and SPREAD, and it is compared with NSGA-II and MOEA/D. Test results verify the superiority and effectiveness of the proposed algorithm.
Ying Yu, Wenjie Yi, Yuanyue Feng, Jia Liu (2017). Understanding the Intention to Use Commercial Bike-sharing Systems: An Integration of TAM and TPB
View abstract View full publication
This article explores the effects of perceived green value, perceived green usefulness, perceived pleasure to use, subjective norms and perceived behavioral control on green loyalty to a public bike system. The mediators between perceived green value and green loyalty and a moderator of general attitude toward protecting the natural environment are also discussed. The aim of this research was to understand how to establish green loyalty via the other dimensions based on the sustainable modified technology acceptance model (modified TAM), the theory of planned behavior (TPB), and a moderator. The findings reveal that perceived pleasure to use and subjective norms have the strongest power to influence loyalty for both users and non-users. The implications of this finding are that fun in people's lives has a strong influence on sustainable continuous use of public bikes, and that subjective norms are more effective for non-users. In addition, environmental attitude has stronger moderating effects for non-users than for users on perceived green usefulness, perceived pleasure and subjective norms. Therefore, governmental policies should promote the attitude of protecting the natural environment, perceptions of pleasure, and subjective norms so as to increase green loyalty to public bike-sharing.
Ben Niu, Lijing Tan, Jing Liu, Jia Liu, Wenjie Yi, Hong Wang (2019). Cooperative bacterial foraging optimization method for multi-objective multi-echelon supply chain optimization problem
View abstract View full publication
Making integrated decisions has a positive effect on component companies' competitiveness and success in the multi-echelon supply chain network. This paper developed a more realistic integrated supply chain model, considering multi-objective, multi-product, multi-period, lead time et al. Minimum total cost and maximum customer satisfactory level are considered as optimization objectives simultaneously. As integrated supply chain optimization problems are typically NP-hard, a heuristic optimization algorithm, referred as Cooperative Multi-objective Bacterial Foraging Optimization, is proposed to solve supply chain optimization problems. In this research, the proposed cooperative evolutionary method is developed to accelerate the search process and enhance search accuracy. The feasibility control mechanism of solutions is implemented to ensure solutions are in the feasible domain. Moreover, an external storage mechanism is adopted to deal with multi-objective features and keep a historical record of the non-dominated individuals found by the algorithm, whereas the algorithm structure redesign method is used to reduce the complexity of the algorithm. The numerical experiments illustrate that the proposed algorithm can successfully optimize the supply chain and find better non-nominated solutions.