Abstract | ||
---|---|---|
It is a common task to construct the reviewer group with diverse background between reviewers. This task is complicated considering the multiple criteria and sizable reviewers and groups. However, it has not been clearly addressed in the current studies. This paper investigates this problem and proposes a solution approach. In our study, this problem is firstly formulated as an integrated model that covers the situations of different group number and group size. Then, considering the computational difficulties of solving this model, the grouping genetic algorithm hybridizing the local neighborhood search heuristic is proposed. In the grouping genetic algorithm, the initialization, crossover and mutation are designed according to our problem's characteristics. Extensive numerical experiments show that the proposed algorithm is computationally efficient. Moreover, the application of the proposed algorithm on a case from NSFC also indicates its effectiveness for practical problems. |
Year | DOI | Venue |
---|---|---|
2011 | 10.1016/j.eswa.2010.08.029 | Expert Syst. Appl. |
Keywords | Field | DocType |
heuristics,reviewer group construction,grouping genetic algorithm,group size,hybrid genetic algorithm,common task,reviewer group,proposed algorithm,computational difficulty,hybrid grouping genetic algorithm,current study,or in government,integrated model,practical problem,different group number,reviewer group construction problem | Data mining,Heuristic,Multiple criteria,Crossover,Computer science,Group Number,Heuristics,Artificial intelligence,Initialization,Population-based incremental learning,Machine learning,Genetic algorithm | Journal |
Volume | Issue | ISSN |
38 | 3 | Expert Systems With Applications |
Citations | PageRank | References |
10 | 0.59 | 18 |
Authors | ||
4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Yuan Chen | 1 | 27 | 1.67 |
Zhi-Ping Fan | 2 | 1617 | 92.73 |
Jian Ma | 3 | 1662 | 103.00 |
Shuo Zeng | 4 | 37 | 2.94 |