Rank aggregation methods for the Web
Proceedings of the 10th international conference on World Wide Web
Condorcet fusion for improved retrieval
Proceedings of the eleventh international conference on Information and knowledge management
SIAM Journal on Discrete Mathematics
Proceedings of the 16th international conference on World Wide Web
Particle Swarm Optimization and Intelligence: Advances and Applications
Particle Swarm Optimization and Intelligence: Advances and Applications
Identifying Differentially Expressed Genes via Weighted Rank Aggregation
ICDM '11 Proceedings of the 2011 IEEE 11th International Conference on Data Mining
| Hi-index | 0.00 |
In microarray analysis, gene relevance is measured according to the level of differential expression of the gene and this level of differential expression decides the rank of that gene. Ranking the disease-related genes has major impact on disease classification and prediction. Several statistical tests exist in literature for ranking the genes with respect to their abilities to distinguish different classes of samples. However, none of these methods can perform well for all kinds of datasets. Therefore the scientists have applied rank aggregation techniques to obtain a consensus ranking of the genes from the rankings produced by different methods. In this article, we have posed the problem of rank aggregation as an optimization problem for minimizing the average distance of the output ranking from the input rankings. In this regard, a novel particle swarm optimization (PSO)-based algorithm is proposed for minimizing this distance. The well-known Kendall's Tau distance measure has been used for this purpose. The performance of the proposed algorithm is compared with that of different existing rank aggregation methods and evaluated using one artificial and two real-life datasets.