International Journal of Advanced Networking and Applications

Open Access Open Access  Restricted Access Subscription Access

Multiple-Criteria Decision Analysis:A Novel Rank Aggregation Method


Affiliations
1 Department of Computer Engineering, Istanbul University, 34320, Avcilar, Istanbul, Turkey
 

Ranking among several objects is a very crucial operation for different applications to find a vote value for each object against the others. Multiple metrics can be combined to get a single vote value of an object. There are many studies in the literature that convert the ranking problem into a graph structure to solve it with a discrete mathematical process. Generally, these studies define multiple metrics as matrix forms and then relate them with the computations of eigenvectors to find the best ranked object. However, due to the dynamic nature of the metric values, ranking approaches should be fast and less complex. In this study a different approach for the ranking process with multiple metrics is proposed. This approach is fast and easy to implement. In order to test the approach, a network scenario is designed with computer programs. The experimental results show that this method outperforms a common conventional method in terms of various metric values, namely transmission time, packet loss rate, jitter, availability, and throughput. As a consequence, the proposed method gives the average value of each individual metric as more advantageous and without rescaling the numerical values.

Keywords

Decision Theory, Multi-Criteria Decision Analysis, Multi-objective Decision, Rank Aggregation, Rank Centrality.
User
Notifications
Font Size

  • V. Conitzer, Making decisions based on the preferences of multiple agents, Communications of the ACM, 53, 2010, 84-94.

  • L.H. Lim, X. Jiang, Y. Yao, and Y. Ye, Graph helmholtzian and rank learning. http://www.stat.uchicago.edu/~lekheng/work/nips.pd f, 2008, Last accessed: March 2018.

  • A. Altman and M. Tennenholtz, Ranking systems: The PageRank axioms, Proc. 6th ACM Conference on Electronic Commerce, Vancouver, British Columbia, Canada, 2005, 1-8.

  • C. Dwork, R. Kumar, M. Naor, and D. Sivakumar, Rank aggregation methods for the web, Proc. Tenth Int. World Wide Web Conference (WWW10), Hong Kong, 2001, 613-622.

  • Z. Gormez, E. Gumus, A. Sertbas, and O. Kursun, Comparison of aggregators for multi-objective SNP selection, Proc. 35th Annual International Conference of the IEEE EMBS, Osaka, Japan, 2013, 3062-3065.

  • F.E.M. Arasi, A. Anand, and S. Kumar, QoS based ranking for composite web services, International Journal of Science, Engineering and Technology Research (IJSETR), 3, 2014, 1041-1046.

  • C. Domshlak, A. Gal, and H. Roitman, Rank aggregation for automatic schema matching, IEEE Transactions on Knowledge and Data Engineering, 19, 2007, 538-553.

  • D. Sculley, Rank aggregation for similar items. http://www.eecs.tufts.edu/~dsculley/papers/mergeSim ilarRank.pdf, 2006, Last accessed: March 2018.

  • L.P. Dinu and F. Manea, An efficient approach for the rank aggregation problem, Theoretical Computer Science, 359, 2006, 455-461.

  • A. Yazdani, L. Dueñas-Osorio, and Q. Li, A scoring mechanism for the rank aggregation of network robustness, Commun Nonlinear Sci Numer Simulat, 18, 2013, 2722–2732.

  • S. Adalı, B. Hill, and M. Magdon-Ismail, Information vs. robustness in rank aggregation: Models, algorithms and a statistical framework for evaluation, http://www.cs.rpi.edu/~magdon/ps/journal/ahm_jdim .pdf, Last accessed: March 2018.

  • H. Shachnai, L. Zhang, and T. Matsui, On rank aggregation of multiple orderings in network design, http://www.cs.technion.ac.il/~hadas/PUB/agg.pdf, Last accessed: March 2018.

  • R. Kumar, Rank aggregation. Lecture Notes, University of Rome, Italy, 2008.

  • S. Negahban, S. Oh, and D. Shah, Rank centrality: Ranking from pair-wise comparisons, http://arxiv.org/abs/1209.1688v2, Cornell University Library. Presented in part at NIPS, 2012 in Lake Tahoe. Last accessed: March 2018.


Abstract Views: 172

PDF Views: 0




  • Multiple-Criteria Decision Analysis:A Novel Rank Aggregation Method

Abstract Views: 172  |  PDF Views: 0

Authors

Yiltas-Kaplan Derya
Department of Computer Engineering, Istanbul University, 34320, Avcilar, Istanbul, Turkey

Abstract


Ranking among several objects is a very crucial operation for different applications to find a vote value for each object against the others. Multiple metrics can be combined to get a single vote value of an object. There are many studies in the literature that convert the ranking problem into a graph structure to solve it with a discrete mathematical process. Generally, these studies define multiple metrics as matrix forms and then relate them with the computations of eigenvectors to find the best ranked object. However, due to the dynamic nature of the metric values, ranking approaches should be fast and less complex. In this study a different approach for the ranking process with multiple metrics is proposed. This approach is fast and easy to implement. In order to test the approach, a network scenario is designed with computer programs. The experimental results show that this method outperforms a common conventional method in terms of various metric values, namely transmission time, packet loss rate, jitter, availability, and throughput. As a consequence, the proposed method gives the average value of each individual metric as more advantageous and without rescaling the numerical values.

Keywords


Decision Theory, Multi-Criteria Decision Analysis, Multi-objective Decision, Rank Aggregation, Rank Centrality.

References