Multiple solutions of an exact algorithm for determination of all Kemeny rankings: preliminary experimental results; 2011 International Conference on Instrumentation, Measurement, Circuits and Systems (ICIMCS 2011)
| Parent link: | 2011 International Conference on Instrumentation, Measurement, Circuits and Systems (ICIMCS 2011).— 2011.— [P. 17-20] |
|---|---|
| Main Author: | |
| Corporate Author: | |
| Other Authors: | |
| Summary: | Title screen The classical problem of a single consensus rankingdetermination for m rankings of n alternatives has a potential ofwide applications in information technologies, and particularlyin measurement and instrumentation. The Kemeny rule is oneof deeply justified ways to solve the problem allowing to findsuch a linear order (Kemeny ranking) of alternatives that adistance (defined in terms of a number of pair-wisedisagreements between rankings) from it to the initial rankingsis minimal. But the approach can give considerably more thanone optimal solutions what can reduce its applicability. Oncomputational experiments outcomes, the paper demonstratesthat a set of Kemeny rankings cardinality can be extremelylarge even in small size cases (m = 4, n = 15…20) and,consequently, special efforts to build an appropriateconvoluting solution are needed. |
| Language: | English |
| Published: |
2011
|
| Subjects: | |
| Online Access: | https://doi.org/10.1115/1.859902.paper5 |
| Format: | Electronic Book Chapter |
| KOHA link: | https://koha.lib.tpu.ru/cgi-bin/koha/opac-detail.pl?biblionumber=641913 |
| Summary: | Title screen The classical problem of a single consensus rankingdetermination for m rankings of n alternatives has a potential ofwide applications in information technologies, and particularlyin measurement and instrumentation. The Kemeny rule is oneof deeply justified ways to solve the problem allowing to findsuch a linear order (Kemeny ranking) of alternatives that adistance (defined in terms of a number of pair-wisedisagreements between rankings) from it to the initial rankingsis minimal. But the approach can give considerably more thanone optimal solutions what can reduce its applicability. Oncomputational experiments outcomes, the paper demonstratesthat a set of Kemeny rankings cardinality can be extremelylarge even in small size cases (m = 4, n = 15…20) and,consequently, special efforts to build an appropriateconvoluting solution are needed. |
|---|---|
| DOI: | 10.1115/1.859902.paper5 |