Voting procedures from the perspective of theory of neural networks

Ibragim Esenovich Suleimenov, Sergey Panchenko, Oleg Gabrielyan, Ivan Pak

2016

Full text Open access: Gold Metadata (CSL-JSON)
DOI 10.1515/eng-2016-0048
Data confidence 70 medium
How this score is calculated

It is built from checkable properties of the record itself: whether the article has a DOI, how many independent sources confirmed its data, whether the required fields are filled in, and how recent the latest confirmation is. Disagreement between sources and a missing identifier lower it. A value of 80 or above means the record was published without further checking; 60 to 79 means it is published at reduced confidence, with a manual check advisable. A record verified by an editor by hand is never pushed below 90 by a recalculation. This scores our data about the article, not the article itself.

Abstract

Abstract It is shown that voting procedure in any authority can be treated as Hopfield neural network analogue. It was revealed that weight coefficients of neural network which has discrete outputs −1 and 1 can be replaced by coefficients of a discrete set (−1, 0, 1). This gives us the opportunity to qualitatively analyze the voting procedure on the basis of limited data about mutual influence of members. It also proves that result of voting procedure is actually taken by network formed by voting members.