[arxiv:1905.06089] 论文 Electre-Score 阅读报告

阅读量：6076 次

发布时间：2019-06-20

本文共 3716 字，大约阅读时间需要 12 分钟。

我有勇气挑战风险, 可惜积累不太够, 打算坚持每天写一个阅读报告 [发抖][怀疑]

时常想起一个学长, 他的研究总结总是非常深刻, 以其为榜样, 勉励自己.

P.S. 又得知那个学长的一些情况, 更为疚心, 想到看过的一段文摘:

一间小屋, 一点念想... 时间流逝, 人过中年, 立于帝都之上, 风卷云涌, 俯瞰万千... 晚年回乡久居, 翻开照片, 些许感慨, 虽无华白墙柱, 无广袤庭园. 但一砖一瓦, 一草一木, 皆是那么熟悉亲切. 望向窗外, 似乎见到故友四海归来的身影, 仿佛回到快乐无邪的童年. --小田先生传

Electre-Score: A first outranking based method for scoring actions

In this paper, we present (to the best of our knowledge) the first outranking method to assign a score to each alternative. It is a method of the Electre family, and we will call it Electre-Score. Contrarily to the Multi-Attribute Value Theory methods, Electre-Score does not construct a value function for each criterion and then proceeds to the aggregation into a single value. It rather makes use of the outranking relations to make a comparison with reference sets of actions (to which a score is assigned) and proposes a score range to each alternative, instead of a single value. This is a more robust way of proceeding given the fragility of a single score. The sets of limiting profiles are defined with the help of Electre Tri-nB and the reference scores are assigned to them through the application of the deck of cards technique. The fact of being able to use outranking relations makes it also possible to take into account the imperfect knowledge of data and avoids systematic compensatory effects. Some fundamental theoretical results guaranteeing the consistency of the method and an illustrative example are also provided in this paper.

Multiple Criteria analysis; Elect Scoring methods; Outranking relations; Decision support systems

Finus

多准则决策辅助 (Multiple Criteria Decision Aiding, MCDA)

three major families of methods:

基于评分的方法(scoring-based): 为每个动作或备选项分配一个分数

基于级别的方法(outranking-based): 利用有序对其元素之间的级别关系 {
Electre method [4]}

基于规则的方法(rule-based): 遵循一套决策规则

In a decision aiding context, knowing the preferences of the Decision Maker (DM) and determining weights of criteria are very hard questions.

一种基于级别的方法

其优点在于:

能定量又能定性地处理评价指标;

指标可以是非常不同的类型: 米, 噪度, 时延, 代价, 反馈... ;

补偿效应? 与系统性不相关: 由于使用一致性指数和(存在避免某些补偿的)否决阈值;

考虑到数据的不完全性(不定确, 不精确, 不准确) 和评价制订的任意性;

借助级别论, 能够说明优胜劣汰的理由.

其痛点在于:

可能出现不传递性(intransitivity)之象 (且出现可能比较高频);

可能出现与非相关备选项间的依赖之象;

如果所有的标准均有定量, 且数据不存在不完全性, 且评价制订有理可循, 且决策者允许系统的补偿, 则多属性价值理论(Multi-Attribute Value Theory, MAVT) 的方法更好;

如果为每一个动作分配一个分数是必要的, 基本所有方法都可行.

痛点3在实际生活中不太可能出现, 那么是否有可能建立一种基于评分的选举方法? (Motivation)

一种基于评分的选举方法

Step:

构建了多个参考操作集。我们使用的限制配置文件, 如在Electre-TrinB(ferna ' ndez等人, 2017) 作为我们的参考集。这类参考行动必须满足一些重要的可分离条件。请注意, 无论是先验地定义一组参考操作的程序, 它都必须满足一定数量的条件, 这些条件应尽可能弱, 以便该方法能够在一个广泛的一般性中得到有效的应用。例。首先, 我们考虑的是必须在以下条件下尽可能放松的限制条件, 但前提是这些限制性较小的条件能够保持方法的有效性.