Three naive Bayes approaches for discrimination-free classification
Three naive Bayes approaches for discrimination-free classification
复制标题
DOI：
10.1007/s10618-010-0190-x
复制
发表时间：
2010-09-01
影响因子：
4.8
通讯作者：
Verwer, Sicco
中科院分区：
文献类型：
--
作者：
Calders, Toon;Verwer, Sicco
在本文中，我们研究了如何修改朴素贝叶斯分类器，以执行分类，被限制为相对于一个给定的敏感属性是独立的。当导致数据集中的标签的决策过程有偏见时，这种独立性限制自然发生;例如，因为性别或种族歧视。这一背景是由于在许多情况下，现行法律不允许部分基于歧视的决定。机器学习技术的天真应用将导致公司面临巨额罚款。我们提出了三种方法来使朴素贝叶斯分类器无歧视：（i）修改决策为正的概率，（ii）为每个敏感属性值训练一个模型并平衡它们，以及（iii）向贝叶斯模型添加一个潜在变量，表示无偏标签并使用期望最大化优化模型参数。我们提出了三种方法在人工和现实生活中的数据实验。
In this paper, we investigate how to modify the naive Bayes classifier in order to perform classification that is restricted to be independent with respect to a given sensitive attribute. Such independency restrictions occur naturally when the decision process leading to the labels in the data-set was biased; e.g., due to gender or racial discrimination. This setting is motivated by many cases in which there exist laws that disallow a decision that is partly based on discrimination. Naive application of machine learning techniques would result in huge fines for companies. We present three approaches for making the naive Bayes classifier discrimination-free: (i) modifying the probability of the decision being positive, (ii) training one model for every sensitive attribute value and balancing them, and (iii) adding a latent variable to the Bayesian model that represents the unbiased label and optimizing the model parameters for likelihood using expectation maximization. We present experiments for the three approaches on both artificial and real-life data.