Attribute weighting is a task of paramount relevance in multi-attribute decision-making (MADM). Over the years, different approaches have been developed to face this problem. Despite the effort of the community, there is a lack of consensus on which method is the most suitable one for a given problem instance. This paper is the second part of a two-part survey on attribute weighting methods in MADM scenarios. The first part introduced a categorization in five classes while focusing on explicit weighting methods. The current paper addresses implicit and hybrid approaches. A total of 20 methods are analyzed in order to identify their strengths and limitations. Toward the end, we discuss possible alternatives to address the detected drawbacks, thus paving the road for further research directions. The implicit weighting with additional information category resulted in the most coherent approach to give effective solutions. Consequently, we encourage the development of future methods with additional preference information.

属性加权是多属性决策（MADM）中至关重要的任务。多年来，已经开发出不同的方法来面对这个问题。尽管社区做出了努力，但对于哪种方法最适合给定问题实例，仍缺乏共识。本文是针对MADM场景中的属性加权方法的两部分调查的第二部分。第一部分介绍了五个类别的分类，同时重点介绍了显式加权方法。本论文探讨了隐式和混合方法。为了确定其优势和局限性，共分析了20种方法。最后，我们讨论了解决已发现缺陷的可能替代方法，从而为进一步的研究方向铺平了道路。具有附加信息类别的隐式加权导致采用最一致的方法来给出有效的解决方案。因此，我们鼓励开发具有更多偏好信息的未来方法。