Explaining Recommendations: Design and Evaluation-猫眼课题宝

DOI：

10.1007/978-1-4899-7637-6_10

发表时间：

2015

期刊：

影响因子：

通讯作者：

中科院分区：

其他

文献类型：

作者：

N. Tintarev;Judith Masthoff

近年来，人们对更多以用户为中心的推荐系统的评估指标越来越感兴趣，如[49]中提到的那些。还认识到，许多推荐系统像黑匣子一样运作，没有为推荐过程的工作提供透明度，也没有提供推荐本身之外的任何额外信息[35]。本章调查解释的作用，如图10.1所示。有时，人们错误地认为，解释总是应该证明为什么推荐这些项目是合理的。一个流行的解释定义是辩护的同义词。然而，解释也意味着“通过提供详细的描述来阐明”[牛津简明词典]。因此，解释可以是帮助用户更好地理解项目的质量以确定它是否与他们相关的项目描述。解释可以服务于多个目标，其中一个是透明度：旨在揭露建议背后的推理和数据。亚马逊上的一些解释就是这种情况，比如：“买了这件商品的顾客也买了……”。解释还可以达到其他目的，比如帮助激发用户的信任和忠诚度，提高满意度，让用户更快、更容易地找到他们想要的东西，并说服他们尝试或购买推荐的商品。通过这种方式，我们区分了不同的解释，例如解释推荐引擎的工作方式(透明度)，以及解释为什么用户可能想要或可能不想尝试一项(有效性)。一个有效的解释可能是

In recent years, there has been an increased interest in more user-centered evaluation metrics for recommender systems such as those mentioned in [49]. It has also been recognized that many recommender systems functioned as black boxes, providing no transparency into the working of the recommendation process, nor offering any additional information to accompany the recommendations beyond the recommendations themselves [35].This chapter investigates the role of explanations, such as the one depicted in Fig. 10.1. It is sometimes erroneously assumed that explanations should always justify why items have been recommended. A popular definition of explanation is synonymous with justification. However, to explain also means “to make clear by giving a detailed description”[Oxford concise dictionary]. So, an explanation can be an item description that helps the user to understand the qualities of the item well enough to decide whether it is relevant to them or not. Explanations can serve multiple aims, out which one is transparency: aiming to expose the reasoning and data behind a recommendation. This is the case with some of the explanations hosted on Amazon, such as:“Customers Who Bought This Item Also Bought...”. Explanations can also serve other aims such as helping to inspire user trust and loyalty, increase satisfaction, make it quicker and easier for users to find what they want, and persuade them to try or purchase a recommended item. In this way, we distinguish between different explanation such as eg explaining the way the recommendation engine works (transparency), and explaining why the user may or may not want to try an item (effectiveness). An effective explanation may be

Explaining Recommendations: Design and Evaluation

文献摘要

被引文献

相似文献