Product reviews play an important role in rankings and impact customers' purchasing decisions on e-commerce sites. There exists a thriving ecosystem of incentivized reviews on e-commerce marketplaces -- reviews written by real customers in exchange for free products. While some e-commerce marketplaces themselves support incentivized review programs to solicit honest high-quality reviews, there are parallel underground services that sellers can use to commission fake positive reviews from real customers in exchange for free products. Despite anecdotal reports, our understanding of how these incentivized services operate and, crucially, how are they able to resist takedown efforts is lacking. In this paper, we conduct a quantitative and qualitative study of incentivized review services by infiltrating an underground incentivized review service geared towards Amazon.com. On a dataset of 1600 products seeking incentivized reviews, we first demonstrate the ineffectiveness of off-the-shelf fake review detection as well as Amazon's existing countermeasures. Through a survey of more than 70 participants of this underground incentivized review service, we uncover fairly sophisticated recruitment, execution, and reporting mechanisms they use to scale their operation while resisting takedown attempts.
翻译:产品审查在对电子商务地点的客户采购决定的排名和影响方面起着重要作用。对电子商务市场进行激励性审查的生态系统正在蓬勃发展,对电子商务市场进行激励性审查 -- -- 实际客户为交换自由产品而编写的审查。虽然一些电子商务市场本身支持激励性审查方案以征求诚实的高质量审查,但也有平行的地下服务,销售者可以用来委托真实客户进行虚假的积极审查以换取自由产品。尽管有传闻报道,但我们了解这些激励性服务如何运作,而且关键是,他们如何能够抵制收缩努力。在本文中,我们通过渗透以亚马逊为对象的地下激励性审查服务,对激励性审查服务进行定量和定性研究。关于寻求激励性审查的1600项产品的数据集,我们首先展示了现成的假审查探测以及亚马逊的现有对策的无效性。通过对这种地下激励性审查服务的70多名参与者的调查,我们发现了他们用来扩大业务规模的精密征聘、执行和报告机制,同时抵制尝试。