标题: 基于情境感知与社群推荐的行动广告机制
A Contextual based Social Referral Mobile Advertising Mechanism
作者: 潘强
Pan, Qiang
李永铭
Li, Yung-Ming
资讯管理研究所
关键字: 行动广告;情景感知;社群推荐;社群过滤;推荐系统;Mobile advertising;Context-aware;Social referral;Social filtering;Recommender systems
公开日期: 2012
摘要: 随着智慧型手机和平板电脑等智慧型行动装置普及和流行,他们已经成为我们生活中不可或缺的部份。行动广告逐渐为厂商和广告商带来巨大的商机。许多成功的公司,像苹果公司和谷歌公司都开始在其服务中设计行动广告推荐的机制以获取利润。然而行动广告也面临着巨大的挑战,的在动态的行动环境中,用户需求变化较快,而现有的广告推荐机制又不能够精确定位用户需求,造成了现在行动广告有效性和效益偏低,从一些指标中我们瞭解到,现在行动广告每千次展示成本(eCPM)的有效性过低,而且其平均每月每用户收入(ARPU)也非常低。所以在本次研究中,为了解决以上的问题,我们提出来一种基于情境资讯和社群行为参考的的行动广告推荐机制。在该机制中,我们采用了情境感知技术和社群行为参考的概念,以提高行动广告的效率和效益。通过我们的系统,我们试图通过在不断变化的环境中为用户找到最合适的广告,并且希望通过朋友的影响来有效提高用户对于广告的接受度。
Mobile devices, such as smartphones and Tablets, have been so popular that they are to be indispensable to daily life, which brings a bright future for mobile advertising. Many great companies have adopted mobile advertising in their services, such as Apple and Google. But the low effective cost per thousand impressions (eCPMs) and the low the average revenue per user (ARPU), caused by inaccurate targeting in dynamic environment and advertising avoidance, are big challenges facing with the mobile advertisers now. In this paper, we propose a novel mobile advertising recommender system by incorporating the concepts of “context-fitness” and “social referral”. By using our system, we could identify most suitable ads for targeted users in the changing environment and improve the ads effectiveness by taking friends’ influence into consideration.
URI: http://140.113.39.130/cdrfb3/record/nctu/#GT070053434
http://hdl.handle.net/11536/71648
显示于类别:Thesis