Recommender system empowered by contextual information

由上下文信息支持的推荐系统

• 批准号: 490782-2015
• 负责人: Bener, Ayse
• 金额: $ 4.95万
• 依托单位: Ryerson University
• 依托单位国家: 加拿大
• 项目类别: Collaborative Research and Development Grants
• 财政年份: 2018
• 资助国家: 加拿大
• 起止时间: 2018-01-01 至 2019-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=645139
• 关键词: Recommender; system; empowered; contextual; information

This proposal is a collaborative effort between IBM Watson Analytics and the Data Science Laboratory at Ryerson University to build a recommender system empowered by contextual information. The fundamental question that a recommender system aims to answer is: "what does the end-user really want?" If the end-user does not know the answer to this question, s/he would benefit from using a recommender system. Our industrial partner - IBM Watson Analytics - is one of the global leaders in the field of recommender systems. IBM Watson Analytics provides cognitive technology processing information by understanding natural language, generating hypothesis based on evidence and learning as it goes. The major deliverable of this proposal is a software prototype recommender system empowered by contextual information that is compatible with the IBM Watson Analytics platform. This project is solution oriented in the sense that its solves an important problem for the flagship artificial intelligence product of our industrial partner - IBM Watson Analytics. IBM Watson is a question answering computer system capable of answering questions posed in natural language. One of the challenges that comes with Watson is how to make the best recommendations based on a limited amount of data, knowing that better recommendations lead to higher user engagements. Up until now, Watson does not use or leverage contextual information. It is with this problem in mind that IBM and that Data Science Laboratory came up with the solution of designing a recommender system empowered by contextual information. The end product will be a software prototype recommender system empowered by contextual information that increases the level of engagement and acts as a catalyst in ultimately helping the end-user find what s/he really wants.

该提案是 IBM Watson Analytics 和瑞尔森大学数据科学实验室之间的合作成果，旨在构建一个由上下文信息支持的推荐系统。推荐系统旨在回答的基本问题是：“最终用户真正想要什么？”如果最终用户不知道这个问题的答案，他/她将受益于使用推荐系统。我们的行业合作伙伴 - IBM Watson Analytics - 是推荐系统领域的全球领导者之一。 IBM Watson Analytics 提供认知技术，通过理解自然语言、基于证据生成假设并进行学习来处理信息。该提案的主要成果是一个由与 IBM Watson Analytics 平台兼容的上下文信息支持的软件原型推荐系统。该项目是面向解决方案的，因为它解决了我们的工业合作伙伴 IBM Watson Analytics 的旗舰人工智能产品的一个重要问题。 IBM Watson 是一个问答计算机系统，能够回答以自然语言提出的问题。 Watson 面临的挑战之一是如何根据有限的数据提出最佳推荐，并且知道更好的推荐会带来更高的用户参与度。到目前为止，Watson 尚未使用或利用上下文信息。正是考虑到这个问题，IBM 和数据科学实验室提出了设计由上下文信息支持的推荐系统的解决方案。最终产品将是一个由上下文信息支持的软件原型推荐系统，该系统可以提高参与度并最终帮助最终用户找到他/她真正想要的东西。