Cross-domain knowledge transfer for data-driven decision making

用于数据驱动决策的跨领域知识转移

• 批准号: DP170101632
• 负责人: A/Prof Guangquan Zhang
• 金额: $ 26.33万
• 依托单位: University of Technology Sydney
• 依托单位国家: 澳大利亚
• 项目类别: Discovery Projects
• 财政年份: 2017
• 资助国家: 澳大利亚
• 起止时间: 2017-01-01 至 2020-05-31
• 项目状态: 已结题
• 来源: https://dataportal.arc.gov.au/RGS/Web/Grants/DP170101632
• 关键词: Cross; domain; knowledge; transfer; data

This project aims to develop a set of cross-domain knowledge transfer methodologies to support Data-Driven Decision-Making (D3M) systems. D3M is essential in business, particularly for ever-changing environments in today’s big data era, but D3Ms for solving new problems may face in-domain data insufficiency. The challenge is to effectively transfer knowledge from multiple heterogeneous source domains. The outcomes are expected to transfer implicit and explicit knowledge, handle discrete and continuous outputs, and support business decision-making, which should advance the discipline of transfer learning and data-driven DSS in dynamically changing environments.

该项目旨在开发一套跨领域的知识转移方法，以支持数据驱动的决策(D3M)系统。D3M在商业中是必不可少的，特别是在当今大数据时代不断变化的环境中，但解决新问题的D3M可能面临域内数据不足。挑战是有效地从多个不同源域转移知识。预期成果将传递隐含和明确的知识，处理离散和连续的产出，并支持业务决策，这将促进在动态变化的环境中传递学习和数据驱动的决策支持的纪律。