Inferencing and synthesizing information from multiple documents using text summarization and question answering models

使用文本摘要和问答模型从多个文档中推断和合成信息

• 批准号: 228139-2011
• 负责人: Chali, Yllias
• 金额: $ 1.46万
• 依托单位: University of Lethbridge
• 依托单位国家: 加拿大
• 项目类别: Discovery Grants Program - Individual
• 财政年份: 2015
• 资助国家: 加拿大
• 起止时间: 2015-01-01 至 2016-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=570357
• 关键词: Inferencing; synthesizing; information; multiple; documents

The term "Google" has become a verb for most of us. Search engines, however, have certain limitations. For example ask it for the impact of the current global financial crisis in different parts of the world, and you can expect to sift through thousands of results for the answer. This motivates the research in complex question answering where the purpose is to create summaries of large volumes of information as answers to complex questions, rather than simply offering a listing of sources. Unlike simple questions, complex questions cannot be answered easily as they often require inferencing and synthesizing information from multiple documents. Inferencing and synthesizing information from multiple documents require extracting, organizing, and inter-relating the pieces of information contained in a set of relevant documents, in order to obtain a comprehensive, non redundant report that satisfies the information need. Answering complex questions can be seen as a topic-oriented, informative multi-document summarization where the goal is to produce a single text as a compressed version of a set of documents with a minimum loss of relevant information. This research proposal deals with ways in which text summarization and question answering technologies can be combined to form truly useful information delivery tools. More precisely, the problems that we are planning to investigate are: developing text summarization and question answering models that answer complex questions, building decomposition models that decompose complex questions into simple questions easy to answer, constructing models that generate questions automatically, capturing users' needs and tailoring the answers as summaries to user needs, and designing intelligent mechanisms to process natural language (i.e., inference, synthesize and summarize information) as well as mechanisms for coherent and tailored information delivery. I am planning to develop models, technologies and tools to deliver to users information that is relevant and appropriate for the users receiving it, information that they can easily assimilate to enable them to perform their tasks, usually enabling better and more efficient decision-making.

“Google”这个词对我们大多数人来说已经成为一个动词。然而，搜索引擎有一定的局限性。例如，问它当前全球金融危机对世界不同地区的影响，你可以期待从数千个结果中筛选答案。这激发了复杂问题回答的研究，其目的是创建大量信息的摘要作为复杂问题的答案，而不是简单地提供来源列表。与简单的问题不同，复杂的问题不容易回答，因为它们通常需要从多个文档中推断和综合信息。从多个文档中推断和合成信息需要提取、组织和相互关联包含在一组相关文档中的信息，以获得满足信息需求的全面、非冗余的报告。对复杂问题的概括可以被看作是一个面向主题的、信息丰富的多文档摘要，其目标是以最小的相关信息损失生成一个单一的文本作为一组文档的压缩版本。本研究建议涉及如何将文本摘要和问答技术结合起来，形成真正有用的信息传递工具。更确切地说，我们计划调查的问题是：开发回答复杂问题的文本摘要和问题回答模型、构建将复杂问题分解成易于回答的简单问题的分解模型、构建自动生成问题的模型、捕获用户需求并将答案定制为用户需求的摘要，以及设计智能机制来处理自然语言（即，推理、综合和总结信息）以及提供连贯和有针对性信息的机制。我正计划开发模型、技术和工具，以便向用户提供与接收用户相关和适当的信息，使用户能够容易地吸收这些信息，从而使他们能够执行任务，通常能够作出更好和更有效的决策。