Automatic Summary Generation for Scientific Data Charts
Automatic Summary Generation for Scientific Data Charts
复制标题
自动生成科学数据图表摘要
DOI：
--
复制
发表时间：
2016
期刊：
影响因子：
--
通讯作者：
C. Lee Giles
中科院分区：
文献类型：
--
作者：
Rabah A. Al;Sagnik Ray Choudhury;C. Lee Giles
网络中的科学图表，无论是作为图像还是嵌入在数字文档中，都包含信息检索工具无法完全获得的有价值的信息。用于描述这些图表的信息通常是从图像元数据中提取的，而不是图形最初设计用来表达的信息。理解学术文献中的数字图表，并从其图形组件中推断出有用的文本信息的问题是本研究的重点。我们提出了一种方法来自动读取图表数据，特别是条形图，并为用户提供图表的文本摘要。所提出的方法遵循的知识发现方法，依赖于一个通用的图形表示的图表。这种表示方法是通过分析图表的原始数据值得出的，从中提取有用的特征。数据特征又被用来构造语义图。为了生成摘要，图表的语义图被映射到适当制作的原型，原型是基于模糊逻辑的构造。我们验证了我们的框架的有效性进行实验，从超过1000个PDF文档中提取的条形图。我们的初步结果表明，在某些假设下，83%的摘要提供了合理的描述条形图。
Scientific charts in the web, whether as images or embedded in digital documents, contain valuable information that is not fully available to information retrieval tools. The information used to describe these charts is typically extracted from the image metadata rather than the information the graphic was initially designed to express. The problem of understanding digital charts found in scholarly documents, and inferring useful textual information from their graphical components is the focus of this study. We present an approach to automatically read the chart data, specifically bar charts, and provide the user with a textual summary of the chart. The proposed method follows a knowledge discovery approach that relies on a versatile graph representation of the chart. This representation is derived from analyzing a chart’s original data values, from which useful features are extracted. The data features are in turn used to construct a semantic-graph. To generate a summary, the semantic-graph of the chart is mapped to appropriately crafted protoforms, which are constructs based on fuzzy logic. We verify the effectiveness of our framework by conducting experiments on bar charts extracted from over 1 , 000 PDF documents. Our preliminary re-sults show that, under certain assumptions, 83% of the produced summaries provide plausible descriptions of the bar charts.