Generation of texts for information graphics
Generation of texts for information graphics
复制标题
生成信息图形的文本
DOI：
--
复制
发表时间：
1999
期刊：
影响因子：
--
通讯作者：
G. Lapalme
中科院分区：
文献类型：
--
作者：
M. Corio;G. Lapalme
我们描述了一个文本生成系统，用于生成短文本和标题，以配合根据作者的意图生成的任何信息图形。SelTex使用的规则是从400多个文本摘录的语料库研究中提取出来的。这项集体研究表明，文字和图形在作者向读者传递信息方面起着互补的作用。然后，我们对与图形相关的文本的自动生成进行了一些观察，其中许多是在seltex中实现的。在过去的几年里，我们研究了postgraph系统背景下统计数据图形的自动生成[5,6]。postgraph以表格形式提供数据，就像在电子表格中一样；另外，输入是对表列中值的类型的声明。然后，用户指示要在图形中传达的意图(例如：比较两个变量或显示一组变量的演变)，系统用相应的PostScript图形文件在L ATEX中生成报告。postgraph还为一些简单的文本模式生成附带的文本。虽然postgraph已经显示了图形生成的潜力，但它的文本基因处理能力远远低于其图形能力水平。因此，我们决定重新设计文本生成模块，以更好地捕捉文本和图形之间交互的微妙本质。在设计seltex（集成在postgraph中的新文本模块）之前，我们对400多个与图形相关的文本进行了语料库研究。初步研究结果发表在b[3]。在描述从语料库研究中推断出的文本生成规则之前，我们将简要回顾这些规则。我们将重点关注图表附带文本中表达的相关事实的选择标准。postgraph概述许多复杂的工具可以用来创建使用统计图表的演示文稿。然而，他们中的大多数专注于制作专业的图形，而不是试图帮助用户组织演示文稿。为了在这方面提供帮助，我们构建了postgraph，它可以从一组作者的意图中生成一个集成图形和文本的报告。在图1中，作者的意图对信息的呈现方式有重要影响。事实上，根据作者想要表达的内容，该图以两种不同的方式展示了相同的数据。示例的一部分展示了如何表示数据的演变，另一部分展示了如何比较数据中的各种元素。作者的意图可以根据结构差异和内容差异这两个基本标准进行分类。图1说明了进化和比较之间的结构差异。我们指的是1971 1972 1973 1974 1975 1976 6 12 18
We describeSelTex a text generation system for producing short texts and captions to accomp any information graphics that are generated according to the writer’s intentio ns.SelTex uses rules that were extracted from a corpus study of more than 400 text excerpts. This corp us study shows that text and graphics play complementary roles in transmitting information from th e writer to the reader. We then derive some observations for the automatic generation of texts associat ed with graphics many of which were implemented inSelTex. For the past few years, we have studied the automatic generat ion of graphics from statistical data in the context of thePostGraphe system [5, 6].PostGraphe is given data in tabular form as might be found in a spreadsheet; also input is a declaration of the types of val ues in the columns of the table. The user then indicates the intentions to be conveyed in the graphics (e.g . compare two variables or show the evolution of a set of variables) and the system generates a report in L ATEX with the appropriate PostScript graphic files. PostGraphe also generates an accompanying text for only a few simple tex t schemas. Although PostGraphe has shown the potential of graphic generation, its text gene ration capabilities are well below its level of graphic competence. So we decided to redesign th e text generation module to better capture the subtle nature of the interaction between text and graphics. Before designingSelTex, the new text module to be integrated inPostGraphe, we did a corpus study of more than 400 texts associated with g raphics. The preliminary findings were presented in [3]. They will be r eviewed here briefly before describing the text generation rules that were deduced from this corpus stu dy. We will focus on the selection criteria for relevant facts that are expressed in the text that accompani es graphic. 1 Overview ofPostGraphe Many sophisticated tools can be used to create presentation s usi g statistical graphs. However, most of them focus on producing professional-looking graphics without trying to help the user to organize the presentation. To help in this aspect, we have built PostGraphe which generates a report integrating graphics and text from a set of writer’s intentions. In figure 1, the writer’s intentions have an important effect on the way information is presented. Indeed, the figure shows the same data presented in two different ways according to what the writer wants to convey. One part of the example shows how to present the evolution of t he data and the other part shows how to compare its various elements. The writer’s intentions can be classified according to two ba sic criteria: structural differences and contents differences. Figure 1 illustrates the structural dif ference between evolution and comparison. We refer 1971 1972 1973 1974 1975 1976 0 6 12 18