The paper presents research on automating the design of graphical presentations of relational information. The goal is to develop an application-independent presentation tool that automatically designs effective graphical presentations, such as bar charts, scatter plots, and connected graphs. Two key challenges are codifying graphic design criteria and generating a wide variety of designs. The approach treats graphical presentations as sentences of graphical languages, with expressiveness and effectiveness criteria for evaluating these languages. Expressiveness criteria determine if a language can express the desired information, while effectiveness criteria assess how well a language exploits the output medium and human visual system. A composition algebra is used to systematically generate a wide variety of designs by combining primitive graphical languages. The paper describes the development of a prototype presentation tool called APT, which uses artificial intelligence techniques to implement this framework. The research addresses the challenges of designing graphical presentations for relational information, including the expressiveness and effectiveness of graphical languages, and the generation of diverse designs using a composition algebra. The paper also discusses related work, the graphical presentation problem, and the effectiveness of different graphical techniques. The research contributes to the development of a systematic approach for automatically designing graphical presentations of relational information.The paper presents research on automating the design of graphical presentations of relational information. The goal is to develop an application-independent presentation tool that automatically designs effective graphical presentations, such as bar charts, scatter plots, and connected graphs. Two key challenges are codifying graphic design criteria and generating a wide variety of designs. The approach treats graphical presentations as sentences of graphical languages, with expressiveness and effectiveness criteria for evaluating these languages. Expressiveness criteria determine if a language can express the desired information, while effectiveness criteria assess how well a language exploits the output medium and human visual system. A composition algebra is used to systematically generate a wide variety of designs by combining primitive graphical languages. The paper describes the development of a prototype presentation tool called APT, which uses artificial intelligence techniques to implement this framework. The research addresses the challenges of designing graphical presentations for relational information, including the expressiveness and effectiveness of graphical languages, and the generation of diverse designs using a composition algebra. The paper also discusses related work, the graphical presentation problem, and the effectiveness of different graphical techniques. The research contributes to the development of a systematic approach for automatically designing graphical presentations of relational information.