Master the Art of Graph-Making

Tips for Effective Graphs in Research Papers

• Graphs are a great way to present numerical results and demonstrate trends in research papers.
• Avoid flooding the paper with statistics, even in graphical form, and focus on presenting interesting new information.
• Graphs should be simple, with a minimum of clutter and no more than a few plotted lines.
• Plotted lines of discrete data should have distinctive marks, such as circles or triangles, and consider using greys or colours to distinguish between lines.
• Greys are preferable to cross-hatching for filled-in areas in a figure, as the latter can create the optical illusion of shimmering and does not always print or photocopy well.
• Minimize unnecessary elements and remove all decoration, such as secondary ticks on the axes, legends, and large captions.
• Axes should be inconspicuous, and ink should be used for data, not dressing.
• Care is needed to ensure that graphs are effective at communicating results and supporting hypotheses.
• Graph-drawing tools should have valuable features such as placing several lines on one graph, a range of symbols for marking points, and the ability to create custom symbols of custom size.
• Optional connection of points with solid, dotted, or dashed lines, and optional omission of the point marks should also be available.
• The ability to place text at specified places in the graph, multiple font sizes and line thicknesses, and availability of greys and colours are also important features.
• Optional logarithmic or exponential scaling on both axes, axis editing, and the ability to apply simple functions or external programs to (x, y) values are also useful.

Tips for Effective Graphs in Research Papers

• Graphs are a great way to present numerical results and demonstrate trends in research papers.
• Avoid flooding the paper with statistics, even in graphical form, and focus on presenting interesting new information.
• Graphs should be simple, with a minimum of clutter and no more than a few plotted lines.
• Plotted lines of discrete data should have distinctive marks, such as circles or triangles, and consider using greys or colours to distinguish between lines.
• Greys are preferable to cross-hatching for filled-in areas in a figure, as the latter can create the optical illusion of shimmering and does not always print or photocopy well.
• Minimize unnecessary elements and remove all decoration, such as secondary ticks on the axes, legends, and large captions.
• Axes should be inconspicuous, and ink should be used for data, not dressing.
• Care is needed to ensure that graphs are effective at communicating results and supporting hypotheses.
• Graph-drawing tools should have valuable features such as placing several lines on one graph, a range of symbols for marking points, and the ability to create custom symbols of custom size.
• Optional connection of points with solid, dotted, or dashed lines, and optional omission of the point marks should also be available.
• The ability to place text at specified places in the graph, multiple font sizes and line thicknesses, and availability of greys and colours are also important features.
• Optional logarithmic or exponential scaling on both axes, axis editing, and the ability to apply simple functions or external programs to (x, y) values are also useful.

Tips for Designing Effective Diagrams in Computing Papers

• Diagrams serve various purposes in computing papers, such as illustrating processes, explaining data structures and algorithms, presenting relationships, visualizing data, and demonstrating interfaces.
• Some areas of computer science have well-defined conventions and standards for diagrams, such as entity-relationship models and automata.
• Diagrams should be created with a clear purpose in mind, either to show a structure, a process, a relationship, or a state, and should not attempt to combine these purposes inappropriately.
• Automatic tools for generating diagrams from data can produce a wide variety of representations, but manually created diagrams are still essential in many papers.
• The first step in designing a manual diagram is often to do initial sketches by hand on paper, but the final version should always be created with professional tools.
• A diagram should be well-proportioned, use space efficiently, be laid out well, and have elements arranged to look reasonable.
• A diagram should be sparse, with no clutter, and not too dark, and fine details can be clarified in the supporting text.
• Labels in diagrams should be meaningful, displayed horizontally, and have similar point size and font to the other text, with no more than two or three fonts and font sizes.
• Lines in diagrams should not be too heavy, with shades of grey used to distinguish between solids, and pictorial elements used consistently.
• Colour can be effective in diagrams if used sparingly, but should not be used if the paper will ultimately be printed in black-and-white.
• Arrows in diagrams should be used consistently, with dashed lines used to show arcs and solid lines used to point at features, and lines should not touch each other unless necessary.
• A well-designed diagram is a powerful tool that can help readers understand complex concepts and should be used effectively in computing papers.