在本帖中，我们来讨论一下如何制作有吸引力的、使文章看起来专业的图片。形式规范的文章会更容易得到读者的重视，如果您制作图片时细心认真，读者会自然而然地认为您在做研究的时候也是严谨的。 制图时，请确保以下几点： 1. 刻度单位和标识清晰易读。 2. 避免不一致的字体以及生僻的刻度。 3. 尽量使用醒目、对比度强的颜色。 4. 如果图片有吸引力，读者就不会忽略图片所传达的信息。 在本帖中，我将给出图片示例。如果您想查看我如何用 Excel 修改图片，请登录我们的博客： 浏览演示稿，其中有截图和详细的解释。 In this post, I will discuss making figures worthy of your research. What do I mean by this? You have spent months building your experiment/designing your study/programming your simulation, collecting your data, and arguing with other researchers about what it means (This last item was a joke). Now you have some data you want to present to the broader community. What should you do? In this post, I will show example graphs. If you would like more details on how I made the changes to these plots using Microsoft Excel, please view the blog post titled: " Making Scienticic Plots: A Microsoft Excel Tutorial ," which has screenshots and detailed explanations. In addition, I would recommend using a full-featured plotting program such as Origin, Igor Pro, KalideGraph, or Matlab. As an author, I used Igor Pro followed by a second round of improvements in Adobe Illustrator. I often spend more time making figures than I do writing manuscript text, because they are more important. However, it is possible to make high quality plots using only Excel. First and foremost, DO NOT under any circumstances use default plot settings, this is particularly true for Excel. Take extra time and care to make your plots readable and attractive. This means you will need to adjust the settings for every graph. Figure 1 . Typical default settings plot. It can be seen from the time that the author (me) took to make this plot that the data included in it is unimportant and should be ignored by the readers. In Fig. 1, I have made only minor changes to the default settings. I changed the spacing of the tics such that the numbers on the axes don’t overlap, and I moved the vertical axis to the left side of the plot such that its associated numbers are not on top of the data. This plot has some clear deficiencies. 1. The axes have no labels. Don’t rely on the caption (or the main text) convey all the details of a plot to the reader. Try to make your plots easy to interpret. 2. The plot does not use space well. The data is concentrated in the center of the plot, and surrounded by areas that are unused. Space is limited in journal submissions (if it is not limited than it is at least expensive). Therefore, you should not waste space. 3. The used in the scatter plot are too large. This makes it difficult to see trends in the data. 4. The figure key, which simply says “Series1”, does not provide useful information to the reader. In this case there is only one data set included, which means that the key is unnecessary. Figure 2 . Minimum effort plot. The author took the time to make a plot that others could read, but this plot still looks unprofessional. The author used scientific notation for small numbers which are all on the same order. Why does the time axis go from -700 ns to -580 ns? Should the intensity floor go to zero? In Fig. 2, I have made these changes. Now the graph can be interpreted by another scientist. This graph might meet the absolute minimum standards expected by journal editors, reviewers, and the broader scientific community. However, it still contains some poor choices, which make it look unprofessional and would reflect poorly on the author. 1. The axes have very strange ranges. These will distract readers that will think you are trying to convey something you are not. For this data, it would be better to normalize the data and set the intensity floor at zero. Be careful when making decisions like this. Setting the intensity floor to zero makes a significant quantitative change to the data. DO NOT alter data for the purposes of improving results! This is unethical and reflects poorly on the entire scientific community. This is far more important than having a clear plot. Be honest first, and then clear second. For the other axis, use a time interval that will not distract the reader. For example or would seem more natural. AVOID just using the ranges given to you by your instrument, because they often will not convey the intended meaning of the data. 2. The bottom axis uses scientific notation for numbers less than 1,000, which are all within one order of magnitude. This make the scale hard to read, and consequently hard to interpret. 3. The text is too small. In a print publication, figures are often compressed. If your text is too small in comparison to the plot it will become illegible when the figure is shrunk for publication. 4. The font choice is inconsistent. Notice that axis labels use two different fonts. This looks careless. In addition, the font choice is not the best. For most of text of the figure, the author used the font “Times New Roman,” which is considered a readable font. These fonts are best used on long sections of printed text. For figures and presentations, it is best to use a legible font such as “Arial.” Legible fonts are easy to read even when the text is small. 5. The figure is unattractive. This figure will not attract readers. Making plots using vibrant high-contrast (easy to see) colors will improve readership (citations!) Figure 3 . Professional looking plot. The author took time planning and making this figure. Readers will interpret the data the way the author intends. In addition, the use of bright red will attract readers to the figure, which may increase citations. This figure could be published in a top-tier international scientific journal. Figure 3 is an example of a plot that I as an author or reviewer would be satisfied with. 1. The scales and labels are easy to read. 2. Distractions like inconsistent fonts and strange scale ranges are eliminated. 3. Vibrant high-contrast (easy to see) colors are used. 4. The figure is attractive. Therefore readers will be interested in the meaning of the figure rather than ignoring it. In addition, I made some personal choices in for this figure. I like to close my plots in a box to separate them from the text of the paper. I also like to have my axis tics on the inside of the axis. I chose red rather than blue or green. Experiment with options like these. Ask yourself, “What looks attractive?” Taking time to think about these details will improve your plots even more. Your plots are more important than the main text of the paper! Give them the time they deserve. All the best, Daniel Broaddus, PhD Physical Sciences Editor, Edanz Group China www.liwenbianji.cn
4th International Symposium on Recurrence Plots We would like to announce the 4th International Symposium on Recurrence Plots which will be held in Hong Kong, December 5-7 2011. The objective of this recurrence symposium is to encourage the exchange of knowledge among scientists working in the disciplines of time and spatial series analyses. Recurrence plots and recurrence quantifications are general methods for visualising and analysing both linear and nonlinear time series data. We continue to witness many new technical developments related to recurrence plots. Some of these include: a framework to treat recurrence plots as a network from which one can obtain network-related statistics; inferring directional couplings; identifying deterministic chaos; obtaining confidence intervals for recurrence quantification analysis; defining recurrence plots for point processes. In addition, applications of recurrence plots are increasing in areas such as mathematics, neuroscience, physiology, psychology, weather and climate, financial systems, and linguistics. This symposium will provide a unique forum to help combine the recent theoretical developments in recurrence science with applications from various fields. We welcome both theoretical and/or applied contributions that use recurrence plots, recurrence quantifications and related methodologies. Participants are invited to present their related work on recurrence plots or recurrence based techniques in the form of contributed talks and posters. We strongly encourage the attendance of junior scientists (PhD and post-doc level). For more information and registration please visit: http://symposium.recurrence-plot.tk/ The deadline for registration and abstract submission is July 31st, 2011. If you have any further questions, please feel free to contact us. We encourage you advertise this meeting in your group, in particular among your PhD students or junior researchers. Organisational and scientific committee: Michael Small (Department of Electronic and Information Engineering, Hong Kong Polytechnic University) Norbert Marwan (Research Domain Transdisciplinary Concepts and Methods, Potsdam Institute for Climate Impact Research) Yoshito Hirata (Aihara Laboratory, Institute of Industrial Science, The University of Tokyo) Charles L. Webber, Jr. (Department of Physiology, Loyola University Chicago) Xu Xiaoke (Department of Electronic and Information Engineering, Hong Kong Polytechnic University) Zhang Jie (Department of Electronic and Information Engineering, Hong Kong Polytechnic University) Reply ForwardReply by chat to xiaoke Beijing Chinese Center - Learn Intensive Chinese in 10 daysAdEffective Method - Enroll Here. www.NewConceptMandarin.com
标签: Plots
