graph representation ppt

• Insert a picture in PowerPoint Article
• Edit pictures Article
• Add SmartArt to a slide Article
• Put a background picture on your slides Article
• Add a background picture to slides Article
• Use charts and graphs in your presentation Article
• Insert icons in PowerPoint Article

Use charts and graphs in your presentation

You can make a chart in PowerPoint or Excel. If you have lots of data to chart, create your chart in Excel , and then copy it into your presentation . This is also the best way if your data changes regularly and you want your chart to always reflect the latest numbers. In that case, when you copy and paste the chart, keep it linked to the original Excel file .

To create a simple chart from scratch in PowerPoint, click Insert > Chart and pick the chart you want.

Click Insert > Chart .

Click the chart type and then double-click the chart you want.

Tip:  For help deciding which chart is best for your data, see Available chart types .

In the worksheet that appears, replace the placeholder data with your own information.

When you’ve finished, close the worksheet.

Create an org chart in PowerPoint

Create charts in Excel

Need more help?

Want more options.

Explore subscription benefits, browse training courses, learn how to secure your device, and more.

Microsoft 365 subscription benefits

Microsoft 365 training

Microsoft security

Accessibility center

Communities help you ask and answer questions, give feedback, and hear from experts with rich knowledge.

Ask the Microsoft Community

Microsoft Tech Community

Windows Insiders

Microsoft 365 Insiders

Was this information helpful?

Thank you for your feedback.

18 Types of Diagrams in PowerPoint: Which is the Right Chart Type for Your Presentation?

Presenting data is one of the most common content types in presentations. Speakers are often faced with the task of presenting their data in a way that directs the audience’s attention to the key messages.

Today, we will show you 18 chart types with examples of their usage. This way, you can find the right diagram for your presentation purposes.

Storytelling & Data Visualization

Speakers should focus on telling a story with data. Storytelling is one of the most effective means of connecting with the audience and capturing their attention. Why? Because stories generate emotions and allow you to better reach your audience.

Presenting raw data without proper preparation will inevitably lead to losing the audience’s interest . The audience will unconsciously begin to orient themselves in the presented data series and interpret it, which consumes a significant portion of their concentration.

The challenge is to integrate complex and dry numbers into the narrative in a way that the audience can follow the argumentation. The key to success lies in communicating through targeted data visualizations.

The most well-known and popular form of data visualization is the diagram . The use of diagrams in PowerPoint presentations is practical due to the convenient integration of PowerPoint with Excel functions .

However, it is important to always consider the message that the presented data is intended to convey and the type of data involved. Not every diagram is suitable for every dataset.

• Is it relative or absolute numbers?
• How many dimensions do I want to represent?
• Am I presenting compositions or developments?

These are just a few examples of the questions you should ask yourself before choosing a diagram for your presentation.

The 18 most important types of diagrams in PowerPoint

We have summarized the most well-known chart types, along with their advantages, applications, and limitations .

Now, let’s explore these diagram types and find the one that best suits your data and goals, allowing you to create a clear and compelling presentation.

1. Column Chart

The bar chart is the most commonly used and simplest type of diagram. By representing data through the varying heights of the bars, you can visually illustrate data and its differences.

The strengths of the bar chart type lie in depicting fluctuations over a period of time or comparing different subjects of investigation.

For example : Revenues of different departments per year.

Feel free to use professionally designed slide templates for bar charts from PresentationLoad!

2. Bar Chart

The bar chart is nothing more than a rotated version of the column chart . Like the column chart, the bar chart represents data and their differences through the distribution of bar sizes.

The major advantage of this chart type is that the horizontal orientation of the bars allows for the use of longer labels, such as survey questions.

Example: This chart type is excellent for representing rankings.

For tips on designing an appealing bar chart , you can refer to the article “ Bar Charts .”

3. Stacked Column Chart

The stacked chart (also known as a cumulative or stacked chart) is a chart type that can represent the individual components of a composite whole. This chart type is suitable when comparing the composition of something over different time periods or with a different composition.

Example: Composition of cost components over a period of time.

Feel free to use professionally designed slide templates for stacked charts from PresentationLoad!

4. Line Chart

The line chart is used for comparing and representing temporal trends . The overlapping lines can be directly compared, making it easy to visualize developments and trends .

Example: Stock prices.

Feel free to use professionally designed slide templates for line charts from PresentationLoad!

5. Area Chart

The area chart is a modified form of the line chart . In this chart, the area between two lines or between the line and the X-axis is filled with color.

This allows for highlighting the relative relationship between two quantities graphically. This type of representation is particularly useful for visualizing operational and strategic gaps.

Example: Gap analysis.

Feel free to use professionally designed slide templates for area charts from PresentationLoad!

6. Pie Chart

Pie and donut charts represent compositions of a whole as slices of a pie. The major strength of these charts is visualizing relative proportions.

Example: However, pie charts are not suitable for representing temporal sequences.

Feel free to use professionally designed slide templates for pie charts from PresentationLoad!

7. Combination Chart

Combination charts are a combination of two different chart types. They are excellent for presenting the relationship between two data series with different scales. The most common variant is the combination of bar and line charts.

Example: Revenue (in millions) and number of employees (up to 100).

Feel free to use professionally designed slide templates for combination charts from PresentationLoad!

8. Radar Chart

The spider chart, also known as a star or radar chart, is particularly useful for displaying the development or characteristics of predefined criteria . Each category has its own axis, with the zero point located at the center.

Example: Comparing two companies based on predefined criteria (including benchmarking).

Feel free to use professionally designed slide templates for spider charts from PresentationLoad!

9. Portfolio Diagram

The bubble chart, also known as a portfolio chart, stands out with its three dimensions. The X and Y axes represent the measurement of a variable defined for each axis. This creates an accurate position of the bubble within the coordinate system. Additionally, the size of the bubble represents a third dimension.

Example : BCG matrix (depicting market growth, relative market share, and revenue).

Feel free to use professionally designed slide templates for bubble charts from PresentationLoad!

10. Waterfall Chart

The waterfall chart is a special form of the bar chart. It shows an initial value that is increased or decreased by additional values . Finally, the end value is depicted.

Example: Breaking down total costs into individual costs.

Feel free to use professionally designed slide templates for waterfall charts from PresentationLoad!

11. Bubble Chart

A bubble chart is used in data visualization to represent relationships between three or more variables . The purpose of a bubble chart is to visualize complex datasets in a simple and easily understandable way.

In a bubble chart, data points are represented as circles (bubbles), where the position of the bubbles on the X and Y axes represents the two main variables. The size of the bubbles represents a third variable, and in some cases, the color of the bubbles can be used to represent a fourth variable.

Companies use bubble charts to illustrate relationships between various financial data, such as in strategic management when visualizing BCG matrices.

Example: Creating a market share overview where revenue and product quantity are represented on the X and Y axes, and the respective market share is indicated by the different sizes of the bubbles.

12. Scatter Diagram

A scatter plot is used to represent the relationship between two continuous variables. The purpose of a scatter plot is to visualize the c orrelation or pattern between these variables in a simple and easily understandable way . If there are dependencies between the two variables, patterns or structures such as linear or quadratic relationships can be observed, revealing average values, trends/developments, or concentrations.

In a scatter plot, data points are represented as dots or symbols, where the position of the points on the X and Y axes represents the two variables. The points are plotted independently, and their distribution in the chart shows the relationship between the variables.

Example: Examining the relationship between age and income.

Feel free to use professionally designed slide templates for scatter plots from PresentationLoad!

13. Sales Funnel

A funnel chart is used to represent the different stages of a process or sales pipeline . The shape of the funnel is crucial in the visualization. The first stage of the process is represented by the wider end, and the narrower end represents the final stage. The size of the sections within the funnel represents the number of items or data points in each stage of the process. The decreasing width of the funnel represents the decreasing magnitude of items transitioning from one stage to the next.

The purpose of a funnel chart is to visualize the number of items or data points going through the different stages of a process in a simple and easily understandable way. Funnel charts are often used to identify and analyze bottlenecks or weaknesses in a process.

Example: Analyzing a sales pipeline.

Feel free to use professionally designed slide templates for funnel charts from PresentationLoad!

14. Pyramid Chart

Pyramid charts are primarily used to represent demographic information in an easily understandable way. The chart depicts a vertically oriented, two-dimensional histogram.

Example: Visualizing the age structure and gender distribution of a population.

Feel free to use professionally designed slide templates for pyramid charts from PresentationLoad!

15. Gantt Chart

A Gantt chart is used to visually represent the activities of a project in a time-oriented table format. The structure of the chart allows for listing all project-related activities and their duration. This is displayed in the form of a bar that indicates both the start and end points of a time-based activity.

The chart provides an overview of how much time is required for each activity and when it will be completed, allowing project managers to have better control over the project timeline. It enables them to identify areas where the project is successful or not, thereby optimizing process flows through appropriate interventions.

Example: Presentation and management of a construction project.

Feel free to use professionally designed slide templates for Gantt charts from PresentationLoad!

16. Venn Diagram

Venn diagrams, using two, three, or more circles, are a practical method for illustrating overlapping or interconnected relationships. They provide a visual representation of the relationships and dependencies within a complex set of elements.

Venn diagrams can be a valuable tool for capturing the entirety of complex situations and understanding the relationships between elements. For more information, feel free to check out our blog post on “ Venn Diagrams “.

Example: Analyzing the similarities and differences between different customer segments in a company.

Feel free to use professionally designed slide templates for Venn diagrams from PresentationLoad!

17. Process Diagram (for example Flowchart)

Process diagrams, such as flowcharts, are excellent for presenting processes and workflows in a clear and organized manner. They can represent both general concepts and specific relationships, making them a valuable tool for any company looking to showcase their business processes and workflows to stakeholders. Algorithms, workflows, and processes can be translated into flowcharts, facilitating analysis, documentation, and management of programs and workspaces.

Flowcharts are widely used and established in sectors such as business, finance, IT, and data processing, thanks to their effective visual representation.

Example: Illustrating and analyzing a customer service process in a company.

For more information, feel free to check out our blog post on “ Flowcharts “.

Feel free to use professionally designed slide templates for flowcharts from PresentationLoad!

18. Organizational Chart

An organizational chart is a chart typed used to structure and organize a company or project , allowing for the clear representation of hierarchies. There are various types of organizational charts to choose from, including the single-line system, multiple-line system, matrix organization, and staff line representation.

The typica l single-line system emphasizes clear responsibilities and a streamlined structure, while the multiple-line system shortens information pathways and contributes to specialization within individual instances.

Example: Presenting a product range or service offering in a clear and organized manner.

For more information, feel free to check out our blog post on “ Organizational Charts “.

Feel free to use professionally designed slide templates for organizational charts from PresentationLoad!

Conclusion: Finding the right chart types for your purposes

In conclusion, you can find the right type of diagram for your purposes by referring to our 18 chart types and determining which one best suits your needs. With the appropriate diagram, you can visualize content much more easily and quickly, making it understandable for your audience.

If you have any questions regarding the article, feel free to contact us via email at [email protected] . We are here to assist you!

If you’re looking for visually supportive and professionally designed slide templates , be sure to check out our shop. We have a wide range of slides available for download on various (business) topics. Visit our shop today! ► Go to Shop

You may also be interested in the following articles:

• Create a Flowchart in PowerPoint
• Create a Venn Diagram
• Create an Organizational Chart
• 8 Tips for better Bar Chart in PowerPoint
• Present Numbers and Tables in an Engaging Way

Share this post

• share  
• save  

Design Thinking: Problem Solving with a Difference

Why Corporate Mission Statements Are So Important

7 Tips & Learnings from the Apple Keynote

How to Create Charts and Graphs to Visualize Data in PowerPoint

The use of data visualization has become increasingly important in today’s digital age, as more and more people have access to large quantities of data. Whether working with data for business, education, research, or personal use, it’s essential to present it in a clear and meaningful way, so it can be easily understood and analyzed. One of the most popular tools for creating visually appealing charts and graphs is Microsoft PowerPoint.

Table of Contents

Understanding the Importance of Data Visualization

Data visualization is the process of representing information in a visual format, such as charts, graphs, and diagrams. When done effectively, it provides a way to quickly understand complex data and identify patterns and trends that may be missed when viewing it in a static table or spreadsheet. Visualizations also allow you to tell a story with the data, making it engaging and memorable for your audience.

Moreover, data visualization can also help in identifying outliers and anomalies in the data, which can be crucial in decision-making processes. It can also aid in identifying correlations and relationships between different variables, which can lead to new insights and discoveries. Additionally, data visualization can be used to communicate data-driven insights to stakeholders and decision-makers, making it an essential tool in various industries such as business, healthcare, and education.

Choosing the Right Chart or Graph for Your Data

There are many different types of charts and graphs to choose from, each with its own strengths and weaknesses. When selecting the right chart or graph for your data, it’s important to consider the following factors:

• The type of data you have (categorical or numerical)
• The relationships between the data points
• The purpose of your presentation

Some of the most common chart types include bar charts, line graphs, pie charts, and scatter plots. Each chart type can display your data in a different way, highlighting specific features depending on the nature of your data.

Another important factor to consider when choosing a chart or graph is the audience you will be presenting to. Different types of charts and graphs may be more effective for different audiences. For example, a pie chart may be more easily understood by a general audience, while a scatter plot may be more appropriate for a technical audience.

It’s also important to consider the context in which your data will be presented. If you are presenting data in a business setting, for example, you may want to choose a chart or graph that emphasizes the financial implications of your data. On the other hand, if you are presenting data in an academic setting, you may want to choose a chart or graph that emphasizes the statistical significance of your data.

Creating a Bar Chart in PowerPoint

Bar charts are one of the most common chart types used in data visualization. They are useful for comparing values across different categories. To create a bar chart in PowerPoint:

• Select the data you want to include in the chart
• Click the ‘Insert’ tab and select the ‘Bar’ chart type
• Choose the specific bar chart subtype you want to use (such as stacked, clustered, or 100% stacked)
• Format the chart by adding labels, titles, and modifying the color scheme

With just a few clicks, you can create a visually appealing bar chart that highlights the differences and similarities between your data categories.

It is important to note that when creating a bar chart, you should carefully consider the data you are presenting and choose the appropriate chart subtype. For example, a stacked bar chart may be useful for showing the total value of each category, while a clustered bar chart may be better for comparing values within each category.

Additionally, you can customize your bar chart further by adding data labels, changing the axis titles, and adjusting the chart layout. Experiment with different options to find the best way to present your data in a clear and visually appealing way.

Making a Line Graph with PowerPoint

Line graphs are another common type of chart used to display numerical data. They are useful for showing trends over time. To create a line graph in PowerPoint:

• Click the ‘Insert’ tab and select the ‘Line’ chart type
• Choose the specific line graph subtype you want to use (such as 2D or 3D)
• Add labels, titles, and customize the color scheme

Line graphs are useful because they allow you to see how your data changes over time. They can make it easier to identify trends or patterns that may be hidden in other types of charts.

One important thing to keep in mind when creating a line graph is to ensure that your data is properly formatted. This means that your data should be organized in a way that makes sense for the type of graph you are creating. For example, if you are creating a line graph to show the sales of a particular product over time, you should organize your data by date and sales figures.

Another useful feature of line graphs in PowerPoint is the ability to add trendlines. Trendlines are lines that are added to a graph to help you see the overall trend of your data. They can be useful for identifying patterns or predicting future trends. To add a trendline in PowerPoint, simply right-click on the data series you want to add the trendline to, and select ‘Add Trendline’ from the menu.

Pie Charts: When and How to Use Them

Pie charts are a popular choice for showing proportions of a whole. They are useful for displaying categorical data and can quickly give an idea of the main contributors for something. However, they can be difficult to interpret when many sections are used. Some tips for making a great pie chart:

• Limit pie charts to 5-7 sections at most
• Make sure percentages add up to 100
• Make labels visible and clear
• Highlight important sections to draw attention

Keep in mind that while pie charts can be visually impactful, they should only be used when they effectively convey the data being presented.

Another important consideration when using pie charts is to ensure that the sections are proportional to the data they represent. If one section is significantly larger than the others, it can skew the overall perception of the data. Additionally, it’s important to choose colors that are easily distinguishable from each other, especially for those who may have color blindness.

While pie charts are a great option for displaying categorical data, they may not be the best choice for showing changes over time or comparing multiple sets of data. In these cases, a line graph or bar chart may be more appropriate. It’s important to consider the type of data being presented and choose the appropriate visualization method to effectively communicate the information.

Creating a Stacked Column Chart in PowerPoint

Stacked column charts are useful for showing how different parts of a whole contribute to the total, while also comparing values for different categories. To create a stacked column chart in PowerPoint:

• Click the ‘Insert’ tab and select the ‘Column’ chart type
• Select the ‘Stacked Column’ subtype
• Format the chart by adding labels, titles, and modifying the colors of the columns

With stacked column charts, you can communicate a lot of information clearly and efficiently.

Using Bubble Charts for Comparative Analysis

Bubble charts are a useful way to show three dimensions of data in a single graph. They are great for comparing two numerical data sets paired with a categorial one. To create a bubble chart in PowerPoint:

• Click the ‘Insert’ tab and select the ‘Bubble’ chart type
• Format the chart by adding labels, titles, and modifying the size, color, and alignment of the bubbles

Bubble charts are perfect for comparing three-dimensional data sets, highlighting the relationships between the various elements being compared.

Adding Labels and Titles to Your Charts and Graphs

Labels and titles are key to effective data visualization. They provide context for the chart or graph you’re presenting and help your audience understand your data. To add labels and titles to your charts and graphs in PowerPoint:

• Select the chart or graph you want to add labels and titles to
• Click on the ‘Chart Elements’ button in the upper-right corner of the chart
• Select the elements you want to add, and choose from the available options for formatting and positioning

By adding labels and titles, you can make your charts and graphs much more informative and easier to understand.

Customizing Colors and Styles for Better Visual Appeal

Colors, styles, and formatting can make a big difference when it comes to the visual appeal of your charts and graphs. Customizing options in PowerPoint allows you to personalize the look of your visualizations. Some tips:

• Use consistent branding colors to help maintain visual consistency
• Choose high-contrast color combinations to help text and graphics stand out
• Avoid too many colors, keeping the chart or graph simple and clear

Customizing colors and styles helps bring cohesion to your presentation while making it more engaging to your audience.

Animating Your Charts and Graphs for Presentations

Animated charts and graphs can be eye-catching and effective for presentations, as they create a sense of dynamism and show how data changes over time. To animate your charts and graphs in PowerPoint:

• Select the chart or graph you want to animate
• Click on the ‘Animations’ tab, and select the type of animation you want to use
• Customize the animation settings to suit your needs, including duration, direction and order of animations

Animations bring data to life, making them more memorable for your audience.

Tips and Tricks for Effective Data Visualization in PowerPoint

Effective data visualization isn’t just about picking the right chart or graph type. There are additional tips and tricks that you can use to make sure your data is presented in the most meaningful way. Here are some things to keep in mind:

• Keep it simple, using plain and unambiguous language
• Choose the right chart or graph type, fitting your data needs as well as your presentation goals
• Make it easy to read, using appropriate font sizes, colors, and layout
• Use engaging visuals, adding icons and images where appropriate
• Tell a story, organizing the chart or graph in a logical and meaningful manner

By paying attention to these tips, you’ll be able to create visually appealing and effective data visualizations that effectively communicate your message to your audience.

By humans, for humans - Best rated articles:

Excel report templates: build better reports faster, top 9 power bi dashboard examples, excel waterfall charts: how to create one that doesn't suck, beyond ai - discover our handpicked bi resources.

Explore Zebra BI's expert-selected resources combining technology and insight for practical, in-depth BI strategies.

We’ve been experimenting with AI-generated content, and sometimes it gets carried away. Give us a feedback and help us learn and improve! 🤍

Note: This is an experimental AI-generated article. Your help is welcome. Share your feedback with us and help us improve.

Jiaxuan You

Do not go gentle into that good night

Jiaxuan You (尤佳轩)

Hi! I am a senior undergraduate in Department of Automation, School of Information Science and Technology, Tsinghua University. I will become a PhD student in the Computer Science Department at Stanford University in 2017 fall. Since April 2016, I've been working with Prof. Stefano Ermon and Prof. David Lobell , and spent a memorable summer in Stanford Artificial Intelligence Laboratory . In Tsinghua, I've been working on Machine Learning since October 2015, supervised by Prof. Jun Zhu . Before stepping into the world of AI, I had been working on modeling and simulation, under the supervision of Prof. Yi Zhang . My research interests lie within Machine Learning .

I am currently an Adjunct Assistant Professor at UIUC CS, and I will join UIUC CS full-time as a tenure-track Assistant Professor in 2024 Fall. I'm taking a gap year as a Senior Research Scientist at NVIDIA. I received my Ph.D. and M.S. degrees from Department of Computer Science, Stanford University, advised by Prof. Jure Leskovec . Jun Zhu at Tsinghua, with Prof. Stefano Ermon and Prof. David Lobell as a summer intern in 2016, with Kaiming He and Saining Xie as a summer intern in Facebook AI Research in 2019. --> I was supported by JPMC PhD Fellowship and Baidu Scholarship during my PhD. My research leads to Kumo AI , where I built the first graph learning predictive system for relational databases as a core founding member from 2021 to 2023.

• AI Agent: Exploring methodologies to enable AI to utilize tools and optimize itself based on those tools.
• ML System: Strategies to enhance the inference and training of Large Language Models (LLMs) & Foundation Models, and facilitate their deployment and application.
• Empowering AI with Relational Data: Investigating the utilization of AI to analyze and comprehend the interconnected digital world.
• AI and Beyond: Delving into how AI research can profoundly reshape the future of scientific research and the broader human society.

Prospective student

Work/teaching experience, professional services, open-source software, publications.

• Handling Missing Data with Graph Neural Networks Jiaxuan You* , Xiaobai Ma*, Daisy Yi Ding*, Mykel Kochenderfer, Jure Leskovec 34th Conference on Neural Information Processing Systems ( NeurIPS 2020b ) [ PDF ] [ Code ] [ Webpage ]
• Graph Structure of Neural Networks Jiaxuan You , Jure Leskovec, Kaiming He, Saining Xie 37th International Conference on Machine Learning ( ICML 2020 ) Long Oral [ PDF ] [ Code ] [ Video Recording ] [ Slides ]
• Redundancy-Free Computation for Graph Neural Networks​ Zhihao Jia, Sina Lin, Rex Ying, Jiaxuan You , Jure Leskovec, Alex Aiken 26th ACM SIGKDD Conference on Knowledge Discovery and Data Mining ( KDD 2020 ) [ PDF ]
• G2SAT: Learning to Generate SAT Formulas Jiaxuan You* , Haoze Wu*, Clark Barrett, Raghuram Ramanujan, Jure Leskovec. 33th Conference on Neural Information Processing Systems ( NeurIPS 2019a ) [ PDF ] [ Code ] [ Webpage ]
• GNNExplainer: A Tool for Post-hoc Explanation of Graph Neural Networks Rex Ying, Dylan Bourgeois, Jiaxuan You , Marinka Zitnik, Jure Leskovec 33th Conference on Neural Information Processing Systems ( NeurIPS 2019b ) [ PDF ] [ Code ] [ Webpage ]
• Position-aware Graph Neural Networks Jiaxuan You , Rex Ying, Jure Leskovec 36th International Conference on Machine Learning ( ICML 2019 ) Long Oral [ PDF ] [ Code ] [ Webpage ] [ Video Recording ]
• Hierarchical Temporal Convolutional Networks for Dynamic Recommender Systems Jiaxuan You , Yichen Wang, Aditya Pal, Pong Eksombatchai, Chuck Rosenberg, Jure Leskovec The Web Conference 2019 ( WWW 2019 ) [ PDF ] [ Code ]
• Graph Convolutional Policy Network for Goal-Directed Molecular Graph Generation Jiaxuan You* , Bowen Liu*, Rex Ying, Vijay Pande, Jure Leskovec 32th Conference on Neural Information Processing Systems ( NeurIPS 2018a ) Spotlight presentation [ PDF ] [ Code ]
• Hierarchical Graph Representation Learning with Differentiable Pooling Rex Ying, Jiaxuan You , Christopher Morris, Xiang Ren, William L. Hamilton, Jure Leskovec 32th Conference on Neural Information Processing Systems ( NeurIPS 2018b ) Spotlight presentation [ PDF ] [ Code ]
• GraphRNN: Generating Realistic Graphs with Deep Auto-regressive Model Jiaxuan You* , Rex Ying*, Xiang Ren, William L. Hamilton, Jure Leskovec 35th International Conference on Machine Learning ( ICML 2018 ) [ PDF ] [ Code ]
• Deep Gaussian Process for Crop Yield Prediction Based on Remote Sensing Data Jiaxuan You , Xiaocheng Li, Melvin Low, David Lobell, Stefano Ermon 31th AAAI Conference on Artificial Intelligence ( AAAI 2017 ) Oral, Best Student Paper Award (Computational Sustainability Track) [ PDF ] [ Code ] [ Project Webpage ]
• Scalable Crop Yield Prediction Approach by Combining Deep Learning with Remote Sensing Data Jiaxuan You , Xiaocheng Li, Stefano Ermon Best Big Data Solution in World Bank Big Data Innovation Challenge 1st place among 180+ teams [ link ] [ Supplementary Materials ]
• An Effective Simulation Model for Multi-line Metro Systems Based on Origin-destination Data Jiaxuan You , Wei Guo, Yi Zhang, et al. 19th IEEE International Conference on Intelligent Transportation Systems ( ITSC 2016 ) As the only undergraduate attendee, I gave talks for 4 papers and was warmly welcomed [ PDF ] [ Photo ]
• Travel Modal Choice Analysis for Traffic Corridors Based on Decision-theoretic Approaches Wei Guo, Yi Zhang, Jiaxuan You , et al. Journal of Central South University (SCI, EI), Nov 2015. [ PDF ]
• Jiaxuan You , An Analysis for the Environmental Impact of Electric Vehicles in China Based on Empirical Investigation and Quantitative Estimation, Review of Economic Research 37 (2015), a Chinese Core Periodical. [ PDF ]

Research Highlights

Latest papers.

Relational Multi-Task Learning: Modeling Relations between Data and Tasks   (ICLR 2022)

Here we introduce a novel relational multi-task learning setting where test data point may present auxiliary task labels. We develop MetaLink, where our key innovation is to build a knowledge graph that connects data points and tasks and thus allows us to leverage labels from auxiliary tasks. [ PDF ] [ Code ]

Identity-aware Graph Neural Networks   (AAAI 2021)

Here we develop a class of message passing GNNs, named Identity-aware Graph Neural Networks (ID-GNNs), with greater expressive power than the 1-WL test. ID-GNN offers a minimal but powerful solution to limitations of existing GNNs. [ PDF ] [ Code ] [ Webpage ]

Design Space for Graph Neural Networks   (NeruIPS 2020a)

Here we define and systematically study the architectural design space for GNNs which consists of 315,000 different designs over 32 different predictive tasks. We release GraphGym, a powerful platform for exploring different GNN designs and tasks. [ PDF ] [ Code ] [ Webpage ]

Handling Missing Data with Graph Neural Networks   (NeruIPS 2020b)

Here, we propose GRAPE, a general framework for feature imputation and label prediction in the presence of missing data. Our key innovation is to formulate the problem using a graph representation, where observations and features are two types of nodes, and the observed feature values are attributed edges. [ PDF ] [ Code ] [ Webpage ]

Graph Structure of Neural Networks   (ICML 2020)

Here we systematically investigate how does the graph structure of neural networks affect their predictive performance. Our work opens new directions for the design of neural architectures and the understanding on neural networks in general. [ PDF ] [ Code ] [ Video Recording ] [ Slides ]

Deep generative models for graphs ("Graph decoder")

• GraphRNN : one of the first deep generative models for graphs

GraphRNN: Generating Realistic Graphs with Deep Auto-regressive Model   (ICML 2018)

Here we propose GraphRNN, a deep autoregressive model that addresses the above challenges and approximates any distribution of graphs with minimal assumptions about their structure. [ PDF ] [ Code ]

• GCPN : generate graph to satisfy user-provided goals, applied to molecule generation

GCPN: Reinforcement Learning for Goal-Directed Molecular Graph Generation   (NeruIPS 2018)

Here we propose Graph Convolutional Policy Network (GCPN), a general graph convolutional network based model for goal-directed graph generation through reinforcement learning. [ PDF ] [ Code ] 

• G2SAT : highly scalable graph generator (over 25K nodes), applied to SAT formula generation

G2SAT: Learning to Generate SAT Formulas (NeurIPS 2019)

Here we present G2SAT, the first deep generative framework that learns to generate SAT formulas from a given set of input formulas. [ PDF ] [ Code ] [ Webpage ]

Advanced representation learning models for graphs ("Graph encoder")

DiffPool: Differentiable Pooling layer for Graph Networks   (NeurIPS 2018)

Here we propose DiffPool, a differentiable graph pooling module that can generate hierarchical representations of graphs and can be combined with various graph neural network architectures in an end-to-end fashion. [ PDF ] [ Code ]

P-GNN: Position-aware Graph Neural Networks   (ICML 2019)

Here we propose Position-aware Graph Neural Networks (PGNNs), a new class of GNNs for computing position-aware node embeddings which existing GNNs cannot represent. [ PDF ] [ Code ] [ Webpage ] [ Video Recording ]

Applications that leverage graph structure

HierTCN: Hierarchical Temporal Convolutional Networks for Dynamic Recommender Systems   (WWW 2019)

Here we propose Hierarchical Temporal Convolutional Networks (HierTCN), a hierarchical deep learning architecture that makes dynamic recommendations based on users' sequential multi-session interactions with items. [ PDF ] [ Code ] [ PDF ] [ Code ]  -->

GNNExplainer: A Tool for Post-hoc Explanation of Graph Neural Networks   (NeurIPS 2019)

Here we propose GNNExplainer, the first general, model-agnostic approach for providing interpretable explanations for predictions of any GNN-based model on any graph-based machine learning task. [ PDF ] [ Code ] [ Webpage ]

HAG: Redundancy-Free Computation Graphs for Graph Neural Networks​   (KDD 2020)

Here we propose Hierarchically Aggregated computation Graphs (HAGs), a new GNN representation technique that explicitly avoids redundancy by managing intermediate aggre- gation results hierarchically and eliminates repeated computations and unnecessary data transfers in GNN training and inference. [ PDF ]

Interdisciplinary research

Crop yield prediction: machine learning over satellite images    (aaai 2017).

Crop yield prediction is central in ensuring the food security. We introduce the first deep learning based method to predict crop yield purely based on publicly available remote sensing data. [ PDF ] [ Code ] [ Project Webpage ]

An Effective Simulation Model for Multi-line Metro Systems    (ITSC 2016)

This paper presents an effective simulation model for multi-line metro systems based on the OD (origin-destination) data and the network connection data. [ PDF ] 

National Scholarship   2015,2016

Shu pin scholarship   2014,2015,2016, min sheng scholarship   2015, tsinghua comprehensive merit scholarship   2014,2015,2016, course work, programming.

• I was born on Jan 5th, 1997.
• I play drones. Some of the highlight videos are in my youtube channel .
• PhD students are financially poor. If you would like to (financially) support me for better research, please feel free to check out these referral links :) Of course, citations are even more welcomed :p
• I've traveled to lots of places. See a map .
• My life is driven by a simple but powerful idea, "Making the world a better place". Therefore, before turning into AI research, I've done researches in Chemistry, Economics, Social Science, Transportation and Simulation. These varied research experiences have equipped me with competence, creativity and perseverance, in the meantime, have convinced me that AI is the right path to fulfill my dream.
• © HTML5 UP 2016

Chapter 2: Representing Motion

Principles and Problems

You can use displacement and velocity to describe an object’s motion.

Representing Motion

Section 2.1 Picturing Motion

Section 2.2 Where and When?

Section 2.3 Position-Time Graphs

Section 2.4 How Fast?

Table Of Contents

Click a hyperlink to view the corresponding slides.

• How do motion diagrams represent motion?
• How can you use a particle model to represent a moving object?

You can use motion diagrams to show how an object’s position changes over time.

Essential Questions

SECTION 2.1

Picturing Motion

New Vocabulary

• Motion diagram
• Particle model

Review Vocabulary

• Model a representation of an idea, event, structure or object to help people better understand it.
• Perceiving motion is instinctive—your eyes pay more attention to moving objects than to stationary ones. Movement is all around you.
• Movement travels in many directions, such as the straight-line path of a bowling ball in a lane’s gutter, the curved path of a tether ball, the spiral of a falling kite, and the swirls of water circling a drain.

All Kinds of Motion

• When an object is in motion, its position changes. Its position can change along the path of a straight line, a circle, an arc, or a back-and-forth vibration.

All Kinds of Motion (cont.)

• Straight-line motion follows a path directly between two points without turning left or right.
• Ex. Forward and backward, up and down, or north and south.
• A description of motion relates to place and time. You must be able to answer the questions of where and when an object is positioned to describe its motion.

Motion Diagrams

Explain how applying the particle model produces a simplified version of a motion diagram?

Section Check

Answer: Keeping track of the motion of the runner is easier if we disregard the movements of the arms and the legs, and instead concentrate on a single point at the center of the body. In effect, we can disregard the fact that the runner has some size and imagine that the runner is a very small object located precisely at that central point. A particle model is a simplified version of a motion diagram in which the object in motion is replaced by a series of single points.

Which statement describes best the motion diagram of an object in motion?

A. a graph of the time data on a horizontal axis and the position on a vertical axis

B. a series of images showing the positions of a moving object at equal time intervals

C. a diagram in which the object in motion is replaced by a series of single points

D. a diagram that tells us the location of the zero point of the object in motion and the direction in which the object is moving

Reason: A series of images showing the positions of a moving object at equal time intervals is called a motion diagram.

What is the purpose of drawing a motion diagram or a particle model?

A. to calculate the speed of the object in motion

B. to calculate the distance covered by the object in a particular time

C. to check whether an object is in motion

D. to calculate the instantaneous velocity of the object in motion

Reason: In a motion diagram or a particle model, we relate the motion of the object with the background, which indicates that relative to the background, only the object is in motion.

• What is a coordinate system?
• How does the chosen coordinate system affect the sign of objects’ positions?
• How are time intervals measured?

What is displacement?

• How are motion diagrams helpful in answering questions about an object’s position or displacement?

A coordinate system is helpful when you are describing motion.

SECTION 2.2

Where and When?

• Coordinate system
• Dimension extension in a given direction;one dimension is along a straight line; three dimensions are height, width and length.
• Time interval
• Displacement
• A coordinate system tells you the location of the zero point of the variable you are studying and the direction in which the values of the variable increase.
• The origin is the point at which both variables have the value zero.

Coordinate Systems

• In the example of the runner, the origin, represented by the zero end of the measuring tape, could be placed 5 m to the left of the tree.

Coordinate Systems (cont.)

• The motion is in a straight line, thus, your measuring tape should lie along that straight line. The straight line is an axis of the coordinate system.
• You can indicate how far away an object is from the origin at a particular time on the simplified motion diagram by drawing an arrow from the origin to the point representing the object, as shown in the figure.
• The two arrows locate the runner’s position at two different times.
• Because the motion in the figure below is in one direction, the arrow lengths represent distance.
• The length of how far an object is from the origin indicates its distance from the origin.
• A position 9 m to the left of the tree, 5 m left of the origin, would be a negative position, as shown in the figure below.
• Quantities that have both size, also called magnitude , and direction, are called vectors , and can be represented by arrows.
• Vector quantities will be represented by boldface letters.
• Quantities that are just numbers without any direction, such as distance, time, or temperature, are called scalars .
• Scalars quantities will be represented by regular letters.

Vectors and Scalars

• The difference between the initial and the final times is called the time interval .

Vectors and Scalars (cont.)

• The common symbol for a time interval is ∆ t , where the Greek letter delta, ∆, is used to represent a change in a quantity.
• The time interval is defined mathematically as follows:
• Although i and f are used to represent the initial and final times, they can be initial and final times of any time interval you choose.
• The time interval is a scalar because it has no direction.
• The figure below shows the position of the runner at both the tree and the lamppost.
• These arrows have magnitude and direction.
• Position is a vector with the arrow’s tail at the origin and the arrow’s tip at the place.
• The symbol x is used to represent position vectors mathematically.
• X i represents the position at the tree, x f represents the position at the lamppost and ∆ x, represents the change in position, displacement , from the tree to the lamppost.
• Displacement is defined mathematically as:
• Remember that the initial and final positions are the start and end of any interval you choose, so a plus and minus sign might be used to indicate direction.

∆ x = x f - x i

• A vector that represents the sum of two other vectors is called a resultant .
• The figure to the right shows how to add and subtract vectors in one dimension.
• To completely describe an object’s displacement, you must indicate the distance it traveled and the direction it moved. Thus, displacement, a vector, is not identical to distance, a scalar; it is distance and direction.
• While the vectors drawn to represent each position change, the length and direction of the displacement vector does not.
• The displacement vector is always drawn with its flat end, or tail, at the earlier position, and its point, or tip, at the later position.

Differentiate between scalar and vector quantities.

Reason: Quantities that have both magnitude and direction are called vectors, and can be represented by arrows. Quantities that are just numbers without any direction, such as time, are called scalars.

A. the vector drawn from the initial position to the final position of the motion in a coordinate system

B. the distance between the initial position and the final position of the motion in a coordinate system

C. the amount by which the object is displaced from the initial position

D. the amount by which the object moved from the initial position

Reason: Options B, C, and D are all defining the distance of the motion and not the displacement. Displacement is a vector drawn from the starting position to the final position.

Refer to the adjoining figure and calculate the time taken by the car to travel from one signal to another signal?

Reason: Time interval Δ t = t f – t i

Here t f = 01:45 and t i = 01:20

Therefore, Δ t = 25 min

• What information do position-time graphs provide?
• How can you use a position-time graph to interpret an object’s position or displacement?
• What are the purposes of equivalent representations of an object’s motion?

You can use position-time graphs to determine an object’s position at a certain time.

SECTION 2.3

Position-Time Graphs

• Position-time graph
• Instantaneous position
• Intersection a point where lines meet and cross.

Finding Positions

Click image to view movie.

• Graphs of an object’s position and time contain useful information about an object’s position at various times. It can be helpful in determining the displacement of an object during various time intervals.

Finding Positions (cont.)

• The data in the table can be presented by plotting the time data on a horizontal axis and the position data on a vertical axis, which is called a position-time graph .
• To draw the graph, plot the object’s recorded positions. Then, draw a line that best fits the recorded points. This line represents the most likely positions of the runner at the times between the recorded data points.
• The symbol x represents the instantaneous position of the object—the position at a particular instant.
• Words, pictorial representations, motion diagrams, data tables, and position-time graphs are all representations that are equivalent. They all contain the same information about an object’s motion.
• Depending on what you want to find out about an object’s motion, some of the representations will be more useful than others.

Multiple Objects on a Position-Time Graph

In the graph, when and where does runner B pass runner A?

Step 1: Analyze the Problem

Multiple Objects on a Position-Time Graph (cont.)

Restate the questions.

Question 1 : At what time do A and B have the same position?

Question 2 : What is the position of runner A and runner B at this time?

Step 2: Solve for the Unknown

In the figure, examine the graph to find the intersection of the line representing the motion of A with the line representing the motion of B.

These lines intersect at 45 s.

The position of both runners is about 190m from the origin.

B passes A about 190 m beyond the origin, 45.0 s after A has passed the origin.

The steps covered were:

Restate the questions.�

Considering the Motion of Multiple Objects

A position-time graph of an athlete winning the 100-m run is shown. Estimate the time taken by the athlete to reach 65 m.

Reason: Draw a horizontal line from the position of 65 m to the line of best fit. Draw a vertical line to touch the time axis from the point of intersection of the horizontal line and line of best fit. Note the time where the vertical line crosses the time axis. This is the estimated time taken by the athlete to reach 65 m.

A position-time graph of an athlete winning the 100-m run is shown. What was the instantaneous position of the athlete at 2.5 s?

Reason: Draw a vertical line from the position of 2.5 m to the line of best fit. Draw a horizontal line to touch the position axis from the point of intersection of the vertical line and line of best fit. Note the position where the horizontal line crosses the position axis. This is the instantaneous position of the athlete at 2.5 s.

From the following position-time graph of two brothers running a 100-m dash, at what time do both brothers have the same position? The smaller brother started the race from the 20-m mark.

Reason: The two brothers meet at 6 s. In the figure, we find the intersection of lines representing the motion of one brother with the line representing the motion of other brother. These lines intersect at 6 s and at 60 m.

• What is velocity?
• What is the difference between speed and velocity?
• How can you determine an object’s average velocity from a position-time graph?
• How can you represent motion with pictorial, physical, and mathematical models?

An object’s velocity is the rate of change in its position.

SECTION 2.4

• Average velocity
• Average speed
• Instantaneous velocity
• Absolute value magnitude of a number, regardless of sign.
• Suppose you recorded two joggers in one motion diagram, as shown in the figure below. The position of the jogger wearing red changes more than the of the jogger wearing blue

Velocity and Speed

• For a fixed time, the magnitude of the displacement ( ∆ x ), is greater for the jogger in red.
• If each jogger travels 100m, the time interval ( ∆ t) would be smaller for the jogger in red.
• Recall from Chapter 1 that to find the slope, you first choose two points on the line.
• Next, you subtract the vertical coordinate ( x in this case) of the first point from the vertical coordinate of the second point to obtain the rise of the line.
• After that, you subtract the horizontal coordinate ( t in this case) of the first point from the horizontal coordinate of the second point to obtain the run.
• Finally, you divide the rise by the run to obtain the slope.

Velocity and Speed (cont.)

• The slopes of the two lines are found as follows:
• A greater slope, shows that the red jogger traveled faster.
• The unit of the slope is meters per second. In other words, the slope tells how many meters the runner moved in 1 s.
• The slope is the change in position, divided by the time interval during which that change took place, or ( x f - x i ) / ( t f - t i ), or Δ x /Δ t .
• When Δ x gets larger, the slope gets larger; when Δ t gets larger, the slope gets smaller.
• The slope of a position-time graph for an object is the object’s average velocity and is represented by the ratio of the change of position to the time interval during which the change occurred.

Average Velocity ≡ _______ = ________

( x f - x i )

( t f - t i )

• The symbol ≡ means that the left-hand side of the equation is defined by the right-hand side.
• It is a common misconception to say that the slope of a position-time graph gives the speed of the object.
• The slope of the position-time graph on the right is –5.0 m/s. It indicates the average velocity of the object and not its speed.
• The object moves in the negative direction at a rate of 5.0 m/s.
• The slope’s absolute value is the object’s average speed , 5.0m/s, which is the distance traveled divided by the time taken to travel that distance.
• If an object moves in the negative direction, then its displacement is negative. The object’s velocity will always have the same sign as the object’s displacement.

The graph describes the motion of a student riding his skateboard along a smooth, pedestrian-free sidewalk. What is his average velocity? What is his average speed?

Step 1: Analyze and Sketch the Problem

Identify the coordinate system of the graph.

Identify the unknown variables.

Find the average velocity using two points on the line.

Use magnitudes with signs indicating directions.

Substitute x 2 = 12.0 m, x 1 = 6.0 m, t 2 = 8.0 s, t 1 = 4.0 s:

Step 3: Evaluate the Answer

Are the units correct?

m/s are the units for both velocity and speed.�

Do the signs make sense?

The positive sign for the velocity agrees with the coordinate system. No direction is associated with speed.

The average speed is the absolute value of the average velocity.�

• A motion diagram shows the position of a moving object at the beginning and end of a time interval. During that time interval, the speed of the object could have remained the same, increased, or decreased. All that can be determined from the motion diagram is the average velocity.
• The speed and direction of an object at a particular instant is called the instantaneous velocity .
• The term velocity refers to instantaneous velocity and is represented by the symbol v .
• Although the average velocity is in the same direction as displacement, the two quantities are not measured in the same units.
• Nevertheless, they are proportional—when displacement is greater during a given time interval, so is the average velocity.
• A motion diagram is not a precise graph of average velocity, but you can indicate the direction and magnitude of the average velocity on it.

Equation of Motion

• Using the position-time graph used before with a slope of -5.0m/s, remember that you can represent any straight line with the equation, y = mx + b .
• y is the quantity plotted on the vertical axis, m is the line’s slope, x is the quantity plotted on the horizontal axis and b is the line’s y -intercept.
• Based on the information shown in the table, the equation y = mx + b becomes x = t + x i , or, by inserting the values of the constants, x = (–5.0 m/s) t + 20.0 m.
• You cannot set two items with different units equal to each other in an equation.

Equation of Motion (cont.)

• An object’s position is equal to the average velocity multiplied by time plus the initial position.
• This equation gives you another way to represent the motion of an object.

Which of the following statements defines the velocity of the object’s motion?

A. the ratio of the distance covered by an object to the respective time interval

B. the rate at which distance is covered

C. the distance moved by a moving body in unit time

D. the ratio of the displacement of an object to the respective time interval

Reason: Options A, B, and C define the speed of the object’s motion. The velocity of a moving object is defined as the ratio of the displacement (Δ x ) to the time interval (Δ t ).

Which of the statements given below is correct?

A. Average velocity cannot have a negative value.

B. Average velocity is a scalar quantity.

C. Average velocity is a vector quantity.

D. Average velocity is the absolute value of the slope of a position-time graph.

Reason: Average velocity is a vector quantity, whereas all other statements are true for scalar quantities.

The position-time graph of a car moving on a street is given here. What is the average velocity of the car?

Reason: The average velocity of an object is the slope of a position-time graph.

Physics Online

Study Guide

Chapter Assessment Questions

Standardized Test Practice

• A motion diagram shows the position of an object at successive equal time intervals.
• In the particle model motion diagram, an object’s position at successive times is represented by a series of dots. The spacing between dots indicates whether the object is moving faster or slower.
• A coordinate system gives the location of the zero point of the variable you are studying and the direction in which the values of the variable increase.
• A vector drawn from the origin of a coordinate system to an object indicates the object’s position in that coordinate system. The directions chosen as positive and negative on the coordinate system.
• A time interval is the difference between two times.
• Change in position is displacement, which has both magnitude and direction.
• On a motion diagram, the displacement vector’s length represents how far the object was displaced. The vector points in the direction of the displacement, from xi to xf.
• Position-time graphs provide information about the motion of objects. They also might indicate where and when two objects meet.
• The line on a position-time graph describes an object’s position at each time.
• Motion can be described using words, motion diagrams, data tables or graphs.
• An object’s velocity tells how fast it is moving and in what direction it is moving.
• Speed is the magnitude of velocity.
• Slope on a position-time graph described the average velocity of the object.
• You can represent motion with pictures and physical models. A simple equation relates an object’s initial position ( x i ), its constant average velocity, its position ( x ) and the time ( t ) since the object was at its initial position.

What should be true about the motion of an object in order for you to treat that object as if it were a particle?

A. The object should be no smaller than your fist.

B. The object should be small compared to its motion.

C. The object should be no larger than you can lift.

D. The object should not be moving faster than the speed of sound.

Chapter Assessment

Reason: you can treat even planets and stars as particles as long as those objects are small compared to the motion you are studying.

Which is the distance and direction from one point to another?

A. Displacement

B. Magnitude of distance

C. Position

D. Velocity

Reason: Velocity is speed and direction.

On a position-time graph, how would you indicate that object A has a greater velocity than object B?

D. Make the y -intercept for object A greater than the y -intercept for object B.

A. Make the slope for object A less than the slope for object B.

B. Make the slope for object A greater than the slope for object B.

C. Make the y -intercept for object A less then the y -intercept for object B.

Answer: The slope of a line on a position-time graph indicates the object’s velocity.

A car is moving at a constant speed of 25 m/s. How far does this car move in 0.2 s, the approximate reaction time for an average person?

Reason: (25m/s)(0.2s) = 5m

Which is a measurement of velocity?

C. 300 km west

D. 7800 m/s north

Reason: Velocity measures both speed and direction.

Which statement about velocity vectors is true?

A. All velocity vectors are positive.

B. Velocity vectors have magnitude but no direction.

C. Velocity vectors and displacement vectors are the same thing.

D. A velocity vector’s length should be proportional to the object’s speed.

What is the average speed of a sprinter who completes a 55-m dash in 6.2 s?

Car A is moving faster than Car B on the highway. Which statement describes the particle model motion diagrams for Car A and Car B?

A. The does for Car A are farther apart than the dots for Car B.

B. The dots for Car A are closer together than the dots for Car B.

C. The slope of the motion diagram is greater for Car A than for Car B.

D. The slope of the motion diagram is less for Car A than for Car B.

An athlete runs four complete laps around a 200-m track. What is the athlete’s displacement?

Which correctly describes a relationship between an object’s particle model motion diagram and that object’s graph of position v. time?

• If the dots on the motion diagram are closer together, then the slope of the graph is greater.
• If the dots on the motion diagram are farther apart, then the slope of the graph is greater.
• If the dots on the motion diagram are closer together, then the y -intercept of the graph is less.
• If the dots on the motion diagram are farther apart, then the y -intercept of the graph is less.

Stock up on Supplies

Test-Taking Tip

Bring all your test-taking tools: number two pencils, black and blue pens, erasers, correction fluid, a sharpener, a ruler, a calculator, and a protractor.

Chapter Resources

Coordinate Systems Showing Position

Motion Diagram Showing Negative Position

Position-Time Graph for the Runner

End of Custom Shows

Graph Representation

Oct 01, 2014

450 likes | 589 Views

Graph Representation. D.J. Duke Department of Computer Science University of Bath, U.K. Overview. The challenge of scale Knowledge and cognition kinds of knowledge human information processing Art and Rendering drawing and denotation systems NPR Enhancing visualization

Share Presentation

• representations
• generic algorithms
• information visualization
• scientific visualization
• generic representations streamlines
• generic algorithms reingold tilford

Presentation Transcript

Graph Representation D.J. Duke Department of Computer Science University of Bath, U.K.

Overview • The challenge of scale • Knowledge and cognition • kinds of knowledge • human information processing • Art and Rendering • drawing and denotation systems • NPR • Enhancing visualization • Some of the open issues …

Visualization • Motivations: • presentation: “See this!” • confirmation: “Can we see this?” • exploration: “What can we see?” • Shift work from logical to perceptual • Utilise latent knowledge

/d VTK tools talks vtkbin Rendering Local Debug bin Wrap vtkAdjList.cxx vtkAdjList.h vtkAdjListTcl.obj vtkAdjListTcl.cxx Vtk.exe vtkAdjList.obj Example /d VTK /d talks /d tools VTK Local VTK Rendering VTK vtkbin vtkbin bin vtkbin Debug vtkbin Wrap Local vtkAdjList.h Local vtkAdjList.cxx bin vtk.exe Debug vtkAdjList.obj Debug vtkAdjListTcl.obj Wrap vtkAdjListTcl.cxx

The challenge of scale • Technology pushing level of ambition: • size of datasets • complexity of relationships within the data • complexity of the underlying domain • Some of the solutions: • streaming, parallelism, cluster hardware • level of detail control • Is the user becoming the bottleneck?

Approaches to scale • Scientific visualization – McCormick et al. • Information visualization – Card et al.

Human inference • Levels of inference (Gahegan / MacEachren) • abduction: classification from experience • induction: classification from attributes • deduction: classification from rules • Sources of meaning • public • private • How and where does a representation work?

Cognitive interpretation • Interacting Cognitive Subsystems • Barnard, 1979 - • generic processing unit • cognition distributed across 9 systems • principles of information processing • Use of ICS (Barnard, May, Scott et al.) • clinical psychology; emotion • CTA (Cognitive Task Analysis) • display decomposition

AC MPL ART PROP BS node p540 +-[parent]- p539 +-[children] – [many] IMPLIC node p741? parent = “down” … OBJ VIS LIM Levels of interpretation

Blending at input array Copied into episodic memory ... Incoming representations Transformed into output representations Subsystem operation … and revived

Memory and learning implic Novel representations require intervention of “central engine” PIP-loop prop Over time subsystem image record forms generalised records, e.g. CTRs Eventually, knowledge becomes proceduralized

Uloomo Takete Implicational channels • Much visualization uses structural channel • Implicational “emotive” interpretation is also available …

Non-Photorealistic Rendering (NPR) Schumann, Strothotte, Raab & Laser: Assessing the affect of non-photorealistic rendered images in CAD. In Proc. CHI’96, ACM Press

Representation: geometric structure attribute mappings Design issues: perceptual cues knowledge assumptions kinds of inference General problem in visualization “‘Good’ visualizations mix metaphors” [Hanrahan] The “art” of visualization

C.elegans cell lineage, Sulston From “To Draw a Tree”, P. Hanrahan, 2001

Machine-part assembly From “To Draw a Tree”, P. Hanrahan, 2001

Drawing on art What can we learn from artistic techniques? Interest in • European school of realism vs • Japanese/Chinese traditions

View of Delft Vermeer

Boating on the river below a Buddhist temple Wu Li

Artistic Traditions • Japanese / Chinese painting • relationship to philosophy? • European school of realism • studies of perspective (Dürer) • studies of physiology (da Vinci) • Photography • 20th century: impressionists, cubists …

Les Tuileries Pissarro

Le Jardin Monet

Katata Hiroshige

Drawing Systems • From “The Draughtsman’s Contract” by J. Willats • Spatial relationship between objects • Different kinds of fidelity: • faithful to appearance • faithful to shape • Visualization • how is space used? • what concept(s) does space capture?

Royere tool (www.cwi.nl/InfoVisu)

Denotation system • Relation between marks and real world. • Painting: • Marks represent pattern of light intensities • Visual system interprets as shapes • Drawing • What do lines stand for? • Visualization • What does an edge in a graph represent?

Example (1/2) • Multiple signs: • Nodes • Edges • Elided regions Latour tool (www.cwi.nl/InfoVisu)

Example (2/2) • combine multiple • sources (context + data) • techniques (stream-surface + scalar) • levels of detail • levels of certainty VTK LOxSurface2 example (www.kitware.com/vtk)

Systems issues • Scientific visualization: • generic representations: streamlines, iso-surfaces, … • generic algorithms: marching cubes, … • modular toolkits: AVS, Iris Explorer, VTK, … • Information visualization: • generic representations: trees, cushion maps, … • generic algorithms: Reingold-Tilford, … • tools?

Modularity • Simple dataset model helps sci.vis.: • generic algorithms • composable representations • Aim to achieve similar in info.vis.: • combine distinct representations • support novel info.vis. algorithms • build on infrastructure, e.g. streaming

A Unified View • Data = geometry + topology + attributes • Geometry = points in the space • Topology = organization of points into cells • Implementation: implicit or explicit

Abstract points and cells • Points = data items • Cells = relationships between data a c k n Points = a … e l p j e Cells = j ... p o m b d

BRS-AMS £110 AMS-ZUR AMS-ZUR £140 £60 LHR-ZUR LHR-ZUR £176 £176 0800 1100 1200 1300 0900 1000 Cells as records • BOZ (Casner, 1991) • Query database of flight information • Design display of results relative to task LHR-AMS £105

from = LHR to = ZUR dept = 0800 arr = 0930 price = £176 avail = yes from = LHR to = AMS dept = 0945 arr = 1055 price = £105 avail = no from = LHR to = ZUR dept = 1130 arr = 1300 price = £176 avail = yes from = AMS to = ZUR dept = 0900 arr = 1030 price = £60 avail = yes from = AMS to = ZUR dept = 1100 arr = 1230 price = £140 avail = yes from = BRS to = AMS dept = 0830 arr = 0955 price = £110 avail = yes BOZ (cont.)

VTK • Visualization Toolkit • C++ class library (~ 600 classes) • Wrappers for Tcl, Python, Java • Open source: public.kitware.com/vtk/ • Separation of data and process objects • Datasets – separate slide • Process objects: source, filter, mapper

datafile vtkPolyDataReader vtkPolyDataNormals vtkMapper vtkGlyph3D vtkMapper vtkSphereSource VTK in action • Process execution demand-driven • Only execute if output needs updating ?

vtkTopology vtkGraph vtkGeometry vtkDataObject vtkDataObject vtkGraph VTK Data Sets vtkObject vtkDataSet vtkRectilinearGrid vtkPointSet vtkImageData vtkPolyData vtkPolyData vtkStructuredGrid vtkUnstructuredGrid

vtkConeTree vtkStrahlerMetric vtkGraphGeometryFilter vtkCubeSource vtkTransform vtkFieldDataToAttributeDataFilter vtkTransformPolyDataFilter vtkLookupTable vtkTubeFilter vtkGlyph3D vtkPolyDataMapper vtkPolyDataMapper vtkActor vtkActor vtkRenderer Tree pipeline Source file vtkProgrammableSource

From trees to graphs • Module for general graph layout: • spanning DAG • tree layout • graph • Management of edge bends: • add pseudo-nodes, ignore for node drawing • apply layout positions to “real nodes” • better: use polyline mechanism? graph DAG layout GenGraphLayout3D graph

Initial results • Web site • cone-tree layout • Strahler metric • splatting • FSM simulation • 3D graph layout • generalized Strahler

Further work • Extensions to basic tools • interaction techniques, e.g. brushing • multiple representations • minimal rendering of overview • attribute management • Significant support already present • level-of-detail control • pluggable interactors • multiple viewports

Summary The brain is not subject to Moore’s law. • Visualization is not a panacea … • users’ mental representations • combine discrete / “continuous” models • Knowledge representation • public / private levels [MacEachren] • integration with other representations • Flexible tools are a first step.

Thanks … • David Auber LABRI, Uni. Bordeaux, France • Phil Barnard MRC Cognition & Brain Science Unit, UK • David Duce Comp. Sci, Oxford Brookes University, UK • Ivan Herman W3C, Amsterdam, The Netherlands • Scott Marshall Glaucus Proteomics B.V., The Netherlands • Jon May Psychology, University of Sheffield, UK • UK Engineering and Physical Sciences Research Council

• More by User

Graph-Based Methods for the Representation and Analysis of Business Workflows

Graph-Based Methods for the Representation and Analysis of Business Workflows Amitava Bagchi Indian Institute of Management Calcutta References

993 views • 58 slides

Web Graph representation and compression

Web Graph representation and compression. Thanks to Luciana Salete Buriol and Debora Donato. Locality : usually most of the hyperlinks are local, i.e, they point to other URLs on the same host. The literature reports that on average 80% of the hyperlinks are local.

498 views • 24 slides

Representation

Representation. 18 USC § 203 Compensation. This statute b ars employees from seeking or accepting compensation for assisting in representing another before the executive branch, or Federal courts or receiving money for anyone else’s representation.

411 views • 19 slides

Graph Operations And Representation

Graph Operations And Representation. Sample Graph Problems. Path problems. Connectedness problems. Spanning tree problems. 2. 4. 3. 8. 8. 1. 6. 10. 2. 4. 5. 4. 4. 3. 5. 9. 11. 5. 6. 7. 6. 7. Path Finding. Path between 1 and 8 . Path length is 20 . 2. 4. 3. 8.

431 views • 33 slides

Graph Theory and Representation

Graph Theory and Representation. Graph Algorithms. Graphs and Theorems about Graphs Graph ADT and implementation Graph Algorithms Shortest paths minimum spanning tree. What can graphs model?. Cost of wiring electronic components together. Shortest route between two cities.

419 views • 28 slides

Entropy-biased Models for Query Representation on the Click Graph

Entropy-biased Models for Query Representation on the Click Graph. Hongbo Deng , Irwin King and Michael R. Lyu Department of Computer Science and Engineering The Chinese University of Hong Kong July 2 1st , 2009. Query suggestion Query classification. Targeted advertising Ranking.

499 views • 34 slides

Graph Representation, DFS and BFS

If you have any question, please ask via Email: [email protected] MSN: [email protected]. Graph Representation, DFS and BFS. Gary Wong. WARNING!. Today’s topics are: Tough Probably the hardest among all intermediate topics Essential

713 views • 47 slides

Temporal Action-Graph Games: A New Representation for Dynamic Games

Temporal Action-Graph Games: A New Representation for Dynamic Games. Albert Xin Jiang University of British Columbia. Kevin Leyton-Brown University of British Columbia. Avi Pfeffer Charles River Analytics. Game Theory In One Slide . A game:

539 views • 35 slides

Representation and Learning in Directed Mixed Graph Models

Representation and Learning in Directed Mixed Graph Models. Ricardo Silva Statistical Science/CSML, University College London [email protected] Networks: Processes and Causality, Menorca 2012. Graphical Models. Graphs provide a language for describing independence constraints

693 views • 57 slides

Representation of System Model with Resource-Allocation Graph

DEADLOCK Date- 12/3/14 By Manoj Kumar (11CS10055). Representation of System Model with Resource-Allocation Graph. A set of edges E and A set of vertices V V is partitioned into two types:

412 views • 12 slides

A Bond-Graph Representation of a Two-Gimbal Gyroscope

A Bond-Graph Representation of a Two-Gimbal Gyroscope. Robert T. Mc Bride Dr. François E. Cellier Raytheon Missile Systems, University of Arizona Tucson, Arizona Tucson, Arizona. ABSTRACT.

385 views • 21 slides

Lecture 3 Graph Representation for Regular Expressions

Lecture 3 Graph Representation for Regular Expressions. digraph (directed graph). A digraph is a pair of sets (V, E) such that each element of E is an ordered pair of elements in V .

391 views • 26 slides

6.1.3 Graph representation

6.1.3 Graph representation. 0. 1. 2. 3. G1. 0. 1. 2. 0. 4. G3. 2. 1. 5. 6. 3. 7. G4. 常見的表示法. Adjacency matrices Adjacency lists Adjacency multilists. 0. 1. 2. 3. G1. 0. 1. 2. G3. Adjacency matrix. n*n 陣列 n*(n-1), 即 O(n 2 ) If the matrix is sparse ? 大部分元素是 0

275 views • 18 slides

First graph Second graph Third graph

DO NOW : Take one from my desk. Complete the first 3 boxes, and make up your own for the last box! 5 minutes; need help? Ask now! Also, take your clicker.

307 views • 20 slides

Graph Undirected graph Directed graph

Lecture 6 Graph Traversal. Graph Undirected graph Directed graph. Overview. Graph Undirected graph DFS, BFS, Application Directed graph DFS, BFS, Application. Graph theory. The Königsberg Bridge problem (Source from Wikipedia). Graph terminology. Undirected graph. Directed graph.

661 views • 61 slides

Representation. A farmer wants to move himself, a hunrgry silver fox, a hungry, fat goose, and a sack of tasty grain across a river. Unfortunately, his boat is so tiny he can move only one of his possessions across on any trip. What is he to do?. Fox, Farmer, Goose, Grain.

174 views • 16 slides

Graph Operations And Representation. Sample Graph Problems. Path problems. Connectedness problems. Spanning tree problems. 2. 4. 3. 8. 8. 1. 6. 10. 2. 4. 5. 4. 4. 3. 5. 9. 11. 5. 6. 7. 6. 7. Path Finding. Path between 1 and 8. Path length is 20. 2. 4. 3. 8. 8.

343 views • 33 slides

231 views • 22 slides

260 views • 26 slides

455 views • 45 slides

487 views • 47 slides