In the Fall 2024 semester, we are meeting on Wednesdays 1 - 2 pm (Central): at UNL we will meet in 354A Hardin Hall (large conference room), at ISU we meet in TBD (ask Tom Lin), we will also meet via Zoom. Please join us for presentations on statistical graphics and computational tools.

Contact Heike Hofmann, hhofmann4 at unl.edu, for access to the Zoom link.

Pivoting between space and time: Spatio-temporal analysis with cubble

R Package
cubble
Visualization
Multivariate spatio-temporal data have a spatial component referring to the location of each observation, a temporal component recorded at regular or irregular time intervals, and multiple variables measured at each spatial and temporal value. Often, such data are fragmented, reflecting a common practice of focusing on either spatial or temporal aspects separately. This fragmentation makes it difficult to handle them coherently and comprehensively. This work introduces a new data structure to facilitate the study of different portions or combinations of spatio-temporal data for exploratory data analysis. The proposed structure, implemented in the R package, cubble, organizes spatial and temporal variables as two facets of a single data object, allowing them to be wrangled separately or combined while ensuring synchronization. Examples will be provided to visualize weather station data with cubble using glyph maps.
Dec 4, 2024
Sherry Zhang

Asymptotic timing and memory analysis with the atime packages

R Package
atime
Visualization
The atime package introduces an approach to benchmarking statistical software by enabling asymptotic performance measurement across different data set sizes denoted by N. Unlike other benchmarking functions that compare performance for a single data set, atime allows researchers to measure time and memory usage for increasingly large input sizes, which helps in estimating computational complexity classes using big-O notation. The package has already been applied to improve efficiency in significant projects like the data.table package, demonstrating its practical utility in software optimization.
Nov 20, 2024
Doris Amoakohene

Visualize your fitted nonlinear dimension reduction model in the high-dimensional data space

R Package
quollr
Visualization
Non-linear dimension reduction (NLDR) techniques such as tSNE, UMAP, PHATE, PaCMAP, and TriMAP provide a low-dimensional representation of high-dimensional data by applying non-linear transformation. The methods and parameter choices can create wildly different representations, so much so that it is difficult to decide which is best, or whether any or all are accurate or misleading. NLDR often exaggerates random patterns, sometimes due to the samples observed, but NLDR views have an important role in data analysis because, if done well, they provide a concise visual (and conceptual) summary of high dimensional distributions. To help evaluate the NLDR we have developed a way to take the fitted model, as represented by the positions of points in 2D, and turn it into a high-dimensional wireframe to overlay on the data, viewing it with a tour. Viewing a model in the data space is an ideal way to examine the fit. One can see whether it fits the points everywhere or fits better in some places, or simply mismatches the pattern. It is used here to help with the difficult decision on which 2D layout is the best representation of the high-dimensional distribution, or whether the 2D layout is displaying mostly random structure. It can also help to see how different layouts made by different methods are effectively the same summary, or how the different methods have some particular quirks. This methodology is available in the R package quollr.
Nov 13, 2024
Jayani Lakshika

Using and contributing to the data.table package for efficient big data analysis

R Package
data.table
big data
data.table is one of the most efficient open-source in-memory data manipulation packages available today. First released to CRAN by Matt Dowle in 2006, it continues to grow in popularity, and now over 1500 other CRAN packages depend on data.table. This talk will start with data reading from CSV, discuss basic and advanced data manipulation topics, and finally will end with a discussion about how you can contribute to data.table.
Nov 6, 2024
Toby Hocking

Creating an R Package for Courtroom Transcript Survey Interface

R Package
courtr
Courtroom Transcript Survey
Many courtroom studies use a transcript format (with Q: and A: speaker prompts) for presenting information to participants. This format may not be engaging for participants and can lead to confusion in terms of speakers. We are developing an R package (‘courtr’) that follows the format of our previous research studies, which includes a text box, speech bubbles, and cartoon characters (we are currently designing a study to validate the use of these cartoon characters. You can view the study and provide feedback here: https://shiny.srvanderplas.com/characters/). In this presentation, I will go through the process of creating an R package, using ‘courtr’ as an example.
Oct 16, 2024
Rachel Rogers

Setting Up Real-Time Object Detection, Classification, Human Pose Estimation, and Open Vocabulary Object Detection In Minutes Using Ultralytics

Object Detection
Classification
Human Pose Estimation
Ultralytics
Computer vision has come a long way in the last decade, and now it is easier than ever to get started. Using the Ultralytics library, you can run pre-trained object detection, classification, human pose estimation, and open vocabulary object detection locally on your computer using only a few simple scripts. The entire process of downloading, installing, and running the models in real time takes only a few minutes, and each model is downloaded automatically, making this one of the fastest and easiest ways to get started with deep learning computer vision models. There are a range of model sizes to choose from depending on your hardware, but each model is designed to run with live video data. Efficient by design, most computers can use the real time features (albeit, possibly with a low frame rate), but the models can be used on downloaded images and videos as well. In this session, I will provide a short tutorial on installing Ultralytics and a set of scripts that can perform object detection, classification, human pose estimation, and open vocabulary object detection.
Oct 9, 2024
Carson Trego

Visualizing for Impact: the role of data visualization expertise in public policy research

Data visualization
Public policy
Applied statistics
Data visualization is a major component of the research process in applied statistics and data science work. Visual exploration of data helps researchers understand characteristics of a data set or model, explore distributions, identify patterns, and even test hypotheses. When research data will be used to make decisions about public policy, it is imperative to communicate data in a clear and accessible way: both decision-makers and the populations their decisions impact should be able to understand data that inform those decisions. This often means that the best way to visualize data may not be the way that researchers find most informative or interesting, but rather the way that will most effectively reach the right audience. In this talk, I’ll share some examples of effective visual communication for public policy and principles for maximizing the impact of your visuals for your audience. I will also discuss how these principles extend to a career in statistics and data science and share lessons learned about effective visual communication in the workplace.
Oct 2, 2024
Kiegan Rice

Let’s go down the rabbit hole!

Research Rabbit
Literature review
Zotero
Finding papers for literature reviews can be a tedious process. In this presentation, I will introduce Research Rabbit, a free application for finding connected papers and discovering networks of research areas. Research Rabbit allows you to create a collection of papers, follow citations/references, and more! With a Zotero integration tool, this application can help speed up the research process and make exploring papers more fun. I will give a live demonstration of Research Rabbit while we explore new papers together.
Sep 25, 2024
Tyler Wiederich

 

The Alchemy of AI: Sparking Creativity with Algorithmic Magic

AI
Machine learning
Explore the dynamic and rapidly evolving field of generative AI, exploring the latest tools, techniques, and advancements that are shaping its development. As generative AI continues to transform creative industries, scientific research, and technological solutions, understanding its capabilities and underlying mechanisms is paramount. We will examine current cutting-edge models that drive AI creativity, highlighting their applications in generating text, images, music, data visualization and more. By integrating examples from my own creative projects and research endeavors, this talk aims to showcase the practical implementations of generative AI and its potential to inspire new forms of artistic and scientific expression. Attendees will gain a comprehensive insight into how generative AI can be harnessed to enhance creativity and solve complex problems in various domains.
Sep 18, 2024
Dan Novy

Making Visualizations to Study Probability Theory

Visualizations
Probability Theory
Graduate programs in statistics usually involve a course or two in the theory of probability to understand the structure and inner workings of common statistical methods. These classes introduce unfamiliar abstract topics and mathematical machinery like measure theory and fourier analysis, which need to be understood at least partially to make the connection to statistics. Is it possible to make graphic visualizations to obtain a partial understanding of these topics? In this talk, I present a few simple animations made using the manim library (used in the popular 3Blue1Brown videos), some of which had to be constructed during my attempt to study Probability Theory over the last year.
Sep 11, 2024
Gautham Venkatasubramanian

Algorithmic assessment of striation similarity between wire cuts

Forensics
Data Visualizations
The comparative evaluation of aluminum wire cuts holds considerable significance within the field of forensic science. Nonetheless, there exists an absence of a systematic algorithmic framework for assessing their degree of similarity. In our recently introduced algorithm, we address this void by undertaking an examination of surface cuts presented in the x3p format.
Sep 21, 2023
Tom Yuhang Lin

The youdrawitR package - a GSOC experience

R package
GSOC
Shiny App
You Draw It’ is a feature that allows users to interact with a chart directly by drawing a line on their computer screen with a mouse. Originally introduced by the New York Times in 2015 for the purpose of interactive reading, we aimed to adapt and demonstrate the use of the ‘You Draw It’ method as a tool for interactive testing of graphics.
Sep 7, 2023
Dillon Murphy and Emily Robinson

Testing website designs and reports with screen readers

accessibility
websites
Designing websites (Quarto reports) and visualizations to be accessible becomes much more challenging, and rewarding, when you test your work using a screen-reader.
Apr 20, 2023
Ian Lyttle

A shiny Model Buildr for Psychophysics Data

psycho-metrics
shiny
model fitting
user experiment
User experiments are essential for informing researchers what an audience sees in a chart. User experiments are generally quite expensive in monetary value and in the time spent getting data. We must make the most out of the data we get from participants. Statistically, the best practice for data with repeated measurements is using (Generalized) Linear Mixed Effects Models (GLME).
Apr 20, 2023
Wangqian Ju

Jury Perception of Bullet Matching Algorithms and Demonstrative Evidence

forensics
user experiment
Subjective pattern comparison has been subject to increased scrutiny by the courts and by the general public, resulting in an increased interest in pattern comparison algorithms that provide quantitative assessments of similarity for use by forensic scientists. While these algorithms would mark an improvement over current subjective comparison methods, individuals without a statistical background may struggle with the statistical concepts and language necessary for describing algorithmic methods.
Apr 13, 2023
Rachel Rogers

 

plotscaper

R package
interactive graphics
Apr 6, 2023
Adam Bartoniczek

 

Striation similarity assessment between wire cuts with functional data analysis

Forensics
Automatic matching algorithms for assessing the similarity between striation marks have been investigated for bullet lands and some tool marks, such as screwdrivers. Here…
Mar 30, 2023
Yuhang (Tom) Lin

Quantifying Writer Variance Through Rainbow Triangle Graph Decomposition of the Common Word ‘the’

Shiny app
graphics design
Handwriting comparative analysis has recently been criticized based off the subjective nature of traditional examinations. To support traditional examination with objective measures, this project provides results from a study where features of handwriting are examined through point decomposition and rainbow triangulation. …
Mar 23, 2023
Alexandra Arabio

Displaying proportions and percentages

user experiments
paper discussion
We will discuss another paper by Ian Spence - this time we focus on the 1991 paper “Displaying proportions and percentages” co-authored with Stephan Lewandowsky. Again, the…
Mar 9, 2023
Heike Hofmann

 

Statistical graphics without R? – A quick look outside

graphics
Python
The grammar of graphics (Wilkinson 1999, 2005) provides a declarative, object-oriented framework to visualize statistical information as a collection of objects. It is noted…
Mar 2, 2023
Gautham Venkatasubramanian

Visual psychophysics of simple graphical elements

paper discussion
user experiments
We will discuss Ian Spence’s 1990 paper “Visual Psychophysics of simple graphical elements”. Make sure to read (or at least read over) the paper below.
Feb 16, 2023
Heike Hofmann

Evaluating Perceptual Judgements on 3D Printed Bar Charts

user experiments
It is well documented that the accuracy of data comparisons is worse for 3D charts, but these studies are almost entirely focused on 2D projections. In this week’s meeting…
Feb 9, 2023
Tyler Wiederich, UNL

Summarizing live traffic incidents in Iowa

Shiny app
Traffic incidents in Iowa are constantly reported through the Advanced Traffic Management System (ATMS). On average, over 100 traffic incidents are daily incidents are…
Feb 2, 2023
Guillermo Basulto, InTrans, ISU

The litr package - it’s turtles all the way down!

literate programming
R package
The litR R package lets you write a complete R package in a single R markdown document. This enables a workflow for writing R packages that is probably very different from what you are used to.
Nov 10, 2022
Jacob Bien, University of Southern California

Pizza Party! Crossover event of NUMBATs and Graphics Group

social
This week are going to have a global crossover shared meeting, with Iowa/Nebraska Graphics Group and Australian NUMBATs, with pizza! We are going to talk about favorite R…
Nov 10, 2022
GG & NUMBATs

The ggdensity package

R package
A popular strategy for visually summarizing bivariate data is plotting contours of an estimated density surface. Most commonly, the density is estimated with a kernel…
Nov 10, 2022
James Otto, Baylor University

Made with yarn, strings, and glue: Making R Markdown work better for you

watch-party
R package
R workflow
This week we will watch Alison Hill talk with the Cleveland R User group about making R Markdown a better integrated tool for your work. Alison Hill is a data scientist…
Oct 20, 2022
Alison Hill, RStudio (presenting indirectly)

A discussion of color ranges

visual communication
colors
This is a discussion of color spaces, optimizing color ranges for aesthetics, light and dark mode, while establishing accessibility and the associated compromises we have to…
Sep 29, 2022
Ian Lyttle

Testing Charts for Accuracy and Interpretation

visual communication
user experiments
The creation of a data visualization depends on both analytic design and graphic design. Analytic designs focus on the structure of a chart and how data are encoded onto…
Sep 23, 2022
Kiegan Rice

Using Python in R through the Reticulate Package and its Integration with R Shiny

Dashboards
Shiny App
Python
R package
When working on a project, sometimes it is necessary to use different programming languages to accomplish a task. The Reticulate Package in R creates a R interface to…
Apr 28, 2022
Alison Kleffner

Introduction to Dash (using Python) and Observable (Using JavaScript)

Dashboards
Shiny App
Python
JavaScript
If you have built a Shiny app, you have some familiarity with the concept of reactivity. For the next two graphics-groups, we’ll talk about two alternatives to Shiny which…
Apr 14, 2022
Ian Lyttle

Exploring Rural Shrink Smart Through Guided Discovery Dashboards

Dashboards
Shiny App
communication
Many small and rural places are shrinking. Interactive dashboards are the most common use cases for data visualization and context for exploratory data tools. In our paper…
Apr 7, 2022
Denise Bradford

Applying unsupervised and supervised learning methods to minimize risk to bald eagles from industrial wind turbines

Machine Learning
collaboration
spatial
In this talk I describe a collaboration with research wildlife biologists and statisticians to analyze over 2 million data points collected from GPS telemetry devices…
Mar 24, 2022
Silas Bergen

Going Analog: Emerging from the Weeds and Broadening our Audience

Data Art
teaching
communication
We could all use a break from our manuscripts (and our screens). Can thinking about our work in a new way help us re-energize? In this talk I’ll share some non-traditional…
Mar 10, 2022
Sara Stoudt

Automatic Image Captioning using Convolutional and Recurrent Neural Networks

Machine Learning
convolutional neural network
image processing
text analysis
python
Automatic image captioning is the process of generating a descriptive text description for an image. Image captioning is one of the few applications of deep neural networks…
Mar 3, 2022
Ved Piyush

Map matching in R

R package
maps
web scraping
API
Map matching is an interesting problem that involves snapping noisy GPS traces to a road network with a high degree of accuracy. It is used by ride-sharing services such as…
Feb 24, 2022
Ashirwad Barnwal

Intergalactic Time Travel: Estimation Pilot Study

visual inference
interactive graphics
User experiment
Questioning matters! The last graphical test for my research involves graph comprehension and extracting information from the chart. My talk will involve an interactive…
Feb 17, 2022
Emily Robinson
No matching items