Students Mine Parking Data to Help You Find a Spot

No parking spot? No problem.

A group of students has teamed up with Duke Parking and Transportation to explore how data analysis and visualization can help make parking on campus a breeze.

As part of the Information Initiative’s Data+ program, students Mitchell Parekh (’19) and Morton Mo (’19) along with IIT student Nikhil Tank (’17), spent 10 weeks over the summer poring over parking data collected at 42 of Duke’s permitted lots.

Under the mentorship of graduate student Nicolas-Aldebrando Benelli, they identified common parking patterns across the campus, with the goal of creating a “redirection” tool that could help Duke students and employees figure out the best place to park if their preferred lot is full.

A map of parking patterns at Duke

To understand parking patterns at Duke, the team created “activity” maps, where each circle represents one of Duke’s parking lots. The size of the circle indicates the size of the lot, and the color of the circle indicates how many people entered and exited the lot within a given hour.

“We envision a mobile app where, before you head out for work, you could check your lot on your phone,” Mo said, speaking with Parekh at the Sept. 23 Visualization Friday Forum. “And if the lot is full, it would give you a pass for an alternate lot.”

Starting with parking data gathered in Fall 2013, which logged permit holders “swiping” in and out from each lot, they set out to map some basic parking habits at Duke, including how full each lot is, when people usually arrive, and how long they stay.

However, the data weren’t always very agreeable, Mo said.

“One of the things we got was a historical occupancy count, which is exactly what we wanted – the number of cars in the facility at a given time – but we were seeing negative numbers,” said Mo. “So we figured that table might not be as trustworthy as we expected it to be.”

Other unexpected features, such as “passback,” which occurs when two cars enter or exit under the same pass, also created challenges with interpreting the data.

However, with some careful approximations, the team was able to estimate the occupancy of lot on campus at different times throughout an average weekday.

They then built an interactive, Matlab-based tool that would suggest up to three alternative parking locations based on the users’ location and travel time plus the utilization and physical capacity of each lot.

“Duke Parking is really happy with the interface that we built, and they want us to keep working on it,” Parekh said.

“The data team worked hard on real world challenges, and provided thoughtful insights to those challenges,” said Kyle Cavanaugh, Vice President of Administration at Duke. “The team was terrific to work with and we look forward to future collaboration.”

Hectic class schedules allowing, the team hopes to continue developing their application into a more user-friendly tool. You can watch a recording of Mo and Parekh’s Sept. 23 presentation here.

The team's algorithm recommends up to three alternative lots if a commuter's preferred lot is full. In this video, suggested alternatives to the blue lot are updated throughout the day to reflect changing traffic and parking patterns. Video courtesy of Nikhil Tank.

Kara J. Manke, PhD

Post by Kara Manke

 

Is Durham’s Revival Pricing Some Longtime Residents Out?

When a 2015 national report on gentrification released its results for the nation’s 50 largest cities, both Charlotte and Raleigh — North Carolina’s top two biggest cities — made the list.

The result was a collection of maps and tables indicating whether various neighborhoods in each city had gentrified or not, based on changes in home values and other factors from 1990 to the present.

Soon Durham residents, business owners, policy wonks and others will have easy access to similar information about their neighborhoods too, thanks to planned updates to a web-based mapping tool called Durham Neighborhood Compass.

Two Duke students are part of the effort. For ten weeks this summer, undergraduates Anna Vivian and Vinai Oddiraju worked with Neighborhood Compass Project Manager John Killeen and Duke economics Ph.D. student Olga Kozlova to explore real-world data on Durham’s changing neighborhoods as part of a summer research program called Data+.

As a first step, they looked at recent trends in the housing market and business development.

Photo by Mark Moz.

Durham real estate and businesses are booming. A student mapping project aims to identify the neighborhoods at risk of pricing longtime residents out. Photo by Mark Moz.

Call it gentrification. Call it revitalization. Whatever you call it, there’s no denying that trendy restaurants, hotels and high-end coffee shops are popping up across Durham, and home values are on the rise.

Integrating data from the Secretary of State, the Home Mortgage Disclosure Act and local home sales, the team analyzed data for all houses sold in Durham between 2010 and 2015, including list and sale prices, days on the market, and owner demographics such as race and income.

They also looked at indicators of business development, such as the number of business openings and closings per square mile.

A senior double majoring in physics and art history, Vivian brought her GIS mapping skills to the project. Junior statistics major Oddiraju brought his know-how with computer programming languages.

To come up with averages for each neighborhood or Census block group, they first converted every street address in their dataset into latitude and longitude coordinates on a map, using a process called geocoding. The team then created city-wide maps of the data using GIS mapping software.

One of their maps shows the average listing price of homes for sale between 2014 and 2015, when housing prices in the area around Duke University’s East Campus between Broad Street and Buchanan Boulevard went up by $40,000 in a single year, the biggest spike in the city

Their web app shows that more businesses opened in downtown and in south Durham than in other parts of the city.

Duke students are developing a web app that allows users to see the number of new businesses that have been opening across Durham. The data will appear in future updates to a web-based mapping tool called Durham Neighborhood Compass.

They also used a programming language called “R” to build an interactive web app that enables users to zoom in on specific neighborhoods and see the number of new businesses that opened, compare a given neighborhood to the average for Durham county as a whole, or toggle between years to see how things changed over time.

The Durham Neighborhood Compass launched in 2014. The tool uses data from local government, the Census Bureau and other state and federal agencies to monitor nearly 50 indicators related to quality of life and access to services.

When it comes to gentrification, users can already track neighborhood-by-neighborhood changes in race, household income, and the percentage of households that are paying 30 percent or more of their income for housing — more than many people can afford.

Vivian and Oddiraju expect the scripts and methods they developed will be implemented in future updates to the tool.

When they do, the team hopes users will be able to compare the average initial asking price to the final sale price to identify neighborhoods where bidding has been the highest, or see how fast properties sell once they go on the market — good indicators of how hot they are.

Visitors will also be able to compare the median income of people buying into a neighborhood to that of the people that already live there. This will help identify neighborhoods that are at risk of pricing out residents, especially renters, who have called the city home.

Vivian and Oddiraju were among more than 60 students who shared preliminary results of their work at a poster session on Friday, July 29 in Gross Hall.

Vivian plans to continue working on the project this fall, when she hopes to comb through additional data sets they didn’t get to this summer.

“One that I’m excited about is the data on applications for renovation permits and historic tax credits,” Vivian said.

She also hopes to further develop the web app to make it possible to look at multiple variables at once. “If sale prices are rising in areas where people have also filed lots of remodeling permits, for example, that could mean that they’re flipping those houses,” Vivian said.

Data+ is sponsored by the Information Initiative at Duke, the Social Sciences Research Institute and Bass Connections. Additional funding was provided by the National Science Foundation via a grant to the departments of mathematics and statistical science.

groupshot

 

 

 

 

Writing by Robin Smith; video by Sarah Spencer and Ashlyn Nuckols

What Makes a Face? Art and Science Team Up to Find Out

From the man in the moon to the slots of an electrical outlet, people can spot faces just about everywhere.

As part of a larger Bass Connections project exploring how our brains make sense of faces, a Duke team of students and faculty is using state-of-the-art eye-tracking to examine how the presence of faces — from the purely representational to the highly abstract — influences our perception of art.

The Making Faces exhibit is on display in the Nasher Museum of Art’s Academic Focus Gallery through July 24th.

The artworks they examined are currently on display at the Nasher Museum of Art in an installation titled, “Making Faces: At the Intersection of Art and Neuroscience.”

“Faces really provide the most absorbing source of information for us as humans,” Duke junior Sophie Katz said during a gallery talk introducing the installation last week. “We are constantly attracted to faces and we see them everywhere. Artists have always had an obsession with faces, and recently scientists have also begun grappling with this obsession.”

Katz said our preoccupation with faces evolved because they provide us with key social cues, including information about another individual’s gender, identity, and emotional state. Studies using functional Magnetic Resonance Imaging (fMRI) even indicate that we have a special area of the brain, called the fusiform face area, that is specifically dedicated to processing facial information.

The team used eye-tracking in the lab and newly developed eye-tracking glasses in the Nasher Museum as volunteers viewed artworks featuring both abstract and representational images of faces. They created “heat maps” from these data to illustrate where viewers gazed most on a piece of art to explore how our facial bias might influence our perception of art.

This interactive website created by the team lets you observe these eye-tracking patterns firsthand.

When looking at faces straight-on, most people direct their attention on the eyes and the mouth, forming a triangular pattern. Katz said the team was surprised to find that this pattern held even when the faces became very abstract.

“Even in a really abstract representation of a face, people still scan it like they would a face. They are looking for the same social information regardless of how abstract the work is,” said Katz.


A demonstration of the eye-tracking technology used to track viewers gaze at the Nasher Museum of Art. Credit: Shariq Iqbal, John Pearson Lab, Duke University.

Sophomore Anuhita Basavaraju pointed out how a Lonnie Holley piece titled “My Tear Becomes the Child,” in which three overlapping faces and a seated figure emerge from a few contoured lines, demonstrates how artists are able to play with our facial perception.

“There really are very few lines being used, but at the same time it’s so intricate, and generates the interesting conversation of how many lines are there, and which face you see first,” said Basavaraju. “That’s what’s so interesting about faces. Because human evolution has made us so drawn towards faces, artists are able to create them out of really very few contours in a really intricate way.”

IMG_8354

Sophomore Anuhita Basavaraju discusses different interpretations of the face in Pablo Picasso’s “Head of a Woman.”

In addition to comparing ambiguous and representational faces, the team also examined how subtle changes to a face, like altering the color contrast or applying a mask, might influence our perception.

Sophomore Eduardo Salgado said that while features like eyes and a nose and mouth are the primary components that allow our brains to construct a face, masks may remove the subtler dimensions of facial expression that we rely on for social cues.

For instance, participants viewing a painting titled “Decompositioning” by artist Jeff Sonhouse, which features a masked man standing before an exploding piano, spent most of their time dwelling on the man’s covered face, despite the violent scene depicted on the rest of the canvas.

“When you cover a face, it’s hard to know what the person is thinking,” Salgado said. “You lack information, and that calls more attention to it. If he wasn’t masked, the focus on his face might have been less intense.”

In connection with the exhibition, Nasher MUSE, DIBS, and the Bass Connections team will host visiting illustrator Hanoch Piven this Thursday April 7th and Friday April 8th  for a lunchtime conversation and hands-on workshop about his work creating portraits with found objects.

Making Faces will be on display in the Nasher Museum of Art’s Academic Focus Gallery through July 24th.

Kara J. Manke, PhD

Post by Kara Manke

The Art of Asking Questions at DataFest 2016

During DataFest, students engaged in intense collaboration. Image courtesy of Rita Lo.

Students engaged in intense collaboration during DataFest 2016, a stats and data analysis competition held from April 1-3 at Duke. Image courtesy of Rita Lo.

On Saturday night, while most students were fast asleep or out partying, Duke junior Callie Mao stayed up until the early hours of the morning pushing and pulling a real-world data set to see what she could make of it — for fun. Callie and her team had planned for months in advance to take part in DataFest 2016, a statistical analysis competition that occurred from April 1 to April 3.

A total of 277 students, hailing from schools as disparate as Duke, UNC Chapel Hill, NCSU, Meredith College, and even one high school, the North Carolina School of Science and Mathematics, gathered in the Edge to extract insight from a mystery data set. The camaraderie was palpable, as students animatedly sketched out their ideas on whiteboard walls and chatted while devouring mountains of free food.

Callie Mao ponders which aspects of data to include in her analysis.

Duke junior Callie Mao ponders which aspects of the data to include in her analysis.

Callie observed that the challenges the students faced at DataFest were extremely unique: “The most difficult part of DataFest is coming up with an idea. In class, we get specific problems, but at DataFest, we are thrown a massive data set and must figure out what to do with it. We originally came up with a lot of ideas, but the data set just didn’t have enough information to fully visualize though.”

At the core, Callie and her team, instead of answering questions posed in class, had to come up with innovative and insightful questions to pose themselves. With virtually no guidance, the team chose which aspects of the data to include and which to exclude.

Another principal consideration across all categories was which tools to use to quickly and clearly represent the data. Callie and her team used R to parse the relevant data, converted their desired data into JSON files, and used D3, a Javascript library, to code graphics to visualize the data. Other groups, however, used Tableau, a drag and drop interface that provided an expedited method for creating beautiful graphics.

Mentors assisted participants with formulating insights and presenting their results

Mentors assisted participants with formulating insights and presenting their results. Image courtesy of Rita Lo.

On Sunday afternoon, students presented their findings to their attentive peers and to a panel of judges, comprised of industry professionals, statistics professors from various universities, and representatives from Data and Visualization Services at Duke Libraries. Judges commended projects based on aspects such as incorporation of other data sources, like Google Adwords, comprehensibility of the data presentation, and the applicability of findings in a real industry setting.

Students competed in four categories:  best use of outside data, best data insight, best visualization, and best recommendation. The Baeesians, pictured below, took first place in best outside data, the SuperANOVA team won best data insight, the Standard Normal team won best visualization, and the Sample Solution team won best recommendation. The winning presentations will be available to view by May 2 at http://www2.stat.duke.edu/datafest/.

Bayesian, the winner of the Best Outside Data category

The Baeasians, winner of the Best Outside Data category at DataFest 2016: Rahul Harikrishnan, Peter Shi, Qian Wang, Abhishek Upadhyaya. (Not pictured Justin Wang) Image courtesy of Rita Lo.

 

By student writer Olivia Zhu  professionalpicture

When the Data Get Tough, These Researchers Go Visual

Ever wondered what a cleaner shrimp can see?

Or how the force of a footstep moves from particle to particle through a layer of sand?

How about what portion of our renewable energy comes from wind versus solar power?

The winning submission, created by Nicholas School PhD candidate Brandon Morrison, illustrates the flow of agricultural and forestry crops from raw materials to consumer products. The colors correspond to the type of crop – brown for wood, green for vegetables, etc. – and the width of the lines correspond to the quantity of the crop. You can check out the full image and caption on the Duke Data Visualization Flickr Gallery.

The winning submission, created by Nicholas School PhD candidate Brandon Morrison, illustrates the flow of agricultural and forestry crops from raw materials to consumer products. The colors correspond to the type of crop – brown for wood, green for vegetables, etc. – and the width of the lines correspond to the quantity of the crop. You can check out the full image and caption on the Duke Data Visualization Flickr Gallery.

The answers to these questions and more are stunningly rendered in the entries to the 2016 Student Data Visualization Contest, which you can check out now on the Duke Data Visualization Flickr Gallery.

“Visualizations take advantage of our powerful ability to detect and process shapes to reveal detailed trends that you otherwise wouldn’t be able to see,” said Angela Zoss, Data Visualization Coordinator at Duke Data and Visualization Services (DVS), who runs the contest. “This year’s winners were all able to take very complex topics and use visualization to make them more accessible.”

One winner and two finalists were selected from the 14 submissions on the basis of five criteria: insightfulness, broad appeal, aesthetics, technical merit, and novelty. The submissions represent data from all areas of research at Duke – from politics and health to fundamental physics and biology.

“This year’s entrants showed a lot of sophistication and advanced scholarship,” Zoss said.  “We’re seeing more advanced graduate work and multi-year research projects that are really benefiting from visualization.”

Eric Monson, a Data Visualization Analyst with DVS, hopes the contest will inspire more students to consider data visualization when grappling with intricate data sets.

“A lot of this work only gets shared within courses or small academic communities, so it’s exciting to give people this opportunity to have their work reach a broader audience,” Monson said.

Posters of the winning submissions will soon be on display in the Brandaleone Lab for Data and Visualization Services in The Edge on the first floor of Bostock Library.

The second-place entry, by Art History PhD student Katherine McCusker, depicts an archaeological site in Viterbo, Italy. The colored lines indicate the likely locations of buried structures like walls, platforms, and pavement, based on an interpretation of data from ground-penetrating radar (represented by a dark red, yellow, white colormap). You can check out the full image and caption on the Duke Data Visualization Flickr Gallery.

The second-place entry, by Art History PhD student Katherine McCusker, depicts an archaeological site in Viterbo, Italy. The colored lines indicate the likely locations of buried structures like walls, platforms, and pavement, based on an interpretation of data from ground-penetrating radar (represented by a dark red, yellow, white colormap). You can check out the full image and caption on the Duke Data Visualization Flickr Gallery.

Kara J. Manke, PhD

Post by Kara Manke

 

Geography and the Web: A new frontier for data vizualization

A GIS Day earth cake made by the Collegiate Baker

You might be forgiven if you missed GIS Day at The Levine Science Research Center Nov. 18, but it was your loss. Students and faculty enjoyed a delightful geography-themed afternoon of professional panels, lightning talks, and even a geospatial research-themed cake contest.

What is GIS and why is it important?

Geographic information systems (GIS) give us the power to visualize, question, analyze, and interpret data to understand relationships, patterns, and trends in the world around us. Those who work with data and analytics have a responsibility to contribute to this change by helping us make the right decisions for our future. As noted during ESRI’s 2015 User Conference in the video below, “We have a unique ability to impact and shape the world around us. [Yet] for all of our wisdom, our vast intellectual marvels, we still choose a path of unsustainability and continue to make decisions that negatively impact the Earth and ourselves. […]We must accept our responsibility as stewards of the Earth. […] We must apply our best technology, our best thinking, our best values. Now is the time to act. Now is the time for change.”

 

How does GIS help?

Doreen Whitley Rogers, Geospatial Information Officer for the National Audubon Society, led a lively discussion about GIS and the World Wide Web at Duke’s GIS Day. She said GIS is essential to understand what is happening in the geographic space around us. As GIS becomes increasingly web-based, efficiently distributing the system to other people is crucial in a time when new data about the environment is being created every second.

3D map displaying the height of buildings that birds hit windows

3D map displaying the height of buildings at which birds fly into windows in Charlotte, NC

Rogers and her team are aiming to move authoritative GIS data to web for visualizations and create a centralized system with the potential to change our culture and transform the world. As the technology manager, she is working on bringing the information to people with proper security and integrity.

In order to get people to use GIS data in a generalized way, Rogers needed to implement several core capabilities to assist those integrating GIS into their workflow. These include socializing GIS as a technology to everybody, creating mobile apps to work with data in real time, and 3D maps such as this one of bird-strikes in downtown Charlotte.

Case Studies

ClimateWatch helps us predict the seasonal behaviour plants and animals.

Mobile apps connecting to the GIS platform promise a strong “return on mission” due to the vast number of people using maps on phones. By mobilizing everyone to use GIS and input data about birds and geography in their area, the platform quickly scales over millions of acres. In the Bahamas, an  app allows users to take pictures to support bird protection programs.

ClimateWatch is an app that gives us a better understanding of how bird habitats are affected during temperature and rainfall variations – motivating people to speak up and act towards minimizing anthropogenic climate change. Developed by Earthwatch with the Bureau of Meteorology and The University of Melbourne, the app enables every Australian to be involved in collecting and recording data to help shape the country’s scientific response to climate change.

Virtual simulation of scenic flights as an endangered bird.

Virtual simulation of scenic flights from the perspective of an endangered bird.

Apps such as the 3-D flight map give users the vicarious thrill of cruising through nature landscapes from the view of endangered birds.

With the movements toward cleaning air and water in our communities, our planet’s birds will once again live in healthier habitats. As the Audubon Society likes to say: “Where birds thrive, people prosper.”

 

 

 

For more information about bird-friendly community programs, you can visit Audubon‘s site or send them a message.

Doreen Rogers after her presentation on National GIS day.

 

 

To learn more about data visualization in GIS, you can contact Doreen Whitley Rogers via email here.

Anika_RD_hed100_2

Post by Anika Radiya-Dixit