Jane Abernathy

Jane Abernathy

Project title:

Understanding Confirmation Bias in Visual Analytics Systems

Project description: The Charlotte Visualization Center is conducting a study on confirmation bias, or the tendency people have to believe or value evidence that supports their current ideas, on decisions made with the help of visual analytics systems, or systems that analyze data and display the information to users. The lab has made a visual analytics system, Verifi, which shows information about Twitter accounts that people can use to decide if Twitter accounts produce real or fake news. I am helping with designing and implementing this study by processing data, producing training videos for Verifi, and conducting some of the trials.

Blog:

July 28

This last week I’ve just been preparing my extended abstract and poster and running user studies for the lab. Its been a great summer overall. I’ve tried a lot of different things and have gotten experience working in a research environment. The most challenging part was not knowing what to expect when I started and taking a while to get into the swing of things. If your thinking of doing research here at Charlotte you should definitely apply.

July 21

The user studies for the lab’s new experiment started this week, and I’ve been running some of them. Since we’ve finally made all the parts that go into the study, running them is just getting people to answer questionnaires and sign forms in the right order and being there to answer any questions they have about the interface. It’s neat to see and participate in the whole process. I’ve also been working on my extended abstract and my poster which are coming along well.

My favorite thing that I did this summer was make the training videos for the user study. It was hard to take the very complicated interface that we built and try to explain all of it clearly and briefly, but I feel that I really understand it better now. Working with video editing software was new to me, but I’m really glad that I have had the experience. The hardest thing that I’ve done so far has been reading academic papers and trying to understand the concepts and apply them.

July 15

This was the last week before the Vis Center starts their experiment, so there was a lot to be done. The visual interface got its final touches and we tried out the task to see how long it would take. I made two training videos to show the users how to use the interface. I wrote a script that explained every aspect of the interface and gave an example of using it. Of course we had to cut it down so the video wouldn’t be too long. Then I took a screen capture of using the interface and recorded the voice over. I used software called Camtasia to edit the sound and video together and add effects to emphasize what I was explaining. It was really exciting to do some video editing, since I haven’t before, and I was really glad to contribute something to the experiment. It’s startling to realize that there are only two weeks left, and that I need to crack down on working on my extended abstract and poster.

July 7

I hope everyone had a good 4th of July! I had a good time relaxing, and it made the week feel a lot shorter than usual. In the lab this week the visual interface for the next experiment has really been coming together. Its been really interesting to see the process of planning the experiment and what needs to be done for it to take place. I’m glad that I’ve been able to help review the interface and right now I’m working on making a training video to explain the interface to the users. I’ve also been discussing the next steps of my project, finding evidence of users though patterns in the user log data from the last visual interface experiment. There’s a whole world of ways to represent and think about data, and its interesting to see even just a part of it.

June 30

This week I exhausted all of the resources available to help me get connected to MongoDB, and still no success. I must have a curse that effects every computer I touch. So, I’ll have to leave the rest of the database cleaning to Kayvan. That means I can focus more on my individual research. I have looked through user logs and am brainstorming ways that I can display it that will show the user’s thought process. In order to display the data, I’ve been learning R. I’m really glad to have a chance to try it and actually apply it to something I’m doing.

Mistakes are always a part of learning something new. If you make a mistake when your writing code, then you learn one way not to solve that problem, which leads you closer to a method that works. So, of course, I’ve been making mistakes, both in coding and in figuring out how to approach the problems overall. I’m sure that will everyone’s help I’ll keep being able to learn from my mistakes in a productive way while I continue my research.

June 23

I’ve spent a lot of time this week trying to set up all the systems I need on various computers. I had problems connecting to MongoDB on my laptop, so I was assigned a lab computer. MongoDB works fine on that computer, but I wasn’t able to install Jupyter in-order to run the python code I needed. I’m not sure what we’re going to do next, maybe find another computer to use, but everyone has been really helpful.

I’ve been helping to clean up the Tweets that the lab will be using for their next experiment. I also started the process of looking at user data from a previous experiment the lab did in-order to find patterns in the thought process of the users. To prepare for this I’ve been reading a lot of papers on the topic of sensemaking, and its a very interesting area of research.

June 16

This week I’ve really started to work on the Literature Review. Though some of the papers are pretty dense, but its been interesting to see some of the academic computer science work that is out there. I also wrote some code to filter the Tweets that are going to be used in the project. As far as problems, my swipe card is finally fixed! It took many trips to ITS and the card office, but we figured out that I was given two accounts. I got a card connected to the right one, and now I can (and have) be been given access to the lab. I’ve been having some problems downloading and using all the new systems and platforms that my team is using to work on their project. Though some systems have worked perfectly, a few have been difficult. Though looking up error messages and troubleshooting a system that your just being introduced to is hard, it forces you to learn a lot quickly. However, everyone has been very helpful in working the problems out.

June 9

In this second week I’ve been getting to know everyone in the lab and settling in. I’m still in the learning stage, so I’ve been doing a lot of tutorials and practice before I get my future work. However, both Kayvan and I have helped review twitter accounts to find good candidates for experiment on fake news and wrote a program to filter and count the tweets we have. We also just started a literature review of relevant papers. Except for the continuing swipe card access problems (I’m getting help on it, don’t worry), things seem to be falling into place. I’m looking forward to having a great time this weekend, with the whitewater trip and watching the Tony Awards.

June 2

This first week of the REU has both busy and exciting. There was a lot of new information to process with moving in, finding my way around, and meeting all of the students and professors, but it was great to learn more about the facility I’m going to be working in. I met Professor Dou and heard about two of the projects going on in the visualization lab. We sat in on a meeting about different ways to analyze and display information about UNCC students and how likely they are complete their degree. We also heard about and started some work on a project that will examine how confirmation bias effects people’s perception of fake news on Twitter.

During my summer of research I hope to acquire new technical skills I don’t currently have. I don’t have a specific list, but I know that anything I learn will be useful in the future. I also want to learn more about what research in computer science is like, both overall and on a day-to-day basis. I’m excited to contribute to a large project and learn more about presenting my work by creating a poster. I’m looking forward to this being a great summer.