Interdisciplinary – Page 6 – UROP Spring Symposium 2021

Interdisciplinary

How Algorithms are Reinforcing Oppressive views towards Black Women

The internet and search engines are used daily by most people in the U.S. However, most users are not familiar with the idea of algorithmic bias and how algorithms impact our communities and society. Algorithmic biases and the ethics of various technological algorithms can affect how gender, race, class systems, and many other categories are viewed and experienced from a societal lens. This project investigates the concepts of bias, discrimination, and ethics in current and emerging technological algorithms. This research will explore how Black Women are viewed from a societal and historical standpoint when searched for in various internet search engines and how these search results affect Black Women from an oppressive standpoint. From this investigation, educational YouTube content is created along with lesson plans based on the topic to be taught in the classroom. These lesson plans are taught in two separate classrooms, one in Detroit, MI with a majority African American community and the other in Ann Arbor, with a majority white population. An analysis of these two interactions along with reports of what has been learned about algorithmic bias and visuals gathered, awareness of the topic will be spread along with looking to find solutions to the algorithmic problems at hand. This research is able to provide a critical understanding of algorithms in technology. From this research one can see how bias, discrimination, and ethics show up in the media we constantly use and how these biases affect various societal communities. Minorities and social groups will also be able to heavily benefit from this research along with the companies that are creating these algorithms.

How Algorithms are Reinforcing Oppressive views towards Black Women Read More »

Sandy Hook Promise Evaluations

The goal of the Sandy Hook Promise is to improve the lives of students by increasing safety and comfort in both schools and the broader community. The program achieves this through the implementation of the Say Something Program, which encourages students to report troubling or abnormal behaviors to the administration before situations spiral out of control. To gather results for further analysis of the effectiveness of the Say Something and Sandy Hook Promise programs, numerous surveys have been conducted on students and teachers involved in the process. Furthermore, interviews have been conducted on administrators inquiring about the ease of implementation and use of the program throughout the school year. These results have been collected across some of the largest school districts in the U.S., including LA and Miami-Dade counties. Although the program is still in effect and results are being collected in real-time, certain conclusions about the effectiveness of the program can already be drawn with data from previous waves of students. Overall, after using STATA and SPSS software to conduct data analysis, it is evident that rates of depression and violence, along with feelings of unsafety, all decrease amongst students after the program has been present for over a year. This shows clear evidence that the program truly does have a positive impact on students’ safety. This research is extremely valuable, as protecting our future generations is amongst the top of our priorities, especially when they are in educational spaces. After numerous threats to student safety following school shootings such as Sandy Hook, programs such as this one have a chance to truly improve the school environment and the lives of students as they seek to better themselves through education.

Sandy Hook Promise Evaluations Read More »

Assessing Men’s Proclivity to Recognizing Subtle Gender Bias Against Women in STEM

Women in Science, Technology, Engineering, and Mathematics (STEM) frequently encounter gender bias (e.g., questioning of their STEM ability, assignment to secretarial roles). Given the subtle and ambiguous nature of contemporary sexism, people vary in their likelihood of recognizing subtly sexist interactions. Past research demonstrates that women are more sensitive to gender bias and more readily recognize it when it occurs. However, there remains a dearth of research related to men’s experiences in witnessing bias. In the present research, we ask: (1) what are the individual difference measures that contribute to men’s proclivity in recognizing subtle gender bias, (2) what are the affective consequences of recognizing subtle gender bias during group tasks, and (3) how do men’s affective states after witnessing subtle gender bias influence their desire to work with women in mixed-gendered groups? STEM identified men (N=275) read a fake transcript depicting a conversation between 3 STEM identified college students (1 woman, 2 men). Participants were exposed to one of two transcripts in which a man either (a) demonstrates subtle gender bias against a woman or (b) engages in a neutral interaction with a woman. After reading the transcript, participants completed measures related to their affect (state and collective), their impressions of the interaction (open and closed ended), and behavioral measures related to the students in the transcript. Open ended responses were coded to determine recognition of bias. Findings and implications for this work are discussed.

Assessing Men’s Proclivity to Recognizing Subtle Gender Bias Against Women in STEM Read More »

Analysis of Wheelchair Dimensions

The goal of this project was to characterize wheelchair dimensions to provide guidance to vehicle manufacturers who are designing integrated wheelchair seating stations in automated vehicles. UMTRI has a database of wheelchair crashes that include front and side view photos of hundreds of wheelchairs. My task on the project was to digitize specific wheelchair points using Image J software, calibrating each photo using a known scale dimension on each photo. These data can be used to define key dimensions for each wheelchair, such as maximum length, width, and height. Forty wheelchairs, including both manual and power styles, were analyzed. Results will be used to create generic 3-dimensional wheelchair models that represent the range of wheelchair sizes available.

Analysis of Wheelchair Dimensions Read More »

Understanding Dynamic Loading of Wheelchairs Secured in Vehicles During Crashes

To develop integrated wheelchair seating stations for automated vehicles (AVs), manufacturers need to understand the loading involved when securing a wheelchair to the vehicle, because wheelchairs can weigh much more than vehicle seats. The University of Michigan Transportation Research Institute (UMTRI) has conducted hundreds of dynamic sled tests of wheelchairs since the 1980s that include data on securement forces. These tests have involved manual, power, and stroller wheelchairs and a range of crash dummy sizes from small children to large adult males. This project first involved updating the wheelchair sled test database with the most recent test results, and then investigated factors that affect load levels. The main finding is that the weight of the wheelchair has the greatest correlation with the amount of floor loading. Results from this analysis will allow vehicle manufacturers to design wheelchair seating stations for AVs that will allow passengers who travel while seated in a wheelchair to ride safely and independently.

Understanding Dynamic Loading of Wheelchairs Secured in Vehicles During Crashes Read More »

3d Online Body Shape Model Development

In the automotive industry, there have been an increasing number of fatalities, easily preventable by building safer vehicles. In the medical field, it is almost impossible to thoroughly test accuracy of medical technology without using a physical model, which are oftentimes very expensive. This study involves the development of realistic user-manipulated 3D body models for usage in the medical, automotive, and engineering fields. Along with my mentor and fellow researcher, we have been building a nuanced body model application using Unity to be put out on the apple store and google play store. While results are inconclusive so far, as the application has not been released yet, it hopefully will facilitate the lives of engineers, researchers, and manufacturers working in a number of industries.

3d Online Body Shape Model Development Read More »

3d Online Body Shape Model Development

In the automotive industry, there have been an increasing number of fatalities, easily preventable by building safer vehicles. In the medical field, it is almost impossible to thoroughly test accuracy of medical technology without using a physical model, which are oftentimes very expensive. This study involves the development of realistic user-manipulated 3D body models for usage in the medical, automotive, and engineering fields. Along with my mentor and fellow researcher, we have been building a nuanced body model application using Unity to be put out on the apple store and google play store. While results are inconclusive so far, as the application has not been released yet, it hopefully will facilitate the lives of engineers, researchers, and manufacturers working in a number of industries.

3d Online Body Shape Model Development Read More »

Video Communication for Brain-Computer Interface Research

This project was to create an instructional video for the Direct Brain Interface Lab, which conducts Brain-Computer Interface research. This lab helps develop and test brain-computer interface devices that help the human brain communicate with software by translating brain activity into commands for electronic devices and computer systems. The video was created to improve communication between researchers in the lab and the lab participants using the brain-computer interface devices. The video format is especially helpful for participants with disabilities who are sometimes unable to understand conventional directions. The video explains how to use the device, and how the user needs to “think” the commands for the computer to respond with the appropriate output. This instructional video was created using videos from the lab’s database, including direct-feed videos of the brain-computer interface device’s output, and videos of people using the devices to create those outputs. These videos were chained together with Adobe Premiere Pro. The final video contained effects and a voiceover to help better communicate the instructions. This video will be used in the lab to help the participants better understand their brain-computer interface devices and how to use them correctly. The video will be updated periodically depending on how the participants respond and how the instructions change. The lab can now communicate their instructions in a video format, which will help better explain how to use the brain-computer interface.

Video Communication for Brain-Computer Interface Research Read More »

Enhanced MEAGA (Minimum distance-based Enrichment Analysis for Genetic Association)

The goal of the MEAGA (Minimum distance-based Enrichment Analysis for Genetic Association) is to help researchers understand complex traits derived from the correlation between genes. More specifically, this project aims at creating a network of genes with edge weights being their correlation with each other. This will allow researchers to test hypotheses between genetic pathways/functions and the correlation between specific genes. There currently exists a genetic network consisting of genes, but their connections/edge weights are binary (weight of 1 or 0). The goal right now is to convert the existing binary network into a continuous network, so the edges can have continuous values indicating the correlation between genes. The project focuses on graph algorithms such as Prim’s, Dijkstra’s, and Kou’s algorithm to create a genetic network that can create Steiner trees of indicated genes. Thus far, a gene to gene correlation file has been created with different correlation cutoffs of 0.3, 0.5, 0.7, and 0.9. This research can be valuable in identifying associations between genes and certain traits or diseases. This research may help the medical field when identifying the causes of conditions.

Enhanced MEAGA (Minimum distance-based Enrichment Analysis for Genetic Association) Read More »

Pollutant Data Prediction: Protecting our Watershed Through Sensing

Watershed health largely influences quality of life, but it is difficult to quantify. Tracking and reducing pollutants is one way to improve watershed health. Total suspended solids (TSS) is an invaluable measurement in determining the amount of pollutants in a body of water. Despite many studies relating TSS to turbidity (water opacity) or doppler backscatter (sound refraction), there is yet to be a way to accurately predict TSS on a long term basis. This study takes previously calculated correlations between suspended solids concentration (SSC) and doppler backscatter using linear regression and conducts a gaussian process on the data to determine a more accurate model. The model created is based on probabilistic computations with a 95% confidence interval. This model will require initial data points from a site before a model can be developed, unlike the previous model, but the confidence interval will decrease in size as more data points are added. Therefore, we will be able to determine a more accurate model for each specific site. The model developed in this study will then be used to predict TSS values for new sites as TSS and SSC are related measurements, and TSS is a more widely used measurement in regulations. This result allows us to continue with further research into the number of true TSS/ SSC samples necessary to determine an accurate model and how the model will react to changes in flow rate implemented by controlled valves.

Pollutant Data Prediction: Protecting our Watershed Through Sensing Read More »

lsa logoum logo