Plastic Waste Awareness Mobile Application; Understanding the Socio-Political Causes and Effects of Bias in Data 286 views

As we increasingly become more data reliant, data increasingly establishes an important role in our life. This motivated my research question to ask what are the causes and effects of biased data in systems and applications. The current understanding in the field is that bias usually occurs during the beginning of the data lifecycle from possible underlying political, religious, or social biases that social actors may have that affect their judgment towards bias. It is also known that outliers and missing data in datasets can offset the other values creating bias, however, different solutions to this issue are simple and applied normally. An example of a possible solution would be to delete outlier values and fill missing values with the average of their respective columns. Even though it is best practice to conduct a validation step before the data gets processed, as we have seen, that is not always effective in detecting and eliminating harmful bias. Therefore I concentrated on building a framework around the general data lifecycle as a result of my findings during the past few weeks. This allows social actors to follow my framework for every step of the data lifecycle and ideally prevent bias from entering the system. As for my technical portion of my research, I am teaming up with a team of the Earth Ambassadors to develop a cross-platform mobile application for primary level students in Jamaica, which ranges from ages 7 to 9, to learn about the effects of plastics on the environment. The application is meant not only to inform students but also allows the students to integrate safe plastic decisions into their daily lives. To do this we have created an activities tab in the app that allows students to participate in fun and educational activities regarding plastic. We also intend to add a feature that allows students to take pictures of areas where garbage needs to be picked up and that picture will get its GPS coordinates geolocated on a map that allows other users to see where the garbage areas are. This mobile application is developed in React Native with a Firebase database integration. “The technical and STS theses are not related”

BS (Bachelor of Science)

Databases; Bias; Data; Data Lifecycle; Fairness; Large Techincal Systems; LTS; React Native; Firebase; Mobile Application; Machine Learning; Plastic Waste; Educational Application

School of Engineering and Applied Science Bachelor of Science in Computer Science Technical Advisor: Daniel Graham STS Advisor: Sharon Ku Technical Team Members: Andrea Clayton

English

All rights reserved (no additional license for public reuse)

2021-05-13