In the First Year Immersion Program at Binghamton University, I am a part of the Image and Acoustic Signals stream working with a team of seven on a pre-proposal studying how machine learning can be applied to detecting misinformation on social media platforms. Our pre-proposal research focuses on detecting misinformation about Covid-19 on Twitter using a supervised learning algorithm called a support vector machine (SVM). Coupled with a text vectorizer, Term-Frequency Inverse Document Frequency (TF-IDF) or Count Vectorizer (CV), that quantifies data, data can be inserted into SVM where it can be accurately identified as true or false information. On average, our research found that SVM coupled with TF-IDF and CV has an accuracy rate of 89% and 87%, respectively. The lowest and highest accuracy value of SVM was 82% and 95.63%, respectively. Indeed, our research shows that TF-IDF is the better text vectorizer when working with SVM. In the future, we hope to research how different text vectorizes can be coupled with SVM and detecting misinformation on other social media platforms such as Facebook, Instagram, etc.

Organized and shelved books. Prepared weekly grab-and-go craft kits for children according to their age. Rewarded children for the hours they spend reading with prizes. Managed tasks such as weekly guesses for raffles, the reading program’s grand raffle, and various tasks to prepare for upcoming library programs.