In the last 15 years, social media has become an integral part of communication, with the intersection of social media and social justice gaining prominence. This research examines how witnessing injustice on social media affects African Americans' participation in digital forms of activism, drawing on hashtag ethnography to analyze how hashtags serve as organizing tools and sites of collective expression in the digital age. It also builds on media witnessing theory, which suggests that viewing social injustices through digital media creates a morally grounded reaction and shapes audience responses, transforming passive spectators into active participants. Furthermore, the study engages with the framework of Black digital practice to explore how African Americans use digital tools and platforms to navigate, resist, and redefine the narratives of social justice in ways that are culturally specific and innovative. By conducting interviews and creating a digital platform as a site for data collection, this research seeks to identify a specific process that connects viral digital moments to social justice activism. In doing so, it provides insights into the dynamics of witnessing, hashtag activism, and the broader implications of digital engagement in fostering movements for racial justice.

This paper seeks to explore the use of LLMs – large language models for sentiment analysis in the quarterly earnings reports of Cummins Inc. This research leverages the Linq platform to predict the impact of press releases of earnings announcements on stock price changes. To carry out this research, data was gathered from Cummin’s press release site across four periods. The Linq platform was then used to test these unstructured data sources for sentiment analysis: positive, negative, or neutral. Price fluctuations were then assessed for 5 trading days before the earnings press release, on the day the press statement was released, and 5 trading days after the release of each earnings report and then compared to the sentiment predictions from Linq. The analyzed data in this paper shows some connection between the predictions made by the Linq platform and real-life market performance. This study emphasizes the value of LLMS in giving real-time financial insights. This paper also stresses the importance of artificial intelligence in financial applications while admitting the complex dynamics associated with financial markets and the need for more development of these AI models for broader applications.

Tracking maternal care behaviors in mice is essential for neuroscience research, particularly in understanding early-life stress. Traditional methods, such as manual tagging with BORIS, are resource-intensive and time-consuming, prompting the need for an automated solution. The objective of this project is to develop TailOR, an automated system that utilizes advanced computer vision techniques to tag specific maternal behaviors in mice models with minimal manual input. This system aims to streamline the process, improve tagging accuracy, and reduce human involvement. We employ SAM2 (Segment Anything Model 2) to generate masks for both the mouse and the nest in each frame, allowing for precise tracking of behaviors. By calculating the centroids of these masks, we monitor displacements across frames to identify patterns of motion. DeepLabCut is used for pose detection, and behavior classification is conducted through a Neural Network based on spatial data. Preliminary prototyping efforts and evaluation reveal that it is possible to detect and track mouse activities with practically usable accuracy. The project is ongoing, and the computer vision algorithm for behavioral classification is under development. Key obstacles include extended training times, low contrast, and partial visibility of pups. TailOR represents a significant advancement in the automation of maternal care behavior analysis in mice, offering a scalable solution that reduces human labor and enhances the accuracy of behavioral studies. This technology has potential applications beyond neuroscience, benefiting any field that requires detailed, reliable behavior tracking in animal models.

Generative artificial intelligence (GenAI) is a type of computer program that can generate new content such as text, images, or music, mimicking human creativity (e.g., ChatGPT). Despite frequent commentary on the topic in social and news media, there are few scales developed and validated to measure public attitudes on GenAI. It is important to understand public attitudes on GenAI because this can guide policy and regulation and inform better GenAI designs and implementation. Based on current literature, 79 survey items were created to measure the public’s attitudes, perceptions, and emotions toward GenAI. A convenience sample of 305 respondents was recruited online to complete the questionnaire. For this project’s aims, only the first scale, which contains 19 items reflecting participants’ attitudes toward using GenAI, was analyzed. Exploratory factor analysis was conducted to identify possible subscales of GenAI attitudes. A 3-factor model was selected to yield the most practical and meaningful estimation. The three factors were identified as acceptance of GenAI, impressiveness of GenAI, and concerns about GenAI. All three factors showed good internal validity. Validity testing for the scale was performed using previously validated scales for psychological stress (K6), attitudes about AI more broadly (AIAS-4), and dispositional optimism (LOT-R). Confirmatory factor analysis (CFA) showed acceptable discriminant validity, excellent concurrent validity, and very poor convergent validity. Limitations and future research directions are discussed.