University of Massachusetts Boston
University of Massachusetts Boston: ScholarWorks at UMassNot a member yet
8618 research outputs found
Sort by
The Sankofa Study: Preliminary Feasibility and Outcomes of an Afrocentric Racism-Related Stress Intervention
No full textRacism has consistently been found to prompt adverse mental health outcomes among Black Americans (Paradies, 2006; Pieterse et al., 2012). Given that Afrocentric values are considered a contributor to the mental health and wellness of descendants of the African diaspora in African-centered psychology, an Afrocentric racism-related stress intervention may be uniquely beneficial for Black Americans (Belgrave & Allison, 2019). This study sought to explore the preliminary feasibility and outcomes of a proposed six-week Afrocentric group intervention for racism-related stress in a sample of four Black American adults. Data from study tracking, participant feedback, and group facilitator feedback were used to assess feasibility. Overall, the proposed Afrocentric group intervention appeared to be an acceptable intervention that could be successfully utilized with Black American adults experiencing racism-related stress. Despite initial challenges, recruitment was deemed to be feasible and found to be most effective via social media. Retention, data collection, resources, and ability to manage the intervention were all determined to be feasible. Preliminary outcome data showed significant decreases in cultural and institutional racism-related stress and perceptions of others as dangerous one-month post-intervention. There were no significant differences in self-reported depressive symptoms, anxious arousal, general anxiety, quality of life, Africentric values, racial pride, cognitive/emotional debriefing, spiritual-centered coping, collective coping, ritual-centered coping, awareness and relational resistance, or participation in resistance activities and organizations one-week or one-month post-intervention, although some changes were of medium to large effects, and may have been significant in a larger sample. Findings from this study support the feasibility of the proposed intervention and warrant additional pilot studies to further refine the intervention and assess efficacy
Muscat, Madrid, Ulster, and the Holy Land: The MEDRC Model of Environmental Peacebuilding in a Revived Middle East Peace Process
Get PDFMandated to assist the Middle East peace process through environmental diplomacy, MEDRC, the last surviving institution of that process, has survived through an institutional and operational approach to conflict resolution separate from the rest of the process. Understanding its transferable approach is important in fields of environmental diplomacy and conflict resolution not only in the context of combating transboundary climate and environmental threats but of using these threats as entry points into a peace process. As the international community grapples with the need for a credible solution to the intractable conflict in Israel and Palestine, the MEDRC approach has implications for the process design of a revived and reformed Middle East peace process. The aim of this article is to present for the first time the detailed elements of the MEDRC Model and underpinning Conflict Resolution Process Guidelines, and to examine implications for environmental peacebuilding in general and a for a revived Middle East peace process
Pioneering the Digital Frontier: CMI\u27s Approach to Forward-Looking Dialogues
Get PDFAs contemporary conflicts grow increasingly complex, new approaches to peacemaking are needed. This article outlines how CMI – Martti Ahtisaari Peace Foundation (CMI) incorporates technology-enhanced foresight methodologies into its dialogue and mediation work. Digital tools, such as software dedicated to data analysis and visualization, play a key role in CMI’s foresight approach by facilitating broad-based data collection and participatory analysis. Interactive visual aids foster collective sense-making and help challenge entrenched mindsets of conflict stakeholders. The article illustrates how foresight approaches can be used to develop shared future visions and facilitate collaboration even in the context of stalled peace processes
Realize that Students are Capable: Perspectives of Students with Intellectual Disability on their College Experiences
Get PDFThe Think College Inclusive Higher Education Network conducted a brief literature review of articles that included the voices and perspectives of college students or college graduates with intellectual disability. The aim of this review was to identify common program elements or experiences that students reflected on positively and/or could be improved upon. In this Insight Brief, we summarize the findings from our research review. We end with a summary of recommendations made by or based on the perspectives of students with intellectual disability that can be used to improve postsecondary education programs
Introduction to the Special Issue
Get PDFThis issue of the New England Journal of Public Policy, with Lord John Alderdice as the guest editor, examines how, with the advent of sophisticated technologies and AI, the conduct of wars and peacemaking in the opening decades of the twenty-first century has changed, with implications for the future of both and society at large
Orienting Volunteers on Cultural Awareness and Empathy While Working With Refugees
Get PDFIt is imperative to recognize that effective volunteer engagement in refugee resettlement hinges upon a deep understanding of cultural dynamics and sensitivities. This paper explores the challenges and complexities of volunteer engagement with refugees, offering insights from the author\u27s experiences, scholarly literature, and theoretical frameworks. The writing highlights the importance of cultural awareness, empathy, and reflective practice in supporting refugees during resettlement while acknowledging the need for ongoing learning and evaluation of volunteer approaches. Relevant concepts and theories from the Critical and Creative Thinking courses are included, integrating perspectives into the author’s analysis of volunteer experiences. Additionally, the paper incorporates metacognitive questions and reflective practices for volunteers, encouraging continuous learning and personal growth. The framework seeks to empower volunteers to navigate complicated interpersonal dynamics and build meaningful connections with refugee families by fostering self-awareness, curiosity, and empathy
Racial Disparities in SNAP Receipt for Eligible Asian Americans in Massachusetts
Get PDFDespite qualifying as income eligible, many Massachusetts families do not access SNAP (Supplemental Nutrition Assistance Program) benefits. Due to the sharp increase in the cost of living, especially the cost of housing and food expenses, more families are facing food insecurity. Thus, it is critical to ensure that families in need receive SNAP benefits. While previous studies have examined racial disparities, there is a limited focus on Asian American families. Even fewer studies disaggregate data to explore disparities among Asian American ethnic subgroups. Further, few studies have addressed disparities in SNAP receipt specifically for income eligible families.
The purpose of this study was to identify racial disparities and disparities in SNAP receipt among Asian American subgroups for income eligible families. This study also explored the extent to which English fluency, immigration status, education level, and employment status determine recipience of SNAP benefits in Massachusetts. American Community Survey 5-year data (ACS 2016-2020) were used to analyze racial-ethnic disparities in receiving SNAP among income eligible families in Massachusetts (those with total incomes at or below 200% of the Federal Poverty Level (FPL)).
The results revealed that among families that are income eligible to receive SNAP benefits, Asian American and White families generally were not different in SNAP receipt. However, income eligible Asian American families with a bachelor’s degree were more likely than White families to receive SNAP. The interaction terms also indicated that even at different levels of education, income eligible Asian American families tend to be less likely to receive SNAP than Black and Hispanic families. The analysis of subgroups revealed that income eligible Cambodian families are more likely than income eligible Chinese families to receive SNAP. These results suggest the counterintuitive recommendation that policymakers should pay closer attention to Asian Americans with higher education levels, who may be reluctant to access SNAP benefits even when food insecure. The report concludes with additional implications for policy and directions for future research
Sequential GBB and Pictet-Spengler Reactions for the Synthesis of Novel Polyheterocyclics
Get PDFOver the last twenty years, the Groebke-Blackburn-Bienaymé (GBB) reaction has emerged as a powerful multicomponent reaction to access various nitrogen-based heterocycles that have potential biological activities. In this works, the three-component reactions of indole 3-carbaldehydes with suitable isocyanides and aromatic amines afford the GBB adducts that then further react with aldehydes for the regio-selective Pictet-Spengler (PS) reaction to afford seven-member instead of other ring-size polyheterocyclic compounds. The product structures were confirmed by x-ray single crystal analysis. Both the GBB and PS reactions went smoothly to afford the product analogs with variations of X, R1 and R2
Learning Semantic Representations for Natural Language Processing
No full textRecent developments in machine learning, especially deep learning, have significantly advanced the progress in Natural Language Processing (NLP). In order to train machines to understand natural languages, learning informative semantic representations from them is a fundamental and crucial step. Although there has been impressive development in semantic representation learning, key challenges still remain. In this dissertation, we address some of these challenges by proposing novel approaches to learning representations at word and sentence levels and developing an automatic short answer grading system that applies the techniques in NLP in the education field. The first section of this dissertation explores semantic representation learning at the word level, which is treated as an undividable atomic information unit. Word embeddings which are dense vectors have been popularly applied in current machine learning in NLP. We propose a novel modular neuro-symbolic approach to learn richer semantic information such as denotation information. It designs a small neural network for each word, treats such representation of a word as a module, utilizes the symbolic information of the dependency parsing tree, and connects word modules to construct a neural network for a sentence which will be trained on some sentence level NLP tasks. Experiments on a Linguistic Acceptability task are run to test this approach’s potential in learning informative semantic representations with much less training data and a much smaller model size. The second section shows our work on representation learning at the sentence level. We develop a framework that applies contrastive learning and utilizes publicly labeled Natural Language Inference corpora to improve the learning of sentence representations. This framework is model agnostic which can be applied on top of any existing encoders. By using BERT as the encoder, experiments on the series of Text Similarity tasks prove this simple yet effective approach. In the last section, we present SteLLA, a structured automatic grading system using Large Language Models (LLMs) with Retrieval-Augmented Generation. This system also applies other NLP techniques such as question generation and answering to provide structured grades and feedback. We experiment with it on a real-world dataset from college-level Biology course exams and show that our grading system can achieve substantial agreement with human graders. A systematic analysis of the outputs from LLMs provides practical insights into their application to the grading task
Effective Neural Network Architecture for Few-Shot Learning
Get PDFDespite the current technological era\u27s ease of data collection, obtaining sufficient data for neural network training remains challenging due to factors like subject rarity, high costs, time constraints, incomplete historical data, and limited data in early research phases. Few-shot learning addresses these issues by enabling effective learning from small datasets, reducing data collection costs, and increasing accessibility to advanced machine-learning techniques. It also enhances training efficiency, improves generalization, and facilitates rapid adaptation to new tasks, proving invaluable in dynamic environments. Few-shot learning faces significant challenges due to data scarcity and the need for robust generalization. Limited training examples hinder effective learning and increase the risk of overfitting, leading to poor performance on new data. This dissertation explores various methodologies to overcome these challenges, aiming to improve model robustness and performance in real-world scenarios. First, we address the data scarcity problem that necessitates a multifaceted approach, with the overwhelming number of parameters in neural networks emerging as a primary culprit. This surplus of parameters intensifies the scarcity issue, leading to suboptimal performance and hindering effective learning. We address this issue through a comprehensive set of methodologies, leveraging domain knowledge to incorporate prior information, employing specialized algorithms tailored to the dataset, integrating Conditional Restricted Boltzmann Machines (CRBMs) with Deep Neural Networks (DNNs) to enhance learning efficiency, and optimizing efficiency through modular sparsification techniques. Secondly, it is also imperative to underscore the significance of generalization and robustness in addressing real-world challenges. Given the limited availability of labeled data typical in such scenarios, the ability of models to generalize effectively from sparse examples becomes pivotal. We endeavor to tackle this issue through a diverse array of methodologies. These encompass network-based regularization, prioritization of techniques leveraging domain knowledge, as well as the strategic utilization of hybrid models. Finally, we will merge the previously outlined methods into a cohesive model to effectively tackle real-world problems. We will perform experiments to compare our approach with SOTA methods, thereby demonstrating its effective performance