AUB ScholarWorks (American Univ. of Beirut)
Not a member yet
33634 research outputs found
Sort by
Brave like the stars
No full textThis heartfelt picture book tells the story of Lea, a young girl wearing her hijab to school for the very first time.
Filled with gentle moments and emotional honesty, the story follows Lea as she navigates feelings of self-doubt and fear of being different. With the loving support of her Mama and Baba—and the kindness of her friends Emma and Adam—Lea begins to see her hijab not as something to hide, but as something that shows her strength, kindness, and bravery.
With soft, hand-drawn illustrations and simple, relatable language, this book offers a powerful message about identity, courage, and belonging. Perfect for classrooms, bedtime reading, or families looking to spark conversations about self-confidence and cultural pride, this story is a warm celebration of being true to yourself.This children's book tells the story of Lea, a young girl wearing her hijab to school for the first time. Through gentle language and hand-drawn-style illustrations, the book explores themes of identity, self-confidence, and belonging. Inspired by Sutherland’s and Maslow’s frameworks of children’s needs, the story supports emotional growth and critical literacy by helping readers reflect on self-acceptance, friendship, and cultural pride
When I Grow Up I Want To Be…
No full textA children's book, created by AUB students from the Education Department, for the course EDUC218 as a final project
The History, Development and Future of Iraq's Popular Mobilisation Forces
No full textProject. M.A. American University of Beirut. Center for Arab and Middle Eastern Studies, 2025.Since 2014, the Popular Mobilisation Forces (PMF) have become an important and
powerful actor within Iraq. The PMF is an umbrella organisation made up of the
militias that fought against the Islamic State (IS) terrorist group in Iraq between 2014
and 2017. This project provides an overview of the origins, ideological alignments and
operational structures of the PMF’s constituent militias. The project views the PMF as a
series of networks that operates within, alongside and outside the official institutions of
the Iraqi state. Since the defeat of IS in 2017, the PMF has consolidated its position
within Iraq and increased its power through military means, economic activities and
participation in parliamentary elections. The PMF is currently well integrated into the
Iraqi state, and its ability to maintain this position will continue to be an issue for the
future of Iraq and the wider region
Abū al-Hudā al-Ṣayyādī and Islamic Conservatism During the Reign of Abdülhamid II (1878-1908)
Full text linkThis dissertation examines the role of Abū al-Hudā al-Ṣayyādī (1847–1909) in shaping Islamic conservatism during the reign of Sultan Abdülhamid II (1876–1909). Through an analysis of his writings, networks, and political influence, it explores how al-Ṣayyādī and other conservative Sufi figures contributed to the consolidation of Hamidian ideology and the sultan-caliph’s legitimacy. Contrary to historiographical narratives that emphasize Islamic modernist thought, this study highlights an alternative intellectual current rooted in traditional Sunni epistemology and Sufi conservatism. It argues that Ottoman conservatism was not merely reactionary but an active and strategic response to the political and ideological challenges posed by nationalism, reformist movements, and European imperial encroachment. By engaging with themes such as obedience to the sultan-caliph, the theological foundations of Islamic universalism, and the instrumentalization of Sufism for state legitimacy, this dissertation provides a reassessment of the Hamidian era's political and intellectual landscape. In doing so, it challenges dominant historiographical paradigms and contributes to a broader understanding of the dynamics between tradition and reform in the late Ottoman Empire
Factors Influencing Ecotourism Development: A Comparative Study of Shouf Biosphere Reserve Villages
No full textThis study explores the factors influencing ecotourism development in the Shouf
Biosphere Reserve (SBR) by conducting a comparative case study of two villages:
Maaser El Shouf and Niha. The comparative case study methodology analyzes
similarities, differences, opportunities, and challenges between the two villages.
While both villages are significant in size and strategically positioned as entry points to
the reserve, their levels of ecotourism development vary.
The research identifies key stakeholders involved in ecotourism development and
examines the environmental, socio-cultural, and economic factors that shape ecotourism
practices in these villages.
The findings indicate that Maaser has benefited from stronger stakeholder collaboration,
more effective governance structures, and better-developed infrastructure, leading to
more successful ecotourism initiatives. In contrast, Niha faces challenges such as
resistance to change, limited local participation, and underdeveloped infrastructure,
which hinder its ecotourism potential. The research also highlights the need for tailored
strategies that consider the unique characteristics of each village and stresses the role of
ecotourism in promoting environmental conservation, cultural heritage, and sustainable
local economies. The findings offer insights for developing ecotourism in biosphere
reserves, contributing to local livelihoods, cultural preservation, and environmental
sustainability by understanding the drivers of ecotourism development.
This study provides recommendations for policymakers and SBR planners and
managers. In addition, this study proposes various directions for future studies
The Politicization of Climate Policies in the Israeli Occupation of Palestine: Environmental Responsibility or Tools of Domination?
No full textThis project critically investigates the politicization of climate policies within the framework of Israel's occupation of Palestine and analyses the duality of environmental responsibility and tools of domination. Embedded within the broader discourse on climate and conflict, it engages with critiques of the securitization and depoliticization of climate narratives that obscure local power dynamics and systems of oppression. Through a discourse analysis of the official statements of Israel at the United Nations Climate Change Conferences (COP) since 2015, this project shows how Israel instrumentalizes environmental discourses in support of its settler colonial project. At the same time as Israel has positioned itself as a leader in climate innovation, afforestation, water management, and the development of renewable energies reshape the landscape and solidify control over Palestinian territories under the rhetoric of environmental stewardship. This project brings postcolonial studies and critical security studies together in an original way to expose the inability of dominant climate-security paradigms to take seriously anything but an abstract threat in view of the lived experiences of marginalized populations. It emphasizes how socio-political contexts must be integrated within climate governance frameworks that offer nuanced insights into how climate policies perpetuate systemic inequalities, especially in occupation contexts
Bridging Cultures: Harnessing Large Language Models for Translation of Lebanese Dialect
No full textMachine translation (MT) of Arabic dialects presents unique challenges, mainly due to their rich cultural context, and the scarcity of linguistic resources. While Large Language Models (LLMs) such as ChatGPT, LLaMA, and BLOOM have demonstrated remarkable capabilities across a range of MT tasks, their effectiveness in translating culturally embedded dialects remains largely unexplored. This thesis specifically investigates the effectiveness of LLMs in translating the Lebanese dialect, a prominent Arabic variant in the Levant region, known for its rich cultural heritage and complex idiomatic language.
A key limitation in dialectal MT is the scarcity of culturally representative datasets needed to develop effective models. The few existing Lebanese-English parallel datasets suffer from cultural misalignment due to their translation from non-native resources. To address this gap, this research introduces two culturally aware resources- LW, and LebEval- derived from authentic Lebanese podcasts, and professionally translated to English. It further investigates the advantage of collecting such authentic datasets by conducting comprehensive experiments comparing the performance of Arabic-centric LLMs against NMT systems. Findings reveal that while both architectures perform similarly on non-native datasets, LLMs demonstrate superior capabilities in preserving cultural nuances, outperforming NMTs by a significant margin on the LW data. Additionally, while fine-tuning LLMs with instructional data has shown promising results in MT tasks, there has been little to no effort dedicated to adapting these techniques specifically for Arabic or its diverse dialects. This work explores fine-tuning the open-source Aya23 model on three types of instructions: 1) parallel Lebanese/English instructions, 2) contrastive instructions, and 3) Grammar-hint instructions. Results demonstrate that models fine-tuned on a smaller but culturally aware Lebanese dataset (LW) consistently outperform those trained on larger, non-native data. They also show the superiority of fine-tuning using contrastive instructions, highlighting the value of leveraging translation errors. Finally, while most studies on the translation of Arabic dialects rely on the statistical evaluation metric BLEU, despite its well-documented limitations, this research takes a different approach by conducting a human correlation analysis with different evaluation metrics. Findings validate the shortcomings of BLEU and showcase xCOMET as a more reliable and culturally sensitive metric for evaluating the quality of MT in this domain.
Overall, this thesis makes significant contributions to culturally aware dialectal MT, highlighting the potential of leveraging LLMs and challenging the prevailing "more data is better" paradigm
Rivers and Rivalries: A Political Ecology Analysis of the Transboundary Assi (Orontes) River Basin in Lebanese-Syrian Relations
Full text linkTransboundary water governance is a critical issue in regions where shared water resources intersect with complex political and socio-ecological dynamics. This thesis explores the Assi (Orontes) River Basin, a transboundary river shared by Lebanon and Syria, through the lens of political ecology and socio-ecological systems thinking. While Lebanon is the upstream riparian, its utilization of the Assi River’s waters remains limited due to historical, political, and governance-related factors. Using a qualitative research approach, this study employs thematic analysis of semi-structured interviews with local resource managers, farmers, aquaculturists, governmental officials, and NGO representatives to assess perceptions of water governance, environmental degradation, and socio-economic impacts. Findings indicate that contrary to dominant academic narratives, Syrian influence is not perceived as a major factor restricting Lebanon’s water development. Instead, internal governance failures, systemic marginalization, and corruption are identified as primary barriers. The research also highlights the declining environmental health of the Assi River, particularly due to pollution from unregulated aquaculture and groundwater over-extraction. The study argues that Lebanon’s political and economic neglect of the Hermel region, where the Assi River flows, has exacerbated underdevelopment, overshadowing transboundary water disputes as a key issue. It challenges existing frameworks of hydro-hegemony, proposing that transboundary water management in smaller river basins should be analyzed not only through geopolitical power asymmetries but also through the lens of domestic governance inefficiencies and local socio-political dynamics. The thesis concludes by recommending institutional reforms, infrastructure investment, and enhanced cooperation between Lebanon and Syria to ensure the sustainable and equitable management of the Assi River. It also calls for a paradigm shift in transboundary water governance studies to account for internal governance structures and localized socio-ecological factors alongside traditional geopolitical analyses
Modeling of CO2 Flow with Impurities
No full textCarbon Capture, Utilization, and Storage (CCUS) plays a pivotal role in reducing greenhouse gas emissions, essential for limiting global warming to below 1.5 °C by 2100 and achieving carbon neutrality by 2050. Modeling carbon dioxide (CO₂) density is crucial for optimizing CO₂ transportation and storage systems. However, captured CO₂ streams, often originating from power sources, contain impurities such as Oxygen (O₂), Nitrogen (N₂), Carbon Monoxide (CO), Argon (Ar), Sulfur Dioxide (SO₂), Hydrogen (H₂), Methane (CH₄), Water (H₂O), and Hydrogen Sulfide (H₂S). These impurities significantly impact transmission properties and challenge the predictive capabilities of current equations of state (EoS). To address these challenges, this study utilized a comprehensive dataset comprising 134,204 density data points to evaluate the performance of 14 EoSs, including cubic, virial, physical, and multi-parameter equations. A versatile computational framework was developed, capable of calculating a range of thermodynamic properties beyond density, such as fugacity and its derivatives, enthalpies, and pressure derivatives. Within the development process, various numerical solvers were implemented to enhance computational efficiency, and genetic algorithms were adopted for better predictive performance. This robust framework provides researchers with a powerful tool for advancing CCUS technology, laying the foundation for future studies in thermodynamic modeling and system optimization.
For density predictions, the research adopted a dual approach. Machine learning (ML) techniques, including Random Forest, Gradient Boosting, and Neural Networks, were employed to enhance predictive accuracy. These models demonstrated robust performance across complex thermodynamic regions by leveraging a combination of experimental and synthetic data (R² > 0.96). Synthetic data, generated within CCUS pipeline operating conditions using the best-performing EoSs, primarily multi-parameter equations, exhibited an Absolute Average Relative Deviation below 3%.
On the other hand, the second approach utilized ranking EoS performance across temperature, pressure, and composition intervals, analyzed using Pareto plots and decision trees with Entropy and Gini indices. This systematic framework identified the optimal applicability ranges for each EoS, shedding light on their strengths and limitations. The integration of ML with interval-based sorting allowed for precise density predictions and reliable assessments of model stability under diverse conditions.
The scope of the study extended to stability analysis, evaluating thermodynamic behavior and ensuring reliable predictions critical for the safe and efficient operation of CCUS systems. By assessing phase behavior, stability limits, and operational conditions, the study addressed key challenges in CO₂ pipeline systems and laid the groundwork for their effective design and management.
Building on these insights, the research applied the findings to optimize CO₂ transmission networks. A case study was conducted, exploring mass and molar balances, energy balances, system stability, pressure and temperature drops, and viscosity modeling. This comprehensive analysis provided actionable insights into the operational dynamics of CO₂ pipelines, enabling precise predictions and optimizations for real-world applications.
To the best of our knowledge, no previous study has utilized such an extensive dataset, evaluated such a diverse array of EoSs, or incorporated this innovative hybrid approach alongside practical case studies and a computational framework of this breadth. These findings provide a robust platform for improving the efficiency, reliability, and scalability of CO₂ transportation and storage systems, further supporting CCUS’s pivotal role in achieving global climate objectives
A Preliminary Validation of The Arabic Multidimensional Dyspnea Profile
No full textRecognizing dyspnea as a complex symptom is advised by the American Thoracic Society. Evaluating not only the underlying causes of dyspnea, rather the sensory and the emotional aspects of dyspnea perception is crucial, prompting a shift from treating solely the physical illness to tailoring therapies based on individual needs. Conventional dyspnea scales offer a limited view of breathlessness, merely quantifying its intensity. However, comprehending dyspnea necessitates a nuanced scale that accounts for its various aspects. The Multidimensional Dyspnea Profile (MDP) offers a comprehensive evaluation of dyspnea perception, dividing its description into sensory and affective components. The aim of this project is to validate the Arabic version of the MDP among individuals suffering from chronic pulmonary illnesses. Ultimately, the goal is to incorporate this version into healthcare facilities to gain a more comprehensive understanding of dyspnea perception and enhance prognostic capabilities. The pilot study will follow a cross-sectional quasi-experimental design at the American University of Beirut Medical Center from April to November 2024. Twenty-five participants answered a baseline questionnaire [T0] that assessed their dyspnea [using the Arabic MDP and other scales]. They then performed a six-minute walking test followed up by a repeat assessment of their dyspnea levels [T1]. The sample consisted of72% males with mean age of 63.6±8.67 years. Reliability analysis of the Immediate Perception domain showed higher consistency results in comparison to the Emotional Response domain (T0: 0.701, 0.364 respectively for both domains and T1: 0.723 for the Immediate Perception domain). Concurrent validity showed strong correlation between Borg and Immediate Perception domain at T1 (r=0.832, p=0.00). Divergent validity showed weak correlations between Immediate Perception domain at T0 with BMI (r=0.051, p=0.81) and Ratio of FEV1/FVC (r=0.118, p=0.573). Age and CCI were found to be negatively correlated with the Emotional Response domain at T0 ( r= -0.575, p= 0.003 and r= -0.563, p =0.003). Fatigue showed a strong negative correlation with the MDP scores across both times T0 and T1 ( Immediate Perception at T1: r= -0.621, p=0.00). A notable finding was the decrease in the MDP scores after conducting the 6MWT (Median score of Immediate Perception at was computed to be 18 at T0 and 10 at T1. Nonetheless, the median scores of the Borg scale were found to be higher in T1 than T0 (0 and 3). These results provide evidence for the psychometric validity of the Arabic version of the MDP and support its use in capturing a holistic assessment of dyspnea