14129 research outputs found
Sort by
The political economy of AI: a content analysis of ChatGPT outputs and anti-Palestinian media bias
No full textPalestine represents a site of digital occupation and a laboratory for the Israeli military–industrial–communications complex that entails collaboration between Israeli military, technology and academia. Considering the current popularity of generative artificial intelligence (AI) and the use of AI in the genocide in Palestine, we investigate how generative AI platforms like ChatGPT are biased towards Palestinians, and how prompting in English and Arabic produces different biases towards Palestinians. A theoretical framework for analysing digital Orientalism is advanced through a historicisation of media imperialism, the military–industrial–communications complex and the political economy of AI in occupied Palestine. Grounded in research-based teaching methods, and employing a mixed-method content and frequency analysis and comparison of ChatGPT outputs in English and Arabic, this article expands research on AI biases, algorithmic oppression and digital divides within the context of digital Orientalism and anti-Palestinianism. We document an anti-Palestinian bias in AI-generated content in English and Arabic that is evident through biased language and the omission of historical and legal contexts, Palestinian human rights, Palestinians’ status as refugees and their right of return. ChatGPT-generated outputs reproduced Orientalist discourse and Israeli narratives entailing Nakba denial, further emphasising the urgency for critical digital media literacy on AI.Publishe
Synthesis and characterization of ZnO nanoparticles for dental applications: Improving the mechanical properties of the temporary dental filling materials
No full textWe investigate the effect of ZnO nanoparticles (ZnO-N) mainly on the mechanical properties of zinc oxide eugenol (ZOE) cements compared to the commercially available bulk ZnO powder (ZnO-B). Our results revealed the polycrystalline structure and well-dispersed nanoparticles (NPs) of regular shapes corresponding to ZnO-N with an average crystallite size of 54.68 nm and energy band gap (Eg) value of 2.7 eV. The Vickers microhardness (
) measurements showed a clear predominance of mechanical parameters for ZOE-N over ZOE-B where an increase by about 60% was recorded. The indentation creep behavior of ZOE-N sample possessed dislocation creep at all investigated loads (2, 3, 5 N) judging from the conducted exponent stress values (15.89, 16.75, 18.41). This work presents an obvious enhancement in the mechanical performance of ZOE using ZnO-N making it more convenient as temporary dental filling materials opening the door for further investigation of their uses in dental applications.Publishe
Masculinity Under Siege: The Evolution of The Lebanese Maronite Identity
No full textThis thesis studies the evolution of the concept of masculinity in Lebanon from the events leading to the Civil War through the present day. It explores how existential threat catalyzed the development of a hypermasculine ideal constructed against the Muslim "Other" during the war years. In the postwar era, the Maronite identity entered a state of crisis, grasping for a strong leadership that can bring back its waning superiority. The formation of this identity is a product of sectarian struggles shaped by a failed power-sharing political system, a deeply engrained fear of the “Other” who is perceived as different in allegiances, agenda, and vision of Lebanon’s future, division amongst the Maronite community, and a longing for a victory that was not achieved, or yet to be achieved.
Acknowledging the patriarchic and sectarian nature of Lebanon, this study examines the historical and contemporary forces shaping Maronite masculinity. It specifically aims to investigate right-wing tendencies and asks: in an era of political transformations, what constitutes Maronite masculinity today
The Construction of a Hypermasculine Dynasty Among Lebanese Young Men: A Gender Critical Discourse Analysis
No full textDiscourse is the act of uniting words to convey a message, transmitting emotions, ideas, and knowledge to a receiver, regardless of their consent. Across various domains, these messages carry immense power—the power to create, resist, manipulate, protect, transform, incite violence, and foster peace. Yet, throughout history, discourse has also been used to sow discord, justify inequalities, and perpetuate unchecked power. This power, often inherited and passed down through generations, shapes the "future other," upholding exaggerated male traits like strength and aggression, fostering hypermasculinity. By devaluing women and femininity, hypermasculine men preserve their dominance at all costs.
This thesis explores the impact of discourse on the education of young Lebanese men, examining how language, culture, and education contribute to the formation of a hypermasculine society. Drawing on existing scholarly work, the study employs gender-critical discourse analysis of the narratives of Human Values and Social Studies coordinators along with investigating language used in certain History and Civic textbooks through the lens of the Social Studies coordinator. Additionally, the study incorporates the perspectives of 280 learners aged 15-18. At last, the findings and analysis affirm the role of education in constructing hypermasculinity among Arab young men
Predicting Android Malware Using Evolution Networks
No full textIn Cybersecurity, a main and persistent issue is the threat of malware. This issue
requires the development of efficient solutions in order to keep up with the continuous
evolution of malware. With this aim, we introduce evolutionary networks, and
particularly the Susceptible-Infectious-Susceptible (SIS) model, as a way to address
the limitations of previous studies which are typically based on traditional machine
learning models. The SIS model is usually used to represent disease spread between
individuals in a population with transition between susceptible and infected
states. We modify the SIS model to include weighted edges and we introduce an
edge-breaking probability. Android malware propagation is thus transformed into
a directed network in which nodes represent IP addresses and edges represent aggregated
multiple packet transmissions weighted by communication frequency. We
combine this model with genetic algorithms to optimize its parameters and return
the best state transition probabilities, and we predict future malware accordingly. Experimental studies clearly show a higher accuracy of our proposed approach in
comparison with existing machine learning models, namely random forest, artificial
neural network, decision tree, and logistic regression
Evaluation of Beauvericin’s Activity and Mode of Action Against All Life Stages of Leishmania tropica
No full textLeishmaniasis is a vector-borne complex malady and causative of a broad spectrum of clinical manifestations. Leishmaniasis has elicited a global health concern, following its spread via the Leishmania parasite. L. tropica, the causative agent of cutaneous leishmaniasis, is an understudied species of Leishmania with various variables yet to be elucidated. Due to the scarce therapeutic agents available in the market, their limited efficacy, and resistance acquisition following their usage, novel interventions are imperative. Our study was executed to delineate the efficacy and resistance acquisition towards BEA by promastigotes and intracellular amastigotes of L. tropica, in association with macrophage-like cells. Moreover, RNA sequencing was employed to identify the differentially expressed genes in response to drug treatment. Our data showcased the high potency of BEA against both developmental stages of the parasite. Additionally, we conclusively demonstrated BEA’s low cytotoxicity to macrophage-like cells, with a lower value to L. tropica developmental stages. Furthermore, BEA mediated low resistance acquisition in the two distinct life stages of the parasite. In addition, RNA-Seq data revealed the overexpression of ten assembled gene sets in both life stages of the parasite following exposure to BEA. Our study elucidated an in-depth view of BEA’s potency in inhibiting the growth of L. tropica, characterized by high efficacy and low resistance acquisition to the drug. Moreover, this study showcased the various mechanisms of action of BEA through the identified differentially expressed genes within the parasite’s genome
Syrian Refugees and the Struggle for Sovereignty in Lebanon: Navigating Political and International Tensions
No full textThis study examines the ongoing tensions surrounding the repatriation of Syrian
refugees and its destabilizing effect on Lebanon’s political system. Since the onset of the
Syrian civil war in 2011, approximately 1.7 million Syrians sought refuge in Lebanon.
Initially, a significant portion of the Lebanese government supported these refugees on
humanitarian grounds. However, the protracted presence of refugees has strained
Lebanon’s fragile political system, leading to conflicting perspectives between Lebanon
and the international community. While international actors advocate for securing refugees'
settlement and upholding their rights, Lebanon maintains a stance prioritizing their return
to Syria, citing national sovereignty and security concerns. The central research question of this study asks: How has the influx of Syrian refugees
influenced Lebanon's political discourse on national sovereignty, and what are the tensions
between Lebanon’s policy responses and its international legal obligations? This study will
explore Lebanon’s internal divisions on the issue and the ways in which external pressures
shape its policy towards refugees. The research also interrogates whether Lebanon’s
sovereignty is being compromised by international interventions and the continued
settlement of refugees, ultimately impacting the country’s political, economic, and
demographic stability. The study argues that without constructive dialogue between the
Lebanese government and international actors, the impasse on Syrian refugees will persist,
exacerbating Lebanon’s ongoing economic crisis and political fragmentation
Nutritional Status and Clinical Outcomes of Children with Medical Complexity in the United Arab Emirates: A Retrospective Cross-Sectional Study
No full textBackground:
Children with medical complexity (CMC) represent a unique pediatric population. Their prevalence is increasing and they are suggested to be at a higher risk of undernutrition. However, limited data exist regarding their nutritional status, predictors of undernutrition, and its association with clinical outcomes, particularly in the Middle East.
Objective:
This study explored the nutritional status of CMC aged 0–24 months, admitted to a dedicated pediatric tertiary hospital in the United Arab Emirates (UAE), and examined the predictors of undernutrition, and its association with clinical outcomes: hospital length of stay, early readmission, recurrent admission, and mortality.
Methods:
This retrospective cross-sectional study included 446 hospitalized CMC at Al Jalila Children’s Specialty Hospital, Dubai, from 2016 to 2023. Nutritional status was assessed using CDC growth z-scores (weight-for-age: WAZ, length-for-age LAZ, weight-for-length: WLZ) and the Composite Index of Anthropometric Failure (CIAF), with diagnoses grouped by medical condition. Descriptive statistics summarized nutritional status. Exploratory bivariate analyses (Chi-square, Pearson correlation, t-tests, ANOVA) identified factors associated with the CMC’s nutritional status, and variables with p < 0.2 were entered into multivariable linear regression models to identify predictors, adjusting for confounders.
Results:
Undernutrition was highly prevalent, with 60.8% of children classified as underweight, 49.1% as wasted, and 41.7% as stunted. According to the CIAF, 75.1% exhibited at least one form of anthropometric failure, while 54.9% presented with multiple concurrent forms. No significant differences in WAZ, LAZ, or WLZ scores were observed based on sex or nationality. Preterm birth was significantly associated with lower WAZ (β = -0.849, p < 0.001) and LAZ (β = -1.464, p < 0.001) scores. Children diagnosed with respiratory and gastrointestinal-related conditions had significantly lower WAZ (β = -0.884, p = 0.024) and WLZ (β = -2.108, p = 0.005) scores compared with those with congenital anomalies. Mixed feeding was linked to decreased WAZ (β= -0.465, p = 0.027) and WLZ (β = -1.329, p = 0.001) scores in comparison with exclusive breastfeeding, while formula feeding was associated with lower WLZ scores (β = -0.874, p = 0.025). Older age at admission was associated with lower WAZ scores (β = -0.053, p < 0.001). Longer hospital length of stay was associated with reduced LAZ scores (β = -0.005, p = 0.006) and increased WLZ scores (β = 0.006, p = 0.035). Although undernourished children exhibited higher rates of early readmission, recurrent admission, and in-hospital mortality, these associations were not statistically significant.
Conclusion:
Undernutrition is alarmingly high among hospitalized CMCs, with a significant association with adverse clinical outcomes, including prolonged hospitalization. Routine nutritional screening, early involvement of dietitians, and targeted, individualized feeding strategies are urgently needed to improve CMC’s health outcomes and reduce the burden of undernutrition on the healthcare systems
Unmasking Hidden Gluten: Evaluating the Compliance of Gluten-Free Products and Consumer Exposure in Lebanon
No full textBackground: Global demand for gluten-free foods has surged as diagnoses of celiac disease
and other gluten-related disorders rise, leading retailers to offer an increasingly wide range
of “gluten-free” products. Although international standards cap gluten at ≤ 20 ppm, surveys
in Europe, North America, and the Middle East continue to detect label violations, pointing
to cross-contact and inconsistent manufacturing controls. Lebanon’s only prior market
study dates to 2014, leaving a decade-long data gap on product integrity and the actual
exposure faced by Lebanese consumers who depend on accurate gluten-free labelling for
their health.
Methods: Following a thorough market screening, a cross-sectional analysis was conducted
on 115 stock-keeping units (SKUs) , including 54 local and 61 imported products. Gluten
was quantified with the RIDASCREEN® R5 ELISA; with levels > 20 ppm were deemed
non-compliant. A qualitative food-frequency questionnaire (FFQ) was administered to 66
self-selected gluten-free shoppers to assess intake frequency of each non-compliant SKU.
Results: 11 SKUs (9.6%) exceeded 20 ppm, while another 15 (13%) fell between the 1
ppm quantification limit of the kit and 20 ppm. Non-compliance was markedly higher in
Lebanese products (18.5%) than in imports (1.6%). 40% of contaminated local SKUs and
the sole contaminated import bore an FSMS logo, signaling implementation gaps. Bread
(6/43) and snack/bars (4/30) accounted for most infractions; a single oat-based “other
grain” also tested positive. Although 87% of products were analytically compliant, the FFQ
revealed 8 daily and 11 weekly consumption events involving contaminated SKUs.
Notably, 65% of respondents were celiac, and 27% purchased for celiac relatives,
increasing clinical risk.
Conclusion: Lebanon’s retail gluten-free sector is largely compliant, yet a substantial risk
persists in domestically produced breads and snack items. Strengthened regulatory
surveillance, dedicated production lines for high-risk categories, and targeted consumer
education must be enforced to safeguard medically vulnerable populations and uphold ethical standards
A Novel Centrality-Based Approach for Link Prediction
No full textLink prediction aims to identify missing or future connections between entities
of a complex system, when modeled as a network. This research problem has attracted
significant attention due to its relevance in numerous fields. In this work, we
propose a novel link prediction approach that integrates various network centrality
metrics in order to predict the likelihood of future connections between entities of a
system. Namely, we consider weighted betweenness, closeness and Katz centralities,
in addition to the Resource Allocation and Adamic-Adar indices. We also use a
genetic algorithm to optimize the weights of these metrics, reflecting their contributions
to link prediction. We tested our method on several real-world benchmark
networks. Our experimental results show that the proposed approach outperforms
various state-of-the-art link prediction approaches, highlighting its effectiveness and
validity