Al-Kindi Center for Research and Development (KCRD) (E-Journals)
Not a member yet
6248 research outputs found
Sort by
Human-AI Collaboration in Customer Behavior Research: Personalizing Financial Services
No full textThis article explores the symbiotic relationship between artificial intelligence systems and human researchers in revolutionizing customer behavior analysis within the financial services sector. By examining how AI\u27s computational capabilities complement human contextual understanding, we demonstrate a framework where machine learning models process vast transactional datasets while human experts provide crucial interpretive insights regarding socioeconomic factors and cultural nuances. The resulting collaborative approach enables financial institutions to develop more sophisticated customer segmentation strategies, deliver precisely tailored product recommendations, and implement proactive retention measures through predictive churn analysis. This human-AI partnership represents a significant advancement over purely algorithmic or exclusively human-driven approaches, offering financial institutions a comprehensive methodology for enhancing customer engagement, improving service personalization, and ultimately driving business growth while addressing the complex needs of diverse customer bases
AI-Powered Patient Benefit Management: A Technical Overview
No full textThe integration of artificial intelligence into Patient Benefit Management (PBM) represents a transformative paradigm shift in healthcare administration and delivery. This technical article explores the multifaceted applications of AI technologies in revolutionizing how healthcare benefits are managed, optimized, and delivered to patients. From sophisticated machine learning algorithms that personalize benefit packages to automation systems that streamline administrative workflows, AI is fundamentally reshaping the PBM landscape. This article examines core technological frameworks enabling these advancements, data-driven approaches to benefit personalization, operational efficiency enhancements, and clinical decision support systems. By addressing both current implementations and future directions, this comprehensive analysis provides healthcare stakeholders with actionable insights for leveraging AI to create more responsive, efficient, and patient-centered benefit management systems
The Future of AI in Digital Search: Towards a Fully Conversational Experience
No full textThe digital search landscape is fundamentally transforming from traditional keyword-based interfaces to fully conversational experiences powered by artificial intelligence. This transformation is driven by advances in large language models, growing user expectations for natural interaction, and the increasing inadequacy of static query-response paradigms. As a result, search is evolving from a tool for retrieving links into an intelligent partner for dynamic, goal-oriented dialogue. This evolution represents more than a mere addition of chat capabilities to existing search engines; it marks a paradigm shift where contextual dialogue becomes the primary mechanism for information discovery. Conversational search systems maintain awareness across multiple interactions, proactively clarify ambiguities, and adapt to evolving user needs through sophisticated language processing, context preservation, and personalization. By distributing the cognitive burden of query formulation between the user and the system, these interfaces enable more natural information-seeking behaviors that mirror human dialogue rather than database queries. The architecture supporting these systems integrates advanced natural language processing, multi-modal capabilities, and dialogue management components that work in concert to deliver coherent, contextually appropriate responses. Despite significant advances, challenges remain in maintaining conversational coherence, developing appropriate evaluation metrics, addressing ethical considerations, and integrating diverse input modalities. As these systems mature, the boundary between search and intelligent assistance continues to blur, promising a future where information discovery becomes an anticipatory, conversational experience
AI at Scale: The Infrastructure Revolution Enabling GPT-Class Large Language Models
No full textThe extraordinary capabilities of Large Language Models (LLMs) like GPT-4 and Llama 3 have redefined the boundaries of artificial intelligence, yet their transformative power rests upon a foundation of breakthrough infrastructure innovations largely invisible to end users. This article examines the critical technological underpinnings enabling today\u27s frontier models, focusing on memory-efficient parallelism strategies that optimize computational resources, high-throughput interconnect technologies that facilitate massive distributed training, and advanced model sharding techniques including 4D parallelism that distribute model components across computational resources. By exploring the integration of these infrastructure elements—from specialized hardware accelerators to sophisticated software orchestration systems—we provide insights into how the AI community has overcome seemingly insurmountable computational barriers to scale training to unprecedented levels. Understanding these infrastructure innovations offers valuable perspective on both current capabilities and future directions as the field continues its rapid evolution toward increasingly capable AI systems
An Intervention-Study on Moroccan High Schoolers’ Difficulty in Comprehending Newly Encountered Vocabulary
No full textVocabulary knowledge is an essential component in language acquisition and learning. Such knowledge facilitates the reception, processing, and production of discourse. When the ‘know-what’ is hindered, learning is not achieved. The present study is conducted to arm EFL learners with ‘know-how’ strategies to decipher the meaning of newly encountered vocabulary. It is the result of attested observations that students’ comprehension and learning are hindered when they do not comprehend a word. Problems persist more when students do reading comprehension, grammar, or communication tasks. The purpose is to enable students with strategies to work out meaning and test the possibility of having those strategies become automated in the students’ cognition. For this, the researcher has adopted a quantitative approach to check the rate of students’ success at deciphering meaning. The study has three stages: pre-intervention, while-intervention, and post-intervention. The first stage is conducted to attest observations concerning students’ comprehension difficulties. Here, the researcher provides a short paragraph with at least 10 difficult words and asks students to clarify them. The while-intervention stage consists of delivering a lesson with a learning objective of training students to use 6 meaning-deciphering strategies adapted from Schmitt’s (1997) and Nation’s (2001) taxonomies. The post-intervention is a complementary step to check the validity and success of the intervention. As the sessions proceed, learners show success in independent meaning-deciphering using an automated approach. This study shows that learners should be helped not by providing ready information but by equipping them with ‘learning how-to-learn’ strategies.
Digital Transformation of English Language Learning in Moroccan Higher Education: A Study of Altissia in Moroccan ISTA Institutes
No full textThis study examines the impact of Altissia as a language learning platform on English proficiency among students in Moroccan vocational institutes (ISTA). While Altissia is intended to support self-paced and blended learning, its effectiveness remains uncertain. Using a quasi-experimental pre/post-test design with control and experimental groups, along with student perception surveys, the study found no significant improvement in English proficiency among Altissia users compared to non-users. Although students expressed moderate satisfaction, concerns about Altissia’s efficacy persisted. The findings suggest that Altissia, in its current implementation, does not significantly enhance language learning in ISTA institutes. The study highlights the need for improved instructional design, structured implementation, and stronger support mechanisms to optimize digital language learning tools in vocational education
The Sociocognitive and Metacognitive Perceptible University Artifacts of Language Proficiency in Scholars\u27 Oral Activity
No full textOral activity serves interactional, transactional, and highly demanding language functions. It is an artifact that recognizes how proficient humans can deal with either socio- or meta-cognitive endeavours. However, while the socio-cognitive schools have emphasized cognitive over strategic ability in deciphering knowledge, the meta-cognitive approach focuses on learning about how humans acquire knowledge during oral activities, using strategies, for instance. This marks a dual processing emphasis not on the socio-cognitive aspect of oral activity, but on how both the act of learning itself and the act of knowing how learning is achieved necessitate tactics. Historically, some of these strategies have been thoroughly identified and investigated, but others remain under research. Therefore, for this knowledge achievement to take place, scholars resort to artifacts such as organisational mechanisms, empowerment tools, value delivery paradigms, teamwork, and reference management. Current research aims to investigate these artifacts in greater detail, shedding light on oral activity more specifically. This research is descriptive, associating what scholars do in specific situations with the impact this can have on their language proficiency. In sum, perceptible university artefacts of language in scholars\u27 oral activity are an attempt to lay the groundwork for linking research on knowledge building to proficiency building
Characteristics and Factors of a Good and Effective English Language Teacher
No full textThis study is carried out to show and shed light on the characteristics and factors that any good English Language teacher should have. Indeed, having these characteristics and factors help in mastering the class and thus, lead to good results. For this purpose, a questionnaire for Jazan University English teachers was used for data collection. The Statistical Package for Social Science Program (SPSS) was used for data analysis. The results and findings of the questionnaire revealed some general and personal characteristics. For example, teacher-student good relationships, teacher’s physical appearance, educational and pedagogical competence, classroom and time management, dealing with the problems inside the classroom, teaching methods and plans diversity, language proficiency, and moral characteristics such as justice and equality among students. Some recommendations were given by the researchers such as: the above mentioned characteristics and others are too important to be obtained by English Language teachers as an essential part in learning process so as to give the required results and outcomes
The Implementation of Mindfulness in Chinese College English Teaching
No full textThis study aims to analyze the application cases of mindfulness in foreign language education overseas and provide suggestions for implementing mindfulness in Chinese college English teaching. It first introduces the application cases of mindfulness in the field of mental health at home and abroad. Then, it analyzes the application cases of mindfulness in foreign language education overseas. Finally, it explores the implementation of mindfulness in college English teaching, aiming to help students more effectively cope with learning pressure, reduce negative emotions, increase positive experiences, and ultimately improve their English learning outcomes
Translation Profession Between the Ethical Challenges and Social Responsibilities
No full textLanguage transfer is an important and complicated process that requires a critical communication component, which is translation. As a language mediator, translation is subject to ethical and social responsibilities. The translator is responsible for transferring the source text in a multilingual setting, and accountability became a vital issue in the translation and interpreting fields. Baker & Maier (2011) claim that there is an increase in responsibility. Thus, this has yielded an increase in visibility, hence greater pressure on the profession to demonstrate that it is cognizant of its impact on society (3). In many situations, a translator faces challenging and complicated tasks; Robinson (1997) raised the following questions from a translator`s point of view; what can a translator do when he/she is asked to translate an offensive text? Or, to put it differently, what can a feminist translator do when she is asked to translate a blatantly sexist text? And what can be done in the case of a liberal translator when he/she is asked to translate a neo-Nazi text? (26). These previous examples carry aspects that may create a clash between professional ethics, loyalty to the person, company, or agency paying the translator, and sometimes the translator`s personal and moral beliefs. In the present paper, I aim to expand my understanding of ethics and social responsibilities in the translation field. Moreover, I seek to analyze various challenges that may face the translator/interpreter while producing a target text that fits into the new cultural setting of the target language, serves its purpose, and considers professional ethics.