This review delves into using e-learning technology and personalized recommendation systems in education. It examines 60 articles from prominent databases and identifies the different methods used in recommendation systems, such as collaborative and content-based approaches with a recent shift towards machine learning. However, the current personalized recommendation system faces challenges such as a lack of understanding of the content, student discontinuity, language barriers, confusion in selecting study materials, inadequate infrastructure and funding. The review proposes using new digital technologies to address these issues, including Fluxy AI, Twin technology, AI-powered virtual proctoring, and Alter Ego. These technologies can create a dynamic and interactive learning environment, providing tailored learning experiences for students and insights for educators to provide targeted support and guidance. The integration of these technologies can improve individualized learning, increase understanding capacity and enhance the learning experience for students with speech disorders.
Autism, also known as Autism Spectrum Disorder (ASD) or Asperger’s syndrome, has an impact on cognition, social relationships, and behavior. Based on these characteristics and their learning interests, e-learning (Electronic Learning) recommendation systems attempt to provide personalized recommendations. This may enhance the learning ability of individuals with ASD. However, some technological challenges limit the ability of the population to use support scenarios and prevent efficient learning. This study tries to review e-learning recommender systems for individuals with ASD. The main identified aspects are 1) lack of design principles while building customized e-learning platforms for those with ASD; 2) technological limitations to developing recommender systems for e-learning, and 3) suggestions that reduce field limitations. The studies additionally discovered that social communication and psychological abilities had the greatest focus of study. A small number of studies mentioned the participant’s ASD level, the majority of the articles highlighted the beneficial effects of designing a content-based recommender system specifically for those with ASD.
In this work we introduce Split-Fed Learning (SFL), a new framework that combines the concepts of Federated Learning (FL) and Split Learning (SL), to provide privacy-aware and scalable training of machine learning models in settings with distributed data. With organizations needing more and more to take advantage of decentralized data sources without compromising on security, the potential of using SFL becomes all the more apparent. We performed extensive experiments to demonstrate the superiority of SFL against FL (both traditional and naive ones) as well as SL with various datasets. Experimental results show that SFL can reduce the communication overhead and training time by a large margin while achieving a competitive accuracy of final model. In contrast, SFL delivered 30% better training speed and 15% accuracy than baseline models particularly for non-IID data distributions challenges. Lastly, a thorough breakdown of the SFL architecture explains its novelty for separating model training between clients and servers with respect to model partitioning. Hence, this optimization not only improves the efficiency of resource utilization but also boosts the resiliency of learning process against data heterogeneity. These results demonstrate the potential of SFL as a scalable and efficient approach to privacy-preserving machine learning, providing an alternative solution to recent challenges faced by organizations that deal with sensitive data. This is the first paper establishing such type of literature for SFL, representing an initial step toward a theoretical background for future research for implementing distributed learning frameworks and next towards practical usage scenarios on testing SFL.
21 August 2024 in EAI Endorsed Transactions on Internet of Things
As we embrace the digital age, artificial intelligence (AI) has converted an essential share of our breaths, and teaching is no allowance. ChatGPT, OpenAI's cutting-edge language processing AI, ChatGPT, stands at the forefront of transforming our approach to education. This article delves into the myriad ways in which ChatGPT can assist educators in reshaping their teaching methodologies and enhancing classroom interactions. in providing personalized learning experiences, simplifying complex concepts, and enhancing student engagement. We also discuss real-world examples of its successful implementation and its potential future in the education sector. However, we also admit the limits of ChatGPT and the need careful consideration before its implementation. This article explores the support and impact of ChatGPT in education. It showcases real-world implementations and discusses the future potential of AI, particularly ChatGPT, in transforming teaching methodologies and classroom interactions. By emphasizing the role of technology in enhancing education, it highlights how AI, such as ChatGPT, can bring about positive transformations in today's classrooms.
Autism is a complex neurodevelopment condition that affects an individual's behavior, communication, and social interaction. Identification is a critical endeavor in health care, necessitating accurate and efficient diagnostic methodologies, early identification is pivotal for timely intervention and improved outcomes in affected individuals. This paper investigates the use of machine learning algorithms, specifically CatBoost, for Autism trait identification using heterogeneous datasets from toddlers, children, adolescents, and adults. The research investigates the performance of CatBoost in handling mixed data types, including categorical features and missing values, without extensive preprocessing. Utilizing gradient boosting on decision trees, CatBoost demonstrates its efficacy in capturing complex relationships between features, facilitating high predictive accuracy in autism identification. Through rigorous evaluation metrics such as accuracy, precision, recall, and Fl score, the designed system achieves a precise accuracy of 92% for adult datasets and 88% for child and adolescent datasets. This study delineates CatBoost's robustness across diverse age groups, providing insightful information on its applicability for Autism Spectrum Disorder diagnosis in the healthcare domain.
21 August 2024 in EAI Endorsed Transactions on Internet of Things
This research paper explores the development of a semantic search engine designed to enhance user query comprehension and deliver contextually applicable research results. Classic search engines basically struggle to catch the nuanced meaning of user queries, giving to suboptimal results. To address this challenge, we give the merge of advanced natural language processing (NLP) techniques with Elasticsearch, and with a specific focus on Bidirectional Encoder Representations from Transformers (BERT), a state-of-the-art pre-trained language model. Our approach involves leveraging BERT's ability to analyze the contextual meaning of words within documents by sentence transformers as (SBERT) , enabling the search engine to grab the user queries and better under- stand semantics of the content as it is converted into vector embeddings making it understandable in the Elasticsearch server. By utilizing BERT's bidirectional attention mechanism, the search engine can discern the relationships between words, thereby capturing the contextual nuances that are crucial for accurate query interpretation. Through experimental validation and performance assessments, we demonstrate the efficacy of our semantic search engine in providing contextually relevant search results. This research contributes to the advancement of search technology by enhancing the intelligence of search engines, ultimately improving the user experience by giving context based research.